index.xml

<?xml version="1.0" encoding="utf-8" standalone="yes" ?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Vienna R User Group </title>
    <link>/</link>
    <language>en-us</language>
    <author>Vienna-R</author>
    <rights>(C) 2019</rights>
    <updated>2019-02-12 00:00:00 &#43;0000 UTC</updated>

    
      
        <item>
          <title>Meetup 02-2019 Minutes</title>
          <link>/blog/2019/02/12/february-2019-meetup/</link>
          <pubDate>Tue, 12 Feb 2019 00:00:00 UTC</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2019/02/12/february-2019-meetup/</guid>
          <description>

&lt;h2 id=&#34;self-service-data-preparation-und-data-science&#34;&gt;Self Service Data Preparation und Data Science&lt;/h2&gt;

&lt;p&gt;&lt;a href=&#34;https://www.linkedin.com/in/peter-jeitschko-3aa02179&#34;&gt;&lt;strong&gt;Peter Jeitschko&lt;/strong&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Peter presented Alteryx, a platform built for Business Analysts to master tasks like data management, data cleaning and modelling. The tool is windows only and will be ported to Linux soon. It can connect to multiple data sources and helps Business Analysts to deploy models in production. Finally, Peter also showed a Demo including data ingestion, an example of the Facebook Face API and some community features.&lt;/p&gt;

&lt;p&gt;You can find the &lt;strong&gt;slides&lt;/strong&gt; of his presentation &lt;a href=&#34;/slides/Alteryx Corporate Intro Presentation.pptx&#34;&gt;here&lt;/a&gt;.&lt;/p&gt;

&lt;h3 id=&#34;questions&#34;&gt;Questions&lt;/h3&gt;

&lt;p&gt;&lt;strong&gt;Do you have scheduling capabilities (for metadata, e.g. &lt;a href=&#34;https://atlas.apache.org/&#34;&gt;Atlas&lt;/a&gt;)?&lt;/strong&gt;&lt;br&gt;
Yes - Alteryx has scheduling capabilities server side and on the desktop.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Who is the target audience?&lt;/strong&gt;&lt;br&gt;
Business Analysts, business users?&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;How does it scale to millions of rows? Does it run in parallel?&lt;/strong&gt;&lt;br&gt;
Yes, to many CPU&amp;rsquo;s -&amp;gt; Large Server&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Which language is behind Alteryx/ what is running behind the scenes?&lt;/strong&gt;&lt;br&gt;
C++?&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;How much?&lt;/strong&gt;&lt;br&gt;
€5195 per user per year, see also&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Who are the competitors?&lt;/strong&gt;&lt;br&gt;
Knime (Open Source), Data Iku&lt;/p&gt;

&lt;h2 id=&#34;grpc-xmlrpc-in-r&#34;&gt;grpc + xmlrpc in R&lt;/h2&gt;

&lt;p&gt;&lt;a href=&#34;https://www.wu.ac.at/statmath/faculty-staff/faculty/fschwendinger&#34;&gt;&lt;strong&gt;Florian Schwendinger&lt;/strong&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Florian presented the package gRPC to interface the popular RPC framework, see also &lt;a href=&#34;https://github.com/nfultz/grpc&#34;&gt;https://github.com/nfultz/grpc&lt;/a&gt;. The reason to use gRPC instead of REST APIs is performance since messages are sent via &lt;a href=&#34;http://dirk.eddelbuettel.com/code/rprotobuf.html&#34;&gt;protobuf&lt;/a&gt; instead of JSON.&lt;/p&gt;

&lt;p&gt;You can find the &lt;strong&gt;slides&lt;/strong&gt; &lt;a href=&#34;/slides/grpc.pdf&#34;&gt;here&lt;/a&gt;.&lt;/p&gt;

&lt;h2 id=&#34;serverless-computing-with-aws-for-data-science&#34;&gt;Serverless Computing with AWS for data science&lt;/h2&gt;

&lt;p&gt;&lt;a href=&#34;https://www.linkedin.com/in/christoph-bodner&#34;&gt;&lt;strong&gt;Christoph Bodner&lt;/strong&gt;&lt;/a&gt; and &lt;a href=&#34;https://www.linkedin.com/in/thomas-laber&#34;&gt;&lt;strong&gt;Thomas Laber&lt;/strong&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Christoph and Thomas presented a way to use R within AWS lambda functions for model deployment - thus making R functions serverless. The reasons for using lambda functions is no administration, scalability and pay per user schemes (GB/secs) allowing for fast turnarounds.&lt;/p&gt;

&lt;p&gt;However, they also mentioned limitations and challenges deploying R as lambda functions including memory and space limits and suggested using managed container services as alternatives.&lt;/p&gt;

&lt;p&gt;Below the materials of their presentation:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Slides&lt;/strong&gt;: &lt;a href=&#34;/slides/Serverless Computing.pdf&#34;&gt;Serverless Computing&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Article&lt;/strong&gt;: &lt;a href=&#34;https://medium.com/bakdata/running-r-on-aws-lambda-9d40643551a6&#34;&gt;https://medium.com/bakdata/running-r-on-aws-lambda-9d40643551a6&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
</description>
        </item>
      
    
      
        <item>
          <title>Meetup 11-2016 Minutes</title>
          <link>/blog/2016/11/22/meetup-11-2016-minutes/</link>
          <pubDate>Tue, 22 Nov 2016 13:00:00 &#43;0200</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/11/22/meetup-11-2016-minutes/</guid>
          <description>&lt;p&gt;&lt;img src=&#34;/img/deepnet.png&#34; title=&#34;**Deep Neural Network**&#34; \
alt=&#34;**Deep Neural Network**&#34; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;p&gt;Deep learning with R using &lt;strong&gt;mxnet&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;During the November ViennaR Meetup Klaudius Kalcher gave an excellent talk about the &lt;strong&gt;mxnet&lt;/strong&gt; package. After covering some theoretical background he included examples to build deep neural networks from scratch.
Finally, Mario showed some time series forecasting examples already discussing during the October Meetup using the &lt;strong&gt;forecast&lt;/strong&gt; package.&lt;/p&gt;

&lt;h2 id=&#34;links&#34;&gt;Links&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;mxnet&lt;/strong&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;/slides/kalcher_deeplearning.pdf&#34;&gt;Slides&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;mxnet&lt;/strong&gt; R-package: &lt;a href=&#34;https://github.com/dmlc/mxnet/tree/master/R-package&#34;&gt;https://github.com/dmlc/mxnet/tree/master/R-package&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;dmlc mxnet for Deep Learning: &lt;a href=&#34;http://mxnet.io&#34;&gt;http://mxnet.io&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;forecast&lt;/strong&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;/slides/forecast_intro2.html&#34;&gt;Forecasting Crashcourse&lt;/a&gt; (&lt;a href=&#34;/slides/forecast_intro2.Rmd&#34;&gt;src&lt;/a&gt;)&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;forecast&lt;/strong&gt; package: &lt;a href=&#34;https://CRAN.R-project.org/package=forecast&#34;&gt;https://CRAN.R-project.org/package=forecast&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;Forecasting: principles and practice book: &lt;a href=&#34;https://www.otexts.org/fpp&#34;&gt;https://www.otexts.org/fpp&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Merry Christmas and a happy new coding year!&lt;/p&gt;

&lt;p&gt;Yours,&lt;/p&gt;

&lt;p&gt;ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 10-2016 Minutes</title>
          <link>/blog/2016/10/19/meetup-10-2016-minutes/</link>
          <pubDate>Wed, 19 Oct 2016 13:00:00 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/10/19/meetup-10-2016-minutes/</guid>
          <description>&lt;p&gt;&lt;img src=&#34;/img/meetup_201610_small.jpg&#34; title=&#34;**sparklyr**&#34; \
alt=&#34;**sparklyr**&#34; style=&#39;width:70%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;p&gt;A &lt;strong&gt;sparklyr&lt;/strong&gt; introduction by Roland Boubela.&lt;/p&gt;

&lt;p&gt;The October Meetup was hosted by Roland Boubela who gave an excellent talk about the new &lt;strong&gt;sparklyr&lt;/strong&gt; package. &lt;strong&gt;sparklyr&lt;/strong&gt; provides an easy-to-use interface to Apache Spark and can even be accessed using the latest and greatest version of RStudio (&lt;a href=&#34;https://www.rstudio.com/products/rstudio/download/preview&#34;&gt;preview release&lt;/a&gt;). It also covered a comparison with the existing &lt;strong&gt;SparkR&lt;/strong&gt; package.&lt;/p&gt;

&lt;h2 id=&#34;links&#34;&gt;Links&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;/slides/boubela_sparklyr.pdf&#34;&gt;Slides&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;sparklyr&lt;/strong&gt;: &lt;a href=&#34;http://spark.rstudio.com&#34;&gt;http://spark.rstudio.com&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;SparkR&lt;/strong&gt;: &lt;a href=&#34;http://spark.apache.org/docs/latest/sparkr.html&#34;&gt;http://spark.apache.org/docs/latest/sparkr.html&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;RStudio Preview: &lt;a href=&#34;https://www.rstudio.com/products/rstudio/download/preview&#34;&gt;https://www.rstudio.com/products/rstudio/download/preview&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;Apache Spark: &lt;a href=&#34;http://spark.apache.org&#34;&gt;http://spark.apache.org&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 09-2016 Minutes</title>
          <link>/blog/2016/09/27/meetup-09-2016-minutes/</link>
          <pubDate>Tue, 27 Sep 2016 13:00:00 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/09/27/meetup-09-2016-minutes/</guid>
          <description>&lt;p&gt;&lt;img src=&#34;/img/meetup_201609_small.jpg&#34; title=&#34;**Brainstorm Forecasting**&#34; \
alt=&#34;**Brainstorm Forecasting**&#34; style=&#39;width:70%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;p&gt;A brainstorming session about forecasting time series in R.&lt;/p&gt;

&lt;p&gt;First we discussed the topic data structures. Regular time series data can be represented using the &lt;em&gt;ts&lt;/em&gt; class which is already included in the basic R installation (as part of the &lt;strong&gt;stats&lt;/strong&gt; package). By contrast, irregular time series data can be represented by the popular &lt;strong&gt;zoo&lt;/strong&gt; objects (or &lt;strong&gt;xts&lt;/strong&gt;). Most (ARIMA-)forecasting methods require regular time series.&lt;/p&gt;

&lt;p&gt;The following R packages have been identified to support time series forecasting:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;https://CRAN.R-project.org/package=forecast&#34;&gt;&lt;strong&gt;forecast&lt;/strong&gt;&lt;/a&gt;: Excellent package by Rob Hyndman supporting all kinds of ARIMA models; even includes an automatic forecasting function (&lt;em&gt;auto.arima&lt;/em&gt;).&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://CRAN.R-project.org/package=tseries&#34;&gt;&lt;strong&gt;tseries&lt;/strong&gt;&lt;/a&gt;: Package supporting basic ARIMA model but also includes GARCH for volatility forecasting.&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://cran.r-project.org/package=x12&#34;&gt;&lt;strong&gt;x12&lt;/strong&gt;&lt;/a&gt;: Interface package to the X12-ARIMA program for sesonal adjustment&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://cran.r-project.org/package=seasonal&#34;&gt;&lt;strong&gt;seasonal&lt;/strong&gt;&lt;/a&gt;: Interface package to the X-13-ARIMA-SEATS program for sesonal adjustment.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&#34;projects&#34;&gt;Projects&lt;/h2&gt;

&lt;p&gt;The following potential projects covering simple time series forecasting examples have been discussed/requested:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Google Trends data (e.g. &lt;em&gt;Influenza&lt;/em&gt;)&lt;/li&gt;
&lt;li&gt;Company data, e.g. Revenues&lt;/li&gt;
&lt;li&gt;Bike Traffic in Vienna&lt;/li&gt;
&lt;li&gt;Stock Prices, Volatility (GARCH)&lt;/li&gt;
&lt;li&gt;Birdlife data, see data for Bavaria e.g.
&lt;a href=&#34;http://www.stunde-der-wintervoegel.de/index.php?id=auswertung&amp;amp;land=Bayern&#34;&gt;2016&lt;/a&gt;,
&lt;a href=&#34;http://sdw2015.lbv.de/index.php?id=auswertung&amp;amp;land=Bayern&#34;&gt;2015&lt;/a&gt;,
&lt;a href=&#34;http://sdw2014.lbv.de/index.php?id=auswertung&amp;amp;land=Bayern&#34;&gt;2014&lt;/a&gt;,
&lt;a href=&#34;http://sdw2013.lbv.de/index.php?id=auswertung&amp;amp;land=Bayern&#34;&gt;2013&lt;/a&gt;,
&lt;a href=&#34;http://sdw2012.lbv.de/index.php?id=auswertung&amp;amp;land=Bayern&#34;&gt;2012&lt;/a&gt;,
&lt;a href=&#34;http://sdw2011.lbv.de/index.php?id=auswertung&amp;amp;land=Bayern&#34;&gt;2011&lt;/a&gt;,
&lt;a href=&#34;http://sdw2010.lbv.de/index.php?id=auswertung&amp;amp;land=Bayern&#34;&gt;2010&lt;/a&gt;,
&lt;a href=&#34;http://sdw2009.lbv.de/index.php?id=auswertung&amp;amp;land=Bayern&#34;&gt;2009&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The projects will be presented during the November R-Meetup (contributions welcome!).&lt;/p&gt;

&lt;h2 id=&#34;announcements&#34;&gt;Announcements&lt;/h2&gt;

&lt;p&gt;The next meetup will cover the &lt;a href=&#34;http://spark.rstudio.com&#34;&gt;&lt;strong&gt;sparklyr&lt;/strong&gt;&lt;/a&gt; package presented by Roland Boubela.&lt;/p&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>igraph Applications for Transportation Networks</title>
          <link>/blog/2016/08/24/igraph-applications-for-transportation-networks/</link>
          <pubDate>Wed, 24 Aug 2016 22:20:00 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/08/24/igraph-applications-for-transportation-networks/</guid>
          <description>


&lt;p&gt;Materials from Max Leodolters talk in May covering the &lt;strong&gt;igraph&lt;/strong&gt; package.&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;require(igraph)
require(data.table)
require(ggplot2)&lt;/code&gt;&lt;/pre&gt;
&lt;div id=&#34;centrality---what-is-it&#34; class=&#34;section level1&#34;&gt;
&lt;h1&gt;Centrality - What is it?&lt;/h1&gt;
&lt;p&gt;Network Centrality gives you an idea of how important a vertex/node and edge/link in your network/graph &lt;span class=&#34;math inline&#34;&gt;\(N(V,E)\)&lt;/span&gt; respectively &lt;span class=&#34;math inline&#34;&gt;\(G(N,L)\)&lt;/span&gt; is. Examples:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Closeness&lt;/li&gt;
&lt;li&gt;Betweenness&lt;/li&gt;
&lt;li&gt;Eigenvalue&lt;/li&gt;
&lt;li&gt;degree&lt;/li&gt;
&lt;li&gt;…&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;Some formulas:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;Closeness Centrality &lt;span class=&#34;math display&#34;&gt;\[ C^{node}(n) = \frac{1}{\sum_{m \in N \setminus \{n\}}d(n,m)} \]&lt;/span&gt;&lt;/li&gt;
&lt;li&gt;Betweenness centrality &lt;span class=&#34;math display&#34;&gt;\[ B^{node}(n) = \sum_{m\neq o \in N\backslash \{n\} } \frac{\sigma_{mo}(n)}{\sigma_{mo}} \]&lt;/span&gt;&lt;/li&gt;
&lt;li&gt;Node-based Closeness for a link &lt;span class=&#34;math display&#34;&gt;\[ \ddot{C}^{link}(l_{n,m}) = \frac{C^{node}(n) + C^{node}(m)}{2} \]&lt;/span&gt;&lt;/li&gt;
&lt;li&gt;Node-based Betweenness for a link &lt;span class=&#34;math display&#34;&gt;\[ \ddot{B}^{link}(l_{n,m}) = \frac{B^{node}(n) + B^{node}(m)}{2} \]&lt;/span&gt;&lt;/li&gt;
&lt;li&gt;Link-based Betweenness for a link &lt;span class=&#34;math display&#34;&gt;\[ B^{link}(l) = \sum_{n \neq m \in N } \frac{\sigma_{nm}(l)}{\sigma_{nm}} \]&lt;/span&gt; *&lt;span class=&#34;math inline&#34;&gt;\(\sigma\)&lt;/span&gt; is the number of traversing OD relations, and &lt;span class=&#34;math inline&#34;&gt;\(d(n,m)\)&lt;/span&gt; is the distance of the shortest route from &lt;span class=&#34;math inline&#34;&gt;\(n\)&lt;/span&gt; to &lt;span class=&#34;math inline&#34;&gt;\(m\)&lt;/span&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;p&gt;What to select, &lt;span class=&#34;math inline&#34;&gt;\(\ddot{B}^{link}(l_{n,m})\)&lt;/span&gt; or &lt;span class=&#34;math inline&#34;&gt;\(B^{link}(l)\)&lt;/span&gt;?&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;igraph-for-centralities&#34; class=&#34;section level1&#34;&gt;
&lt;h1&gt;igraph for centralities&lt;/h1&gt;
&lt;div id=&#34;node-betweenness&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Node Betweenness&lt;/h2&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;g &amp;lt;- make_star(5, mode = &amp;quot;undirected&amp;quot;, center=3)
V(g)$name &amp;lt;- letters[1:length(V(g))]
plot(g)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-2-1.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)
b &amp;lt;- betweenness(g, directed=F)
V(g)$name &amp;lt;- b
plot(g)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-2-2.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;p&gt;Edge Betweenness:&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;g &amp;lt;- make_star(5, mode = &amp;quot;undirected&amp;quot;, center=3)
V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)

bg &amp;lt;- edge_betweenness(g, directed=F)
plot(g, edge.label = bg)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-3-1.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;local-vs.global&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Local vs. Global&lt;/h2&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;g &amp;lt;- make_graph(c(1, 2, 2, 3,#horizontal connections
                  4,5,5,6,6,7,
                  1,6,#vertical connections
                  2,7), directed = FALSE)
#g &amp;lt;- make_star(10, mode = &amp;quot;undirected&amp;quot;, center=3)
V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)
set.seed(2)
lay &amp;lt;- layout.auto(g)
bl &amp;lt;- estimate_edge_betweenness(g, cutoff = 1.5, directed=F);bl&lt;/code&gt;&lt;/pre&gt;
&lt;pre&gt;&lt;code&gt;## [1] 3 3 2 4 3 3 3&lt;/code&gt;&lt;/pre&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;bg &amp;lt;- edge_betweenness(g, directed=F)


# set plotting parameters
vs &amp;lt;- 15# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
hd &amp;lt;- paste(rep(&amp;quot; &amp;quot;,0), collapse=&amp;quot;&amp;quot;)
cm &amp;lt;- 3
vc &amp;lt;- &amp;quot;orange&amp;quot;#gray(0.8)#palette(&amp;quot;default&amp;quot;)#&amp;quot;grey&amp;quot;#vertex.color

#windows(width = 18, height=6)
#par(mfrow=c(1,3))

plot(g, edge.label=paste(hd, round(E(g)$weight,1)), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-4-1.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;plot(g, edge.label=paste(hd, round(bl,1)),          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Local Betweenness&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-4-2.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;plot(g, edge.label=paste(hd, round(bg,1)),          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) Global Betweenness&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-4-3.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;node-based-vs.link-based-betweenness-for-links&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Node-based vs. link-based Betweenness for Links&lt;/h2&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;g &amp;lt;- make_graph(c(1,2, 2,3, 3,4, 4,5,#horizontal connections
                  6,7, 7,8, 8,9, 9,10,
                  2,7,#vertical connections
                  3,8,
                  4,9), directed = FALSE)

V(g)$name &amp;lt;- letters[1:length(V(g))]
set.seed(1210)
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;, index= c(10), value= 2)
#plot(g,edge.label=paste(hd, round(E(g)$weight,1)))
lay &amp;lt;- layout.auto(g)
eb &amp;lt;- edge_betweenness(g, directed=F)
vb &amp;lt;- betweenness(g, directed=F)

tmp1 &amp;lt;- as.data.table(as.data.frame(cbind(get.edgelist(g))))
tmp1[,id:=1:.N]
tmp2 &amp;lt;- as.data.table(data.frame(value=vb, name=names(vb)))
setnames(tmp1, &amp;quot;V1&amp;quot;, &amp;quot;name&amp;quot;)
tmp3 &amp;lt;- merge(tmp1, tmp2, by =&amp;quot;name&amp;quot;)
setnames(tmp3, &amp;quot;name&amp;quot;, &amp;quot;V1&amp;quot;)
setnames(tmp3, &amp;quot;V2&amp;quot;, &amp;quot;name&amp;quot;)
tmp4 &amp;lt;- merge(tmp3, tmp2, by =&amp;quot;name&amp;quot;)
setnames(tmp4,  &amp;quot;name&amp;quot;, &amp;quot;V2&amp;quot;)
tmp4[,m:=mean(c(value.x, value.y)) , by=c(&amp;quot;V2&amp;quot;,&amp;quot;V1&amp;quot;)]
evb &amp;lt;- tmp4[order(id),list(V1,V2,m)]$m


# set plotting parameters
vs &amp;lt;- 15# vertex.size
ec &amp;lt;- &amp;quot;grey&amp;quot;#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
hd &amp;lt;- paste(rep(&amp;quot; &amp;quot;,0), collapse=&amp;quot;&amp;quot;)
cm &amp;lt;- 3
vc &amp;lt;- &amp;quot;orange&amp;quot;# vertex.color

X11(width = 18, height=6)
#par(mfrow=c(1,3))

plot(g, edge.label=paste(hd, round(E(g)$weight,1)), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;,cex.main=cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-5-1.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;plot(g, edge.label=paste(hd, round(eb,1)),          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Link based Betweenness&amp;quot;,cex.main=cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-5-2.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;V(g)$name &amp;lt;- paste(vb)

plot(g, edge.label=paste(hd, round(evb,1)),          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) Node based Betweenness&amp;quot; ,cex.main=cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-5-3.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;bugsfeaturesspecial-experiences&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Bugs/Features/Special experiences&lt;/h2&gt;
&lt;ul&gt;
&lt;li&gt;demonstrate the parameter ‘lay’ with setting it NULL&lt;/li&gt;
&lt;li&gt;cutoff parameter&lt;/li&gt;
&lt;li&gt;nodes need to be named for using induced.subgraph(), otherwise problems in matching results&lt;/li&gt;
&lt;/ul&gt;
&lt;div id=&#34;cutoff&#34; class=&#34;section level3&#34;&gt;
&lt;h3&gt;cutoff&lt;/h3&gt;
&lt;p&gt;stackoverflow: &lt;a href=&#34;http://stackoverflow.com/questions/34180036/how-does-the-cutoff-parameter-influence-betweenness-calculation&#34; class=&#34;uri&#34;&gt;http://stackoverflow.com/questions/34180036/how-does-the-cutoff-parameter-influence-betweenness-calculation&lt;/a&gt;&lt;/p&gt;
&lt;p&gt;discussion: &lt;a href=&#34;https://lists.nongnu.org/archive/html/igraph-help/2012-11/msg00083.html&#34; class=&#34;uri&#34;&gt;https://lists.nongnu.org/archive/html/igraph-help/2012-11/msg00083.html&lt;/a&gt;&lt;/p&gt;
&lt;p&gt;github project: &lt;a href=&#34;https://github.com/maxar/igraph_testing/blob/master/test_vertex_betweenness_local.R&#34; class=&#34;uri&#34;&gt;https://github.com/maxar/igraph_testing/blob/master/test_vertex_betweenness_local.R&lt;/a&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;require(igraph)
packageVersion(&amp;quot;igraph&amp;quot;) # my installed version is 1.0.1&lt;/code&gt;&lt;/pre&gt;
&lt;pre&gt;&lt;code&gt;## [1] &amp;#39;1.2.2&amp;#39;&lt;/code&gt;&lt;/pre&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;#----------------------------------------------------------------------------------------
#---------------    EXAMPLE 1    --------------------------------------------------------
#----------------------------------------------------------------------------------------

#--- set plotting parameters
vs &amp;lt;- 25# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
cm &amp;lt;- 2
vc &amp;lt;- &amp;quot;orange&amp;quot;


g &amp;lt;- make_graph(c(1, 2, 2, 3, 3, 4, 4,5, 5, 6, 6, 7) , directed = FALSE)

set.seed(1210)
lay &amp;lt;- layout.auto(g)

graphics.off()
X11(width = 30, height=10)
#par(mfrow=c(1,3))

V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= c(2,3,5,5,3,2))
plot(g, edge.label=round(E(g)$weight,1), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-6-1.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;bl.1 &amp;lt;- estimate_betweenness(g, cutoff=10)
V(g)$name &amp;lt;- bl.1
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Local Betweenness\ncutoff 10&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-6-2.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;bl.2 &amp;lt;- estimate_betweenness(g, cutoff=Inf)
V(g)$name &amp;lt;- bl.2
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Global Betweenness\ncutoff Inf&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-6-3.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;# Comments and Discussion:
# This simple example illustrates, why it is questionable
# that the cutoff parameter actually measures distance independent of the edge weights
#
# 1. define edge weights as in plot (a)
# 2. calculate vertex betweenness with cutoff 10 (b)
# 3. calculate vertex betweenness with cutoff Inf (c)






#----------------------------------------------------------------------------------------
#---------------    EXAMPLE 2    --------------------------------------------------------
#----------------------------------------------------------------------------------------

#--- set plotting parameters
vs &amp;lt;- 25# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
cm &amp;lt;- 2
vc &amp;lt;- &amp;quot;orange&amp;quot;


set.seed(1210)
g &amp;lt;- make_ring(8)
lay &amp;lt;- layout.auto(g)

graphics.off()
X11(width = 12, height=12)
par(mfrow=c(2,2))

V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;, value= c(.1 , .2, rep(.1, 2), rep(0.2,4)))
plot(g, edge.label=round(E(g)$weight,1), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)

bl.1 &amp;lt;- estimate_betweenness(g, cutoff=0.29)
V(g)$name &amp;lt;- bl.1
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Local Betweenness\ncutoff 0.29&amp;quot;, cex.main = cm)

V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;, value=.2)
plot(g, edge.label=round(E(g)$weight,1), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) new weights&amp;quot;, cex.main = cm)

bl.2 &amp;lt;- estimate_betweenness(g, cutoff=0.29)
V(g)$name &amp;lt;- bl.2
plot(g,          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(d) Local Betweenness\ncutoff 0.29&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-6-4.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;# Comments and Discussion:
# This simple example illustrates, that vertex betweenness values change,
# if the edge weights are changed and the cutoff parameters are chosen constant.
# Ring shape to reconstruct the calculation easier;
#
# 1. define edge weights as in plot (a)
# 2. calculate vertex betweenness with cutoff 0.29 (b)
# 3. define new edge weights as in plot (c)
# 4. calculate vertex betweenness with cutoff 0.29 (d)





#----------------------------------------------------------------------------------------
#---------------    EXAMPLE 3    --------------------------------------------------------
#----------------------------------------------------------------------------------------

#--- set plotting parameters
vs &amp;lt;- 20# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 1# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 1#vertex.label.cex
ew &amp;lt;- 2#edge.width
cm &amp;lt;- 2
vc &amp;lt;- &amp;quot;orange&amp;quot;


set.seed(1210)
g &amp;lt;- make_ring(20)
lay &amp;lt;- layout.auto(g)

graphics.off()
#windows(width = 30, height=30)
#par(mfrow=c(2,2))

V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;, value= c(rep(2:3,10)))
plot(g, edge.label=round(E(g)$weight,1), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)

bl.1 &amp;lt;- estimate_betweenness(g, cutoff=6)
V(g)$name &amp;lt;- bl.1
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Local Betweenness\ncutoff 6&amp;quot;, cex.main = cm)

bl.2 &amp;lt;- estimate_betweenness(g, cutoff=9)
V(g)$name &amp;lt;- bl.2
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) Local Betweenness\ncutoff 9&amp;quot;, cex.main = cm)

bl.3 &amp;lt;- estimate_betweenness(g, cutoff=Inf)
V(g)$name &amp;lt;- bl.3
plot(g,          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(d) Global Betweenness\ncutoff Inf&amp;quot;, cex.main = cm)


# Comments and Discussion:
# This simple example illustrates, that vertex betweenness values change,
# if the edge weights are constant and the cutoff parameters change.
# Ring shape to reconstruct the calculation easier;
#
# 1. define edge weights as in plot (a)
# 2. calculate vertex betweenness with cutoff 6 (b)
# 3. calculate vertex betweenness with cutoff 9 (c)
# 4. calculate vertex betweenness with cutoff Inf (d)&lt;/code&gt;&lt;/pre&gt;
&lt;/div&gt;
&lt;div id=&#34;normalization&#34; class=&#34;section level3&#34;&gt;
&lt;h3&gt;normalization&lt;/h3&gt;
&lt;p&gt;stackoverflow: &lt;a href=&#34;http://stackoverflow.com/questions/34000326/normalized-local-closeness-centrality-in-r-igraphestimate-closeness&#34; class=&#34;uri&#34;&gt;http://stackoverflow.com/questions/34000326/normalized-local-closeness-centrality-in-r-igraphestimate-closeness&lt;/a&gt;&lt;/p&gt;
&lt;p&gt;I am trying to calculate a normalized local closeness centrality. But setting the parameter ‘normalized = T’ for igraph::estimate_closeness() only multiplies the results with (N-1). Is it possible to define N_i for vertex i individually by it’s neighborhood, that is determined by the cut-off parameter (and of course by the graph itself)?&lt;/p&gt;
&lt;p&gt;The mini example demonstrates, that setting the parameter ‘normalized=T’ does not work for this purpose. It normalizes with one and the same (N-1) for all vertices.&lt;/p&gt;
&lt;p&gt;Thanks for help.&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;set.seed(1210)
require(&amp;#39;igraph&amp;#39;)
g &amp;lt;- random.graph.game(20,3/10)
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)
cnt &amp;lt;- estimate_closeness(g, cutoff = 3, normalized = T );cnt&lt;/code&gt;&lt;/pre&gt;
&lt;pre&gt;&lt;code&gt;## Warning in estimate_closeness(g, cutoff = 3, normalized = T): At
## centrality.c:2617 :closeness centrality is not well-defined for
## disconnected graphs&lt;/code&gt;&lt;/pre&gt;
&lt;pre&gt;&lt;code&gt;##  [1] 0.6333333 0.2467532 0.3220339 0.5000000 0.5277778 0.5588235 0.6129032
##  [8] 0.5135135 0.5428571 0.5588235 0.4634146 0.5277778 0.6129032 0.5000000
## [15] 0.6333333 0.5000000 0.3392857 0.5135135 0.5135135 0.5135135&lt;/code&gt;&lt;/pre&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;cnf &amp;lt;- estimate_closeness(g, cutoff = 3, normalized = F );cnf&lt;/code&gt;&lt;/pre&gt;
&lt;pre&gt;&lt;code&gt;## Warning in estimate_closeness(g, cutoff = 3, normalized = F): At
## centrality.c:2617 :closeness centrality is not well-defined for
## disconnected graphs&lt;/code&gt;&lt;/pre&gt;
&lt;pre&gt;&lt;code&gt;##  [1] 0.03333333 0.01298701 0.01694915 0.02631579 0.02777778 0.02941176
##  [7] 0.03225806 0.02702703 0.02857143 0.02941176 0.02439024 0.02777778
## [13] 0.03225806 0.02631579 0.03333333 0.02631579 0.01785714 0.02702703
## [19] 0.02702703 0.02702703&lt;/code&gt;&lt;/pre&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;#print results
cbind(cnf * (length(V(g))-1), cnt)&lt;/code&gt;&lt;/pre&gt;
&lt;pre&gt;&lt;code&gt;##                       cnt
##  [1,] 0.6333333 0.6333333
##  [2,] 0.2467532 0.2467532
##  [3,] 0.3220339 0.3220339
##  [4,] 0.5000000 0.5000000
##  [5,] 0.5277778 0.5277778
##  [6,] 0.5588235 0.5588235
##  [7,] 0.6129032 0.6129032
##  [8,] 0.5135135 0.5135135
##  [9,] 0.5428571 0.5428571
## [10,] 0.5588235 0.5588235
## [11,] 0.4634146 0.4634146
## [12,] 0.5277778 0.5277778
## [13,] 0.6129032 0.6129032
## [14,] 0.5000000 0.5000000
## [15,] 0.6333333 0.6333333
## [16,] 0.5000000 0.5000000
## [17,] 0.3392857 0.3392857
## [18,] 0.5135135 0.5135135
## [19,] 0.5135135 0.5135135
## [20,] 0.5135135 0.5135135&lt;/code&gt;&lt;/pre&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;sum(abs(cnf * (length(V(g))-1) - cnt))&lt;/code&gt;&lt;/pre&gt;
&lt;pre&gt;&lt;code&gt;## [1] 8.326673e-17&lt;/code&gt;&lt;/pre&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;#for visualization
V(g)$name &amp;lt;- paste(&amp;quot;v&amp;quot;, 1:length(V(g)), sep=&amp;quot;&amp;quot;)#letters[1:length(V(g))]
set.seed(2)
lay &amp;lt;- layout.auto(g)

## set plotting parameters
vs &amp;lt;- 15# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
hd &amp;lt;- paste(rep(&amp;quot; &amp;quot;,0), collapse=&amp;quot;&amp;quot;)
cm &amp;lt;- 3
vc &amp;lt;- &amp;quot;orange&amp;quot;#gray(0.8)#palette(&amp;quot;default&amp;quot;)#&amp;quot;grey&amp;quot;#vertex.color

X11(width = 18, height=6)
#par(mfrow=c(1,3))

plot(g, edge.label=paste(hd, round(E(g)$weight,1)), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-7-1.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;V(g)$name &amp;lt;- paste(round(cnt,3))
plot(g,          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) local closeness normalized&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-7-2.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;pre class=&#34;r&#34;&gt;&lt;code&gt;V(g)$name &amp;lt;- paste(round(cnf,3))
plot(g,          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) local closeness&amp;quot;, cex.main = cm)&lt;/code&gt;&lt;/pre&gt;
&lt;p&gt;&lt;img src=&#34;/post/20160524_igraph_applications_files/figure-html/unnamed-chunk-7-3.png&#34; width=&#34;672&#34; /&gt;&lt;/p&gt;
&lt;div id=&#34;my-solution&#34; class=&#34;section level4&#34;&gt;
&lt;h4&gt;My solution&lt;/h4&gt;
&lt;p&gt;easy workaround for local centralities: define neighborhood for each node, calcualte ‘global’ centralities for these neighborhoods and normalize it individually.&lt;/p&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;div id=&#34;what-for-did-we-use-centralities&#34; class=&#34;section level1&#34;&gt;
&lt;h1&gt;What for did we use centralities?&lt;/h1&gt;
&lt;p&gt;Thanks to Anita Graser for the beautiful plots!&lt;/p&gt;
&lt;p&gt;&lt;span class=&#34;math display&#34;&gt;\[\hat{y}_{t,\gamma}(s,b,c) = \beta_t  + \beta_{\gamma,s} \cdot s  + \beta_{\gamma,b} \cdot b + \beta_{\gamma,c} \cdot c + \beta_{\gamma,bc} \cdot b \cdot c\]&lt;/span&gt;&lt;/p&gt;
&lt;p&gt;some plots:&lt;/p&gt;
&lt;div id=&#34;global-betweenness&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Global Betweenness&lt;/h2&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic1.png&#34; title=&#34;**Global Betweenness** \
*Global Betweenness*&#34; alt=&#34;**Global Betweenness** \
*Global Betweenness*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;local-betweenness&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Local Betweenness&lt;/h2&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic2.png&#34; title=&#34;**Local Betweenness** \
*Local Betweenness*&#34; alt=&#34;**Local Betweenness** \
*Local Betweenness*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;global-closeness&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Global Closeness&lt;/h2&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic3.png&#34; title=&#34;**Global Closeness** \
*Global Closeness*&#34; alt=&#34;**Global Closeness** \
*Global Closeness*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;local-closeness&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Local Closeness&lt;/h2&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic4.png&#34; title=&#34;**Local Closeness** \
*Local Closeness*&#34; alt=&#34;**Local Closeness** \
*Local Closeness*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;real-world-difference-of-node-based-and-link-based-betweenness-for-a-link&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Real world difference of node-based and link-based Betweenness for a link&lt;/h2&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic5.png&#34; title=&#34;**Real world difference 1** \
*Real world difference 1*&#34; alt=&#34;**Real world difference 1** \
*Real world difference 1*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic6.png&#34; title=&#34;**Real world difference 2** \
*Real world difference 2*&#34; alt=&#34;**Real world difference 2** \
*Real world difference 2*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;improvement-travel-time-model&#34; class=&#34;section level2&#34;&gt;
&lt;h2&gt;Improvement travel time model&lt;/h2&gt;
&lt;div id=&#34;mape-difference-global-minus-local&#34; class=&#34;section level3&#34;&gt;
&lt;h3&gt;MAPE Difference Global minus local&lt;/h3&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic7.png&#34; title=&#34;**MAPE Difference** \
*MAPE Difference*&#34; alt=&#34;**MAPE Difference** \
*MAPE Difference*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;detail-mape-base&#34; class=&#34;section level3&#34;&gt;
&lt;h3&gt;Detail MAPE base&lt;/h3&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic8.png&#34; title=&#34;**MAPE base** \
*MAPE base*&#34; alt=&#34;**MAPE base** \
*MAPE base*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;div id=&#34;detail-mape-with-local-centralities&#34; class=&#34;section level3&#34;&gt;
&lt;h3&gt;Detail MAPE with local centralities&lt;/h3&gt;
&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic9.png&#34; title=&#34;**MAPE local** \
*MAPE local*&#34; alt=&#34;**MAPE local** \ *MAPE local*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;
&lt;/div&gt;
&lt;/div&gt;
&lt;/div&gt;
</description>
        </item>
      
    
      
        <item>
          <title>igraph Applications for Transportation Networks</title>
          <link>/blog/2016/08/24/igraph-applications-for-transportation-networks/</link>
          <pubDate>Wed, 24 Aug 2016 22:20:00 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/08/24/igraph-applications-for-transportation-networks/</guid>
          <description>&lt;p&gt;Materials from Max Leodolters talk in May covering the &lt;strong&gt;igraph&lt;/strong&gt;
package.&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;require(igraph)
require(data.table)
require(ggplot2)
&lt;/code&gt;&lt;/pre&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;#centrality---what-is-it&#34;&gt;Centrality - What is it?&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#igraph-for-centralities&#34;&gt;igraph for centralities&lt;/a&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;#node-betweenness&#34;&gt;Node Betweenness&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#local-vs.-global&#34;&gt;Local vs. Global&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#node-based-vs.-link-based-betweenness-for-links&#34;&gt;Node-based vs. link-based Betweenness for
Links&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#bugsfeaturesspecial-experiences&#34;&gt;Bugs/Features/Special
experiences&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#what-for-did-we-use-centralities&#34;&gt;What for did we use
centralities?&lt;/a&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;#global-betweenness&#34;&gt;Global Betweenness&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#local-betweenness&#34;&gt;Local Betweenness&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#global-closeness&#34;&gt;Global Closeness&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#local-closeness&#34;&gt;Local Closeness&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#real-world-difference-of-node-based-and-link-based-betweenness-for-a-link&#34;&gt;Real world difference of node-based and link-based Betweenness
for a
link&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;#improvement-travel-time-model&#34;&gt;Improvement travel time model&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h1 id=&#34;centrality-what-is-it&#34;&gt;Centrality - What is it?&lt;/h1&gt;

&lt;p&gt;Network Centrality gives you an idea of how important a vertex/node and
edge/link in your network/graph &lt;em&gt;N&lt;/em&gt;(&lt;em&gt;V&lt;/em&gt;, &lt;em&gt;E&lt;/em&gt;) respectively &lt;em&gt;G&lt;/em&gt;(&lt;em&gt;N&lt;/em&gt;, &lt;em&gt;L&lt;/em&gt;)
is. Examples:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Closeness&lt;/li&gt;
&lt;li&gt;Betweenness&lt;/li&gt;
&lt;li&gt;Eigenvalue&lt;/li&gt;
&lt;li&gt;degree&lt;/li&gt;
&lt;li&gt;&amp;hellip;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Some formulas:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Closeness Centrality
$$ C^{node}(n) = \frac{1}{\sum_{m \in N \setminus \{n\}}d(n,m)} $$&lt;/li&gt;
&lt;li&gt;Betweenness centrality
$$ B^{node}(n) = \sum_{m\neq o \in N\backslash \{n\} } \frac{\sigma_{mo}(n)}{\sigma_{mo}} $$&lt;/li&gt;
&lt;li&gt;Node-based Closeness for a link
$$ \ddot{C}^{link}(l_{n,m}) = \frac{C^{node}(n) + C^{node}(m)}{2} $$&lt;/li&gt;
&lt;li&gt;Node-based Betweenness for a link
$$ \ddot{B}^{link}(l_{n,m}) = \frac{B^{node}(n) + B^{node}(m)}{2} $$&lt;/li&gt;
&lt;li&gt;Link-based Betweenness for a link
$$ B^{link}(l) = \sum_{n \neq m \in N } \frac{\sigma_{nm}(l)}{\sigma_{nm}} $$
 *&lt;em&gt;σ&lt;/em&gt; is the number of traversing OD relations, and &lt;em&gt;d&lt;/em&gt;(&lt;em&gt;n&lt;/em&gt;, &lt;em&gt;m&lt;/em&gt;)
is the distance of the shortest route from &lt;em&gt;n&lt;/em&gt; to &lt;em&gt;m&lt;/em&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;What to select, $\ddot{B}^{link}(l_{n,m})$ or $ B^{link}(l)$?&lt;/p&gt;

&lt;h1 id=&#34;igraph-for-centralities&#34;&gt;igraph for centralities&lt;/h1&gt;

&lt;h2 id=&#34;node-betweenness&#34;&gt;Node Betweenness&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;g &amp;lt;- make_star(5, mode = &amp;quot;undirected&amp;quot;, center=3)
V(g)$name &amp;lt;- letters[1:length(V(g))]
plot(g)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-2-1.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)
b &amp;lt;- betweenness(g, directed=F)
V(g)$name &amp;lt;- b
plot(g)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-2-2.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;p&gt;Edge Betweenness:&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;g &amp;lt;- make_star(5, mode = &amp;quot;undirected&amp;quot;, center=3)
V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)

bg &amp;lt;- edge_betweenness(g, directed=F)
plot(g, edge.label = bg)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-3-1.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;h2 id=&#34;local-vs-global&#34;&gt;Local vs. Global&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;g &amp;lt;- make_graph(c(1, 2, 2, 3,#horizontal connections
                  4,5,5,6,6,7,
                  1,6,#vertical connections
                  2,7), directed = FALSE)
#g &amp;lt;- make_star(10, mode = &amp;quot;undirected&amp;quot;, center=3)
V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)
set.seed(2)
lay &amp;lt;- layout.auto(g)
bl &amp;lt;- estimate_edge_betweenness(g, cutoff = 1.5, directed=F);bl

## [1] 3 3 2 4 3 3 3

bg &amp;lt;- edge_betweenness(g, directed=F)


# set plotting parameters
vs &amp;lt;- 15# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
hd &amp;lt;- paste(rep(&amp;quot; &amp;quot;,0), collapse=&amp;quot;&amp;quot;)
cm &amp;lt;- 3
vc &amp;lt;- &amp;quot;orange&amp;quot;#gray(0.8)#palette(&amp;quot;default&amp;quot;)#&amp;quot;grey&amp;quot;#vertex.color

#windows(width = 18, height=6)
#par(mfrow=c(1,3))

plot(g, edge.label=paste(hd, round(E(g)$weight,1)), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-4-1.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;plot(g, edge.label=paste(hd, round(bl,1)),          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Local Betweenness&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-4-2.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;plot(g, edge.label=paste(hd, round(bg,1)),          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) Global Betweenness&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-4-3.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;h2 id=&#34;node-based-vs-link-based-betweenness-for-links&#34;&gt;Node-based vs. link-based Betweenness for Links&lt;/h2&gt;

&lt;pre&gt;&lt;code&gt;g &amp;lt;- make_graph(c(1,2, 2,3, 3,4, 4,5,#horizontal connections
                  6,7, 7,8, 8,9, 9,10,
                  2,7,#vertical connections
                  3,8,
                  4,9), directed = FALSE)

V(g)$name &amp;lt;- letters[1:length(V(g))]
set.seed(1210)
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;, index= c(10), value= 2)
#plot(g,edge.label=paste(hd, round(E(g)$weight,1)))
lay &amp;lt;- layout.auto(g)
eb &amp;lt;- edge_betweenness(g, directed=F)
vb &amp;lt;- betweenness(g, directed=F)

tmp1 &amp;lt;- as.data.table(as.data.frame(cbind(get.edgelist(g))))
tmp1[,id:=1:.N]
tmp2 &amp;lt;- as.data.table(data.frame(value=vb, name=names(vb)))
setnames(tmp1, &amp;quot;V1&amp;quot;, &amp;quot;name&amp;quot;)
tmp3 &amp;lt;- merge(tmp1, tmp2, by =&amp;quot;name&amp;quot;)
setnames(tmp3, &amp;quot;name&amp;quot;, &amp;quot;V1&amp;quot;)
setnames(tmp3, &amp;quot;V2&amp;quot;, &amp;quot;name&amp;quot;)
tmp4 &amp;lt;- merge(tmp3, tmp2, by =&amp;quot;name&amp;quot;)
setnames(tmp4,  &amp;quot;name&amp;quot;, &amp;quot;V2&amp;quot;)
tmp4[,m:=mean(c(value.x, value.y)) , by=c(&amp;quot;V2&amp;quot;,&amp;quot;V1&amp;quot;)]
evb &amp;lt;- tmp4[order(id),list(V1,V2,m)]$m


# set plotting parameters
vs &amp;lt;- 15# vertex.size
ec &amp;lt;- &amp;quot;grey&amp;quot;#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
hd &amp;lt;- paste(rep(&amp;quot; &amp;quot;,0), collapse=&amp;quot;&amp;quot;)
cm &amp;lt;- 3
vc &amp;lt;- &amp;quot;orange&amp;quot;# vertex.color

windows(width = 18, height=6)
#par(mfrow=c(1,3))

plot(g, edge.label=paste(hd, round(E(g)$weight,1)), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;,cex.main=cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-5-1.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;plot(g, edge.label=paste(hd, round(eb,1)),          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Link based Betweenness&amp;quot;,cex.main=cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-5-2.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;V(g)$name &amp;lt;- paste(vb)

plot(g, edge.label=paste(hd, round(evb,1)),          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) Node based Betweenness&amp;quot; ,cex.main=cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-5-3.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;h2 id=&#34;bugs-features-special-experiences&#34;&gt;Bugs/Features/Special experiences&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;demonstrate the parameter &amp;lsquo;lay&amp;rsquo; with setting it NULL&lt;/li&gt;
&lt;li&gt;cutoff parameter&lt;/li&gt;
&lt;li&gt;nodes need to be named for using induced.subgraph(), otherwise
problems in matching results&lt;/li&gt;
&lt;/ul&gt;

&lt;h3 id=&#34;cutoff&#34;&gt;cutoff&lt;/h3&gt;

&lt;p&gt;stackoverflow:
&lt;a href=&#34;http://stackoverflow.com/questions/34180036/how-does-the-cutoff-parameter-influence-betweenness-calculation&#34;&gt;http://stackoverflow.com/questions/34180036/how-does-the-cutoff-parameter-influence-betweenness-calculation&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;discussion:
&lt;a href=&#34;https://lists.nongnu.org/archive/html/igraph-help/2012-11/msg00083.html&#34;&gt;https://lists.nongnu.org/archive/html/igraph-help/2012-11/msg00083.html&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;github project:
&lt;a href=&#34;https://github.com/maxar/igraph_testing/blob/master/test_vertex_betweenness_local.R&#34;&gt;https://github.com/maxar/igraph_testing/blob/master/test_vertex_betweenness_local.R&lt;/a&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;require(igraph)
packageVersion(&amp;quot;igraph&amp;quot;) # my installed version is 1.0.1

## [1] &#39;1.0.1&#39;

#----------------------------------------------------------------------------------------
#---------------    EXAMPLE 1    --------------------------------------------------------
#----------------------------------------------------------------------------------------

#--- set plotting parameters
vs &amp;lt;- 25# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
cm &amp;lt;- 2
vc &amp;lt;- &amp;quot;orange&amp;quot;


g &amp;lt;- make_graph(c(1, 2, 2, 3, 3, 4, 4,5, 5, 6, 6, 7) , directed = FALSE)

set.seed(1210)
lay &amp;lt;- layout.auto(g)

graphics.off()
windows(width = 30, height=10)
#par(mfrow=c(1,3))

V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= c(2,3,5,5,3,2))
plot(g, edge.label=round(E(g)$weight,1), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-6-1.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;bl.1 &amp;lt;- estimate_betweenness(g, cutoff=10)
V(g)$name &amp;lt;- bl.1
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Local Betweenness\ncutoff 10&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-6-2.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;bl.2 &amp;lt;- estimate_betweenness(g, cutoff=Inf)
V(g)$name &amp;lt;- bl.2
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Global Betweenness\ncutoff Inf&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-6-3.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;# Comments and Discussion:
# This simple example illustrates, why it is questionable
# that the cutoff parameter actually measures distance independent of the edge weights
#
# 1. define edge weights as in plot (a)
# 2. calculate vertex betweenness with cutoff 10 (b)
# 3. calculate vertex betweenness with cutoff Inf (c)






#----------------------------------------------------------------------------------------
#---------------    EXAMPLE 2    --------------------------------------------------------
#----------------------------------------------------------------------------------------

#--- set plotting parameters
vs &amp;lt;- 25# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
cm &amp;lt;- 2
vc &amp;lt;- &amp;quot;orange&amp;quot;


set.seed(1210)
g &amp;lt;- make_ring(8)
lay &amp;lt;- layout.auto(g)

graphics.off()
windows(width = 12, height=12)
par(mfrow=c(2,2))

V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;, value= c(.1 , .2, rep(.1, 2), rep(0.2,4)))
plot(g, edge.label=round(E(g)$weight,1), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)

bl.1 &amp;lt;- estimate_betweenness(g, cutoff=0.29)
V(g)$name &amp;lt;- bl.1
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Local Betweenness\ncutoff 0.29&amp;quot;, cex.main = cm)

V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;, value=.2)
plot(g, edge.label=round(E(g)$weight,1), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) new weights&amp;quot;, cex.main = cm)

bl.2 &amp;lt;- estimate_betweenness(g, cutoff=0.29)
V(g)$name &amp;lt;- bl.2
plot(g,          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(d) Local Betweenness\ncutoff 0.29&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-6-4.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;# Comments and Discussion:
# This simple example illustrates, that vertex betweenness values change,
# if the edge weights are changed and the cutoff parameters are chosen constant.
# Ring shape to reconstruct the calculation easier;
#
# 1. define edge weights as in plot (a)
# 2. calculate vertex betweenness with cutoff 0.29 (b)
# 3. define new edge weights as in plot (c)
# 4. calculate vertex betweenness with cutoff 0.29 (d)





#----------------------------------------------------------------------------------------
#---------------    EXAMPLE 3    --------------------------------------------------------
#----------------------------------------------------------------------------------------

#--- set plotting parameters
vs &amp;lt;- 20# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 1# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 1#vertex.label.cex
ew &amp;lt;- 2#edge.width
cm &amp;lt;- 2
vc &amp;lt;- &amp;quot;orange&amp;quot;


set.seed(1210)
g &amp;lt;- make_ring(20)
lay &amp;lt;- layout.auto(g)

graphics.off()
#windows(width = 30, height=30)
#par(mfrow=c(2,2))

V(g)$name &amp;lt;- letters[1:length(V(g))]
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;, value= c(rep(2:3,10)))
plot(g, edge.label=round(E(g)$weight,1), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)

bl.1 &amp;lt;- estimate_betweenness(g, cutoff=6)
V(g)$name &amp;lt;- bl.1
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) Local Betweenness\ncutoff 6&amp;quot;, cex.main = cm)

bl.2 &amp;lt;- estimate_betweenness(g, cutoff=9)
V(g)$name &amp;lt;- bl.2
plot(g,        layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) Local Betweenness\ncutoff 9&amp;quot;, cex.main = cm)

bl.3 &amp;lt;- estimate_betweenness(g, cutoff=Inf)
V(g)$name &amp;lt;- bl.3
plot(g,          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(d) Global Betweenness\ncutoff Inf&amp;quot;, cex.main = cm)


# Comments and Discussion:
# This simple example illustrates, that vertex betweenness values change,
# if the edge weights are constant and the cutoff parameters change.
# Ring shape to reconstruct the calculation easier;
#
# 1. define edge weights as in plot (a)
# 2. calculate vertex betweenness with cutoff 6 (b)
# 3. calculate vertex betweenness with cutoff 9 (c)
# 4. calculate vertex betweenness with cutoff Inf (d)
&lt;/code&gt;&lt;/pre&gt;

&lt;h3 id=&#34;normalization&#34;&gt;normalization&lt;/h3&gt;

&lt;p&gt;stackoverflow:
&lt;a href=&#34;http://stackoverflow.com/questions/34000326/normalized-local-closeness-centrality-in-r-igraphestimate-closeness&#34;&gt;http://stackoverflow.com/questions/34000326/normalized-local-closeness-centrality-in-r-igraphestimate-closeness&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;I am trying to calculate a normalized local closeness centrality. But
setting the parameter &amp;lsquo;normalized = T&amp;rsquo; for igraph::estimate_closeness()
only multiplies the results with (N-1). Is it possible to define N_i
for vertex i individually by it&amp;rsquo;s neighborhood, that is determined by
the cut-off parameter (and of course by the graph itself)?&lt;/p&gt;

&lt;p&gt;The mini example demonstrates, that setting the parameter &amp;lsquo;normalized=T&amp;rsquo;
does not work for this purpose. It normalizes with one and the same
(N-1) for all vertices.&lt;/p&gt;

&lt;p&gt;Thanks for help.&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;set.seed(1210)
require(&#39;igraph&#39;)
g &amp;lt;- random.graph.game(20,3/10)
g &amp;lt;- set.edge.attribute(g, &amp;quot;weight&amp;quot;,  value= 1)
cnt &amp;lt;- estimate_closeness(g, cutoff = 3, normalized = T );cnt

##  [1] 0.6333333 0.2467532 0.3220339 0.5000000 0.5277778 0.5588235 0.6129032
##  [8] 0.5135135 0.5428571 0.5588235 0.4634146 0.5277778 0.6129032 0.5000000
## [15] 0.6333333 0.5000000 0.3392857 0.5135135 0.5135135 0.5135135

cnf &amp;lt;- estimate_closeness(g, cutoff = 3, normalized = F );cnf

##  [1] 0.03333333 0.01298701 0.01694915 0.02631579 0.02777778 0.02941176
##  [7] 0.03225806 0.02702703 0.02857143 0.02941176 0.02439024 0.02777778
## [13] 0.03225806 0.02631579 0.03333333 0.02631579 0.01785714 0.02702703
## [19] 0.02702703 0.02702703

#print results
cbind(cnf * (length(V(g))-1), cnt)

##                       cnt
##  [1,] 0.6333333 0.6333333
##  [2,] 0.2467532 0.2467532
##  [3,] 0.3220339 0.3220339
##  [4,] 0.5000000 0.5000000
##  [5,] 0.5277778 0.5277778
##  [6,] 0.5588235 0.5588235
##  [7,] 0.6129032 0.6129032
##  [8,] 0.5135135 0.5135135
##  [9,] 0.5428571 0.5428571
## [10,] 0.5588235 0.5588235
## [11,] 0.4634146 0.4634146
## [12,] 0.5277778 0.5277778
## [13,] 0.6129032 0.6129032
## [14,] 0.5000000 0.5000000
## [15,] 0.6333333 0.6333333
## [16,] 0.5000000 0.5000000
## [17,] 0.3392857 0.3392857
## [18,] 0.5135135 0.5135135
## [19,] 0.5135135 0.5135135
## [20,] 0.5135135 0.5135135

sum(abs(cnf * (length(V(g))-1) - cnt))

## [1] 8.326673e-17

#for visualization
V(g)$name &amp;lt;- paste(&amp;quot;v&amp;quot;, 1:length(V(g)), sep=&amp;quot;&amp;quot;)#letters[1:length(V(g))]
set.seed(2)
lay &amp;lt;- layout.auto(g)

## set plotting parameters
vs &amp;lt;- 15# vertex.size
ec &amp;lt;- gray(0.8)#edge.color
elx &amp;lt;- 2# edge.label.cex
elc &amp;lt;- &amp;quot;black&amp;quot;#,edge.label.color
vlc &amp;lt;- 2#vertex.label.cex
ew &amp;lt;- 2#edge.width
hd &amp;lt;- paste(rep(&amp;quot; &amp;quot;,0), collapse=&amp;quot;&amp;quot;)
cm &amp;lt;- 3
vc &amp;lt;- &amp;quot;orange&amp;quot;#gray(0.8)#palette(&amp;quot;default&amp;quot;)#&amp;quot;grey&amp;quot;#vertex.color

windows(width = 18, height=6)
#par(mfrow=c(1,3))

plot(g, edge.label=paste(hd, round(E(g)$weight,1)), layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(a) Weights&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-7-1.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;V(g)$name &amp;lt;- paste(round(cnt,3))
plot(g,          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(b) local closeness normalized&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-7-2.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;pre&gt;&lt;code&gt;V(g)$name &amp;lt;- paste(round(cnf,3))
plot(g,          layout=lay, vertex.size=vs, edge.color=ec, edge.label.cex=elx, vertex.label.cex=vlc, edge.width=ew, edge.label.color=elc, vertex.color=vc)
title(&amp;quot;(c) local closeness&amp;quot;, cex.main = cm)
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/unnamed-chunk-7-3.png&#34; alt=&#34;&#34; /&gt;&lt;/p&gt;

&lt;h4 id=&#34;my-solution&#34;&gt;My solution&lt;/h4&gt;

&lt;p&gt;easy workaround for local centralities: define neighborhood for each
node, calcualte &amp;lsquo;global&amp;rsquo; centralities for these neighborhoods and
normalize it individually.&lt;/p&gt;

&lt;h1 id=&#34;what-for-did-we-use-centralities&#34;&gt;What for did we use centralities?&lt;/h1&gt;

&lt;p&gt;Thanks to Anita Graser for the beautiful plots!&lt;/p&gt;

&lt;p&gt;$$\hat{y}_{t,\gamma}(s,b,c) = \beta_t  + \beta_{\gamma,s} \cdot s  + \beta_{\gamma,b} \cdot b + \beta_{\gamma,c} \cdot c + \beta_{\gamma,bc} \cdot b \cdot c$$&lt;/p&gt;

&lt;p&gt;some plots:&lt;/p&gt;

&lt;h2 id=&#34;global-betweenness&#34;&gt;Global Betweenness&lt;/h2&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic1.png&#34; title=&#34;**Global Betweenness** \
*Global Betweenness*&#34; alt=&#34;**Global Betweenness** \
*Global Betweenness*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;h2 id=&#34;local-betweenness&#34;&gt;Local Betweenness&lt;/h2&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic2.png&#34; title=&#34;**Local Betweenness** \
*Local Betweenness*&#34; alt=&#34;**Local Betweenness** \
*Local Betweenness*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;h2 id=&#34;global-closeness&#34;&gt;Global Closeness&lt;/h2&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic3.png&#34; title=&#34;**Global Closeness** \
*Global Closeness*&#34; alt=&#34;**Global Closeness** \
*Global Closeness*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;h2 id=&#34;local-closeness&#34;&gt;Local Closeness&lt;/h2&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic4.png&#34; title=&#34;**Local Closeness** \
*Local Closeness*&#34; alt=&#34;**Local Closeness** \
*Local Closeness*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;h2 id=&#34;real-world-difference-of-node-based-and-link-based-betweenness-for-a-link&#34;&gt;Real world difference of node-based and link-based Betweenness for a link&lt;/h2&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic5.png&#34; title=&#34;**Real world difference 1** \
*Real world difference 1*&#34; alt=&#34;**Real world difference 1** \
*Real world difference 1*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic6.png&#34; title=&#34;**Real world difference 2** \
*Real world difference 2*&#34; alt=&#34;**Real world difference 2** \
*Real world difference 2*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;h2 id=&#34;improvement-travel-time-model&#34;&gt;Improvement travel time model&lt;/h2&gt;

&lt;h3 id=&#34;mape-difference-global-minus-local&#34;&gt;MAPE Difference Global minus local&lt;/h3&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic7.png&#34; title=&#34;**MAPE Difference** \
*MAPE Difference*&#34; alt=&#34;**MAPE Difference** \
*MAPE Difference*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;h3 id=&#34;detail-mape-base&#34;&gt;Detail MAPE base&lt;/h3&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic8.png&#34; title=&#34;**MAPE base** \
*MAPE base*&#34; alt=&#34;**MAPE base** \
*MAPE base*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;h3 id=&#34;detail-mape-with-local-centralities&#34;&gt;Detail MAPE with local centralities&lt;/h3&gt;

&lt;p&gt;&lt;img src=&#34;/img/20160524_igraph_applications_files/figure-markdown_strict/pic9.png&#34; title=&#34;**MAPE local** \
*MAPE local*&#34; alt=&#34;**MAPE local** \ *MAPE local*&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 06-2016 Minutes</title>
          <link>/blog/2016/06/28/meetup-06-2016-minutes/</link>
          <pubDate>Tue, 28 Jun 2016 13:00:00 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/06/28/meetup-06-2016-minutes/</guid>
          <description>&lt;p&gt;&lt;img src=&#34;/img/sun.png&#34; alt=&#34;SummerOfCode&#34; /&gt;&lt;/p&gt;

&lt;p&gt;The last meetup before the summer break concluded the igraph session with a talk by Rainer. Mario summed up his impressions from the R/Finance 2016 conference.&lt;/p&gt;

&lt;h2 id=&#34;rainer-stuetz-igraph-applications&#34;&gt;Rainer Stuetz: igraph Applications&lt;/h2&gt;

&lt;p&gt;Rainer presented interesting &lt;a href=&#34;https://cran.r-project.org/web/packages/igraph/index.html&#34;&gt;igraph&lt;/a&gt; applications with an example to extract (clean) subnetworks, e.g. motorways, from the Austrian &lt;a href=&#34;http://gip.gv.at&#34;&gt;graph integration platform&lt;/a&gt;. The routing graph was parsed using the &lt;a href=&#34;https://cran.r-project.org/web/packages/readr/index.html&#34;&gt;readr&lt;/a&gt; package and pre-processed with &lt;a href=&#34;https://cran.r-project.org/web/packages/dplyr/index.html&#34;&gt;dplyr&lt;/a&gt;. For the extraction of the subnetworks, shortest path algorithms from igraph were used.&lt;/p&gt;

&lt;h2 id=&#34;mario-annau-r-finance-2016-impressions&#34;&gt;Mario Annau: R/Finance 2016 Impressions&lt;/h2&gt;

&lt;p&gt;Mario summed up his impressions on the last &lt;a href=&#34;http://www.rinfinance.com/agenda&#34;&gt;R/Finance conference&lt;/a&gt; including a small &lt;em&gt;mashup&lt;/em&gt; &lt;a href=&#34;/slides/rfinance2016_mashup.pdf&#34;&gt;presentation&lt;/a&gt;.&lt;/p&gt;

&lt;h2 id=&#34;organizational&#34;&gt;Organizational&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Next Meetup: September 27th, check &lt;a href=&#34;http://www.meetup.com/ViennaR&#34;&gt;meetup page&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;Ideas for projects / talks:

&lt;ul&gt;
&lt;li&gt;Finance related, forecasting&lt;/li&gt;
&lt;li&gt;Fraud detection, e.g. Panama papers (&lt;a href=&#34;https://github.com/dgrtwo/rpanama&#34;&gt;https://github.com/dgrtwo/rpanama&lt;/a&gt;)&lt;/li&gt;
&lt;li&gt;Company database&lt;/li&gt;
&lt;li&gt;Visualization tutorial covering &lt;a href=&#34;https://cran.r-project.org/web/packages/ggplot2/index.html&#34;&gt;ggplot2&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;For further comments and suggestions for projects and talks just drop a mail at &lt;a href=&#34;mailto:host@viennar.org&#34;&gt;host@viennar.org&lt;/a&gt;!
&lt;br /&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&#34;announcements&#34;&gt;Announcements&lt;/h2&gt;

&lt;p&gt;The first &lt;a href=&#34;http://budapest.satrdays.org&#34;&gt;satRday&lt;/a&gt; conference is taking place in Budapest on September 3rd 2016. Thanks to sponsor support (e.g. &lt;a href=&#34;R Consortium&#34;&gt;https://www.r-consortium.org/&lt;/a&gt;) the conference is quite cheap (10-30€).&lt;/p&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 05-2016 Minutes</title>
          <link>/blog/2016/05/24/meetup-05-2016-minutes/</link>
          <pubDate>Tue, 24 May 2016 13:00:00 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/05/24/meetup-05-2016-minutes/</guid>
          <description>&lt;p&gt;&lt;img src=&#34;/img/igraph.jpeg&#34; alt=&#34;IGraph&#34; /&gt;&lt;/p&gt;

&lt;p&gt;This meetup extensively covered the &lt;strong&gt;igraph&lt;/strong&gt; package with talks by Florian Sobieczky and Max Leodolter.&lt;/p&gt;

&lt;h2 id=&#34;florian-sobieczky-igraph-visualising-random-walks-on-graphs&#34;&gt;Florian Sobieczky: igraph - Visualising Random Walks on Graphs.&lt;/h2&gt;

&lt;p&gt;The igraph package for examples from the theory of randomly generated graphs. Questions from percolation theory and random walks such as connectivity, isoperimetry, amenability and phase transitions have been discussed in the framework of the routines from igraph. Examples from social network theory, traffic networks, and production environments have been presented.&lt;/p&gt;

&lt;p&gt;You can check out the code &lt;a href=&#34;/code/igraph_randomwalks.R&#34;&gt;here&lt;/a&gt;.&lt;/p&gt;

&lt;h2 id=&#34;max-leodolter-igraph-applications-for-transportation-networks&#34;&gt;Max Leodolter: igraph Applications for Transportation Networks&lt;/h2&gt;

&lt;p&gt;Max gave an introduction to the &lt;strong&gt;igraph&lt;/strong&gt; package including real-world applications (slides follow asap).&lt;/p&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 04-2016 Minutes</title>
          <link>/blog/2016/05/06/meetup-04-2016-minutes/</link>
          <pubDate>Fri, 06 May 2016 13:00:00 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/05/06/meetup-04-2016-minutes/</guid>
          <description>&lt;p&gt;&lt;img src=&#34;/img/wordcloudtm.png&#34; alt=&#34;Wordcloud&#34; /&gt;&lt;/p&gt;

&lt;p&gt;This meetup included an extensive Text Mining in R session with an &lt;em&gt;Introduction to&lt;/em&gt; &lt;strong&gt;tm&lt;/strong&gt;  by Ingo Feinerer and a talk about &lt;em&gt;Text Mining with Hadoop&lt;/em&gt; by Stefan Theussl.&lt;/p&gt;

&lt;p&gt;After a creative break for the last month Ingo and Stefan gave great talks covering &lt;strong&gt;tm&lt;/strong&gt; in greater detail after the brief introduction in February.&lt;/p&gt;

&lt;h2 id=&#34;ingo-feinerer-introduction-to-tm&#34;&gt;Ingo Feinerer: Introduction to &lt;strong&gt;tm&lt;/strong&gt;&lt;/h2&gt;

&lt;p&gt;Ingo started right away with a nice bottom-up introduction covering &lt;strong&gt;tm&lt;/strong&gt;&amp;rsquo;s building blocks like Sources, Readers and Corpora. The creation of Document-TermMatrices was also motivated with a small clustering example for 3 documents.&lt;/p&gt;

&lt;p&gt;CRAN Package &lt;a href=&#34;https://cran.r-project.org/web/packages/tm/index.html&#34;&gt;Link&lt;/a&gt;
Package &lt;a href=&#34;https://cran.r-project.org/web/packages/tm/vignettes/tm.pdf&#34;&gt;Vignette&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;The word cloud shown above was created from the &lt;strong&gt;tm&lt;/strong&gt; package vignette as follows:&lt;/p&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;library(tm)
library(wordcloud)
uri &amp;lt;- sprintf(&amp;quot;file://%s&amp;quot;, system.file(file.path(&amp;quot;doc&amp;quot;, &amp;quot;tm.pdf&amp;quot;), package = &amp;quot;tm&amp;quot;))
stopifnot(all(file.exists(Sys.which(c(&amp;quot;pdfinfo&amp;quot;, &amp;quot;pdftotext&amp;quot;)))))
corp &amp;lt;- Corpus(URISource(uri), readerControl = list(reader = readPDF))  
tmvignette &amp;lt;- paste(content(corp[[1]]), collapse = &amp;quot;\n&amp;quot;)
vigclean &amp;lt;- stripWhitespace(removePunctuation(removeNumbers(tmvignette)))
vigclean &amp;lt;- removeWords(vigclean, stopwords())
wordcloud(vigclean)
&lt;/code&gt;&lt;/pre&gt;

&lt;h2 id=&#34;stefan-theussl-text-mining-with-hadoop&#34;&gt;Stefan Theussl: Text Mining with Hadoop&lt;/h2&gt;

&lt;p&gt;Stefan gave a solution to the problem when things (i.e. text corpora) get big using a set of Hadoop R-packages he created in collaboration with Ingo.&lt;/p&gt;

&lt;p&gt;CRAN packages:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;https://cran.r-project.org/web/packages/DSL/index.html&#34;&gt;&lt;strong&gt;DSL&lt;/strong&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://cran.r-project.org/web/packages/tm.plugin.dc/index.html&#34;&gt;&lt;strong&gt;tm.plugin.dc&lt;/strong&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://cran.r-project.org/web/packages/hive/index.html&#34;&gt;&lt;strong&gt;hive&lt;/strong&gt;&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Documentation:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;https://cran.r-project.org/web/packages/DSL/vignettes/DSL.pdf&#34;&gt;&lt;strong&gt;DSL&lt;/strong&gt;&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;http://www.jstatsoft.org/v51/i05&#34;&gt;A tm Plug-In for Distributed Text Mining in R&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
    
      
        <item>
          <title>Wordcloud Wienerwald</title>
          <link>/blog/2016/02/24/wordcloud-wienerwald/</link>
          <pubDate>Wed, 24 Feb 2016 22:20:00 &#43;0200</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/02/24/wordcloud-wienerwald/</guid>
          <description>&lt;p&gt;&lt;img src=&#34;/img/wordcloudwienerwald.png&#34; title=&#34;**Wordcloud** \
*Ödön von Horvath - Geschichten aus dem Wienerwald*&#34; alt=&#34;**Wordcloud** \
*Ödön von Horvath - Geschichten aus dem Wienerwald*&#34; style=&#39;width:70%;&#39; border=&#34;0&#34; /&gt;&lt;/p&gt;

&lt;p&gt;After a very nice hands-on introduction in yesterday&amp;rsquo;s Vienna R meetup meeting from Mario Annau, I created an example of textmining ending with a wordcloud. As the blog is called ViennaR, I chose to use a play strongly related to Vienna - Ödön von Horváth - Geschichten aus dem Wienerwald.&lt;/p&gt;

&lt;p&gt;The code documentation is in German, but with a little R-experience it should be easily understood.
In a first step all the required libraries are loaded.&lt;/p&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;library(knitr) # Erzeugen eines HTML-Dokument
library(rvest) # Einlesen von HTML-Dokumenten
library(tm) # Erzeugen und Manipulieren von Textcorpora 
library(stringi) # Umwandeln von schlecht kodierten Sonderzeichen 
library(SnowballC) # Wortstammoperationen 
library(wordcloud) # Erzeugen der Wordcloud
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;As of next, the textparts are loaded from the project Gutenberg homepage. As the play was divided this had to be repeated four times. A function was defined for the import from each url with the appropriate CSS-node and ideal encoding. However, the forced transformation to UTF-8 did not do anything actually.&lt;/p&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;# Ödön von Horváth - Geschichten aus dem Wienerwald 
url1 &amp;lt;- &amp;quot;http://gutenberg.spiegel.de/buch/geschichten-aus-dem-wiener-wald-volksstuck-in-drei-teilen-2900/1&amp;quot;
url2 &amp;lt;- &amp;quot;http://gutenberg.spiegel.de/buch/geschichten-aus-dem-wiener-wald-volksstuck-in-drei-teilen-2900/2&amp;quot;
url3 &amp;lt;- &amp;quot;http://gutenberg.spiegel.de/buch/geschichten-aus-dem-wiener-wald-volksstuck-in-drei-teilen-2900/3&amp;quot;
url4 &amp;lt;- &amp;quot;http://gutenberg.spiegel.de/buch/geschichten-aus-dem-wiener-wald-volksstuck-in-drei-teilen-2900/4&amp;quot;

# Einlesefunktion des Textes; Kodierung zu UTF-8
  gut_les &amp;lt;-   function(x){
    step1 &amp;lt;- read_html(x, encoding = &amp;quot;ISO-8859-1&amp;quot;)
    step2 &amp;lt;- html_nodes(step1,&amp;quot;#gutenb&amp;quot;)
    step3 &amp;lt;- iconv(html_text(step2),from = &amp;quot;ISO-8859-1&amp;quot;, to = &amp;quot;UTF-8&amp;quot;)
    return(step3)}
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;Next the four text parts were integrated into one Corpus, a data type used in the tm package. In this case, the corpus was created from four char-vectors. Afterwards steps to clean up the text were performed. As in the presentation from Mario discussed, the order of these operations should be considered thoroughly.
First, I adapted the wrongly encoded signs with the function stri_replace_all_fixed(). German Umlaute are a real pain, I really can say that, as I have one in my surname. Then the names of the figures had to be removed, otherwise they would have overwhelmed the output.&lt;/p&gt;

&lt;p&gt;Typical steps for allowing for meaningful text operation are alse the removal of Whitespace, unifying to lower cases, removing punctuations, stopwords (&amp;rdquo;&lt;em&gt;meaningless&lt;/em&gt;&amp;rdquo; words as conjunctions) and word endings. All four texts are still in the corpus, which can be adressed like list items. They were though recoded as PlainTextDocument, which was a necessary step for being used in the wordcloud() function.&lt;/p&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;# Verbinden der 4 Texte zu einem Corpus 
GWW_corp &amp;lt;- Corpus(VectorSource(c(gut_les(url1),gut_les(url2),gut_les(url3),gut_les(url4))))

# Kodierungsfehler 
sz_fehler &amp;lt;- c(&amp;quot;ü&amp;quot;,&amp;quot;ö&amp;quot;,&amp;quot;ä&amp;quot;,&amp;quot;Ã\u009f&amp;quot;,&amp;quot;â\u0080\u0093&amp;quot;,&amp;quot;Â&amp;quot;)
sz_korrekt &amp;lt;- c(&amp;quot;ü&amp;quot;,&amp;quot;ö&amp;quot;,&amp;quot;ä&amp;quot;,&amp;quot;ß&amp;quot;,&amp;quot;–&amp;quot;,&amp;quot;&amp;quot;)

# Ersetzen der Kodierungsfehler; deshalb das Stringi Package 
gwwc0 &amp;lt;- tm_map(GWW_corp, function(x) stri_replace_all_fixed(x, sz_fehler, sz_korrekt, vectorize_all = FALSE))

# Die Personen werden bei einem Theaterstrück sinnvollerweise entfernt. 
gwwc1 &amp;lt;- tm_map(gwwc0, removeWords, 
        c(&amp;quot;Alfred&amp;quot;,&amp;quot;Die Mutter&amp;quot;,&amp;quot;Die Großmutter&amp;quot;,&amp;quot;Der Hierlinger Ferdinand&amp;quot;,
          &amp;quot;Valerie&amp;quot;,&amp;quot;Oskar&amp;quot;,&amp;quot;Ida&amp;quot;,&amp;quot;Havlitschek&amp;quot;,&amp;quot;Rittmeister&amp;quot;,&amp;quot;Eine gnädige Frau&amp;quot;,
          &amp;quot;Marianne&amp;quot;,&amp;quot;Zauberkönig&amp;quot;,&amp;quot;Zwei Tanten&amp;quot;,&amp;quot;Erich&amp;quot;,&amp;quot;Emma&amp;quot;,&amp;quot;Helene&amp;quot;,
          &amp;quot;Der Dienstbot&amp;quot;,&amp;quot;Baronin&amp;quot;,&amp;quot;Beichtvater&amp;quot;, &amp;quot;Der Mister&amp;quot;,&amp;quot;Der Conferencier&amp;quot;))
gwwc2 &amp;lt;- tm_map(gwwc1, stripWhitespace) # Entfernen von Leerzeichen
gwwc3 &amp;lt;- tm_map(gwwc2, tolower) # Kleinschrift
gwwc4 &amp;lt;- tm_map(gwwc3, PlainTextDocument) # Umcodierung der Char-Vektoren zu Textdokumenten (für Wordlcloud)
gwwc5 &amp;lt;- tm_map(gwwc4, removeWords, stopwords(&amp;quot;german&amp;quot;)) # Entfernen von Füllwörtern
gwwc6 &amp;lt;- tm_map(gwwc5, removePunctuation) # Entfernen von Sonderzeichen 
gwwc7 &amp;lt;- tm_map(gwwc6, stemDocument) # Entfernen der Endungen 
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;After all this cleanup the wordcloud can be drawn. I used a variety of twelve rather light colors with a black background. It is not possible to access the background color inside the wordcloud function, but with the par-options.&lt;/p&gt;

&lt;p&gt;Et voilà - l&amp;rsquo;illustration!&lt;/p&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;farbs &amp;lt;- brewer.pal(12,&amp;quot;Set3&amp;quot;) # Farbpalette für die Wordcloud

par(mar=c(0,0,0,0),bg=&amp;quot;black&amp;quot;) # Ausfüllen des Plotfensters, Hintergrund schwarz
# Die eigentliche Wordcloud
wordcloud(gwwc6, max.words=180, 
          random.order=FALSE,random.color=FALSE, # Farben und Reihenfolge nach Anzahl der Wörter geordnet 
          rot.per=0.35, # 35 % der Wörter sind senkrecht  
          colors=(farbs))
&lt;/code&gt;&lt;/pre&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 02-2016 Minutes</title>
          <link>/blog/2016/02/22/meetup-02-2016-minutes/</link>
          <pubDate>Mon, 22 Feb 2016 13:00:00 &#43;0200</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/02/22/meetup-02-2016-minutes/</guid>
          <description>&lt;p&gt;This meeting included an introductory tutorial by Mario covering &lt;strong&gt;tm&lt;/strong&gt; and sentiment anaylsis. The slides are available &lt;a href=&#34;/slides/tm101_slides.pdf&#34;&gt;here&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;The &lt;a href=&#34;/slides/tm101_slides.pdf&#34;&gt;tutorial&lt;/a&gt; was structured as follows:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Data Reading&lt;/li&gt;
&lt;li&gt;Data Structures&lt;/li&gt;
&lt;li&gt;Preprocessing Pipeline&lt;/li&gt;
&lt;li&gt;removePunctuation, tolower, removeWords, stripWhitespace, stemDocument&lt;/li&gt;
&lt;li&gt;Examples&lt;/li&gt;
&lt;li&gt;Known Weaknesses and Outlook&lt;/li&gt;
&lt;li&gt;Plans for &lt;strong&gt;SentimentAnalysis&lt;/strong&gt; (tm.plugin.sentiment 2.0)&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Further talks covering advanced text mining topics in R are planned for the next meeting(s), so stay tuned&amp;hellip;&lt;/p&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 01-2016 Minutes</title>
          <link>/blog/2016/02/03/meetup-01-2016-minutes/</link>
          <pubDate>Wed, 03 Feb 2016 13:00:00 &#43;0200</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2016/02/03/meetup-01-2016-minutes/</guid>
          <description>&lt;p&gt;The goal of this tutorial was to demonstrate how to use Apache &lt;a href=&#34;https://spark.apache.org/docs/1.5.2/sparkr.html&#34;&gt;SparkR&lt;/a&gt; for analyzing large-scale datasets in R. For demo purposes data extracted from the Bitcoin blockchain were used to produce a time series plot similar to &lt;a href=&#34;/img/blockchain_transactions.png&#34;&gt;this&lt;/a&gt; one.&lt;/p&gt;

&lt;p&gt;The talk was structured as follows:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Intro + Architecture Spark [Roland]&lt;/li&gt;
&lt;li&gt;Intro Bitcoin Use Case [Bernhard]&lt;/li&gt;
&lt;li&gt;Demo Standard R (+ some extra packages) [Bernhard]&lt;/li&gt;
&lt;li&gt;Demo SparkR [Bernhard]&lt;/li&gt;
&lt;li&gt;Demo SparkR - Cluster [Roland]&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Materials are available at &lt;a href=&#34;https://github.com/behas/sparkR-tutorial&#34;&gt;https://github.com/behas/sparkR-tutorial&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 11-2015 Minutes</title>
          <link>/blog/2015/11/25/meetup-11-2015-minutes/</link>
          <pubDate>Wed, 25 Nov 2015 21:00:00 &#43;0200</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2015/11/25/meetup-11-2015-minutes/</guid>
          <description>&lt;p&gt;During the last meetup a few interesting ideas/discussions came up like interactive visualizations of &lt;a href=&#34;https://github.com/ViennaR/OpenGovernmentVienna&#34;&gt;Open Government Data&lt;/a&gt;, Geocoding APIs, derstandard.at datasets and Hadoop vs Spark. During the second part of the meetup Mateuz held a great talk covering R-package creation, Github and CI.&lt;/p&gt;

&lt;h2 id=&#34;projects&#34;&gt;Projects&lt;/h2&gt;

&lt;p&gt;Even without my full memory capacity I&amp;rsquo;ll try to sum them up by project ;):&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;https://github.com/ViennaR/OpenGovernmentVienna&#34;&gt;Open Government Data Vienna&lt;/a&gt;

&lt;ul&gt;
&lt;li&gt;With the nice visualizations from Christian in mind (&lt;a href=&#34;/blog/2015/10/30/population-densities-in-vienna&#34;&gt;Population Densities&lt;/a&gt; and &lt;a href=&#34;/blog/2015/11/08/bike-parking-lots-in-vienna&#34;&gt;Bike Parking Lots&lt;/a&gt;) it would be nice to make up a list of features to be plotted from &lt;a href=&#34;https://www.data.gv.at&#34;&gt;https://www.data.gv.at&lt;/a&gt; and make this visualization interactive (e.g. using &lt;a href=&#34;http://shiny.rstudio.com/&#34;&gt;Shiny&lt;/a&gt;).&lt;/li&gt;
&lt;li&gt;Mateuz also introduced us to the topic of Geocoding and its numerous APIs (e.g. from &lt;a href=&#34;https://developers.google.com/maps/documentation/geocoding/intro&#34;&gt;Google&lt;/a&gt;) and suggested to write a package for it.&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://github.com/ViennaR/TextMiningVienna&#34;&gt;Text Mining Vienna&lt;/a&gt;

&lt;ul&gt;
&lt;li&gt;After the discussion about puclic newspapers in Vienna Bernhard already provided a small &lt;a href=&#34;https://github.com/ViennaR/TextMiningVienna/tree/master/NewspapersAustria&#34;&gt;dataset&lt;/a&gt; to play around with. We&amp;rsquo;ll get our hands dirty and provide ideas (topic modelling, mood index, popular words, etc.) next time!&lt;/li&gt;
&lt;li&gt;Bernhard &amp;amp; Roland mentioned interest to talk about Hadoop -&amp;gt; Spark -&amp;gt; SparkR. How to set things up, integrate R, etc.
Sounds like a great topic next years first meetup.&lt;/li&gt;
&lt;li&gt;Hillary Clintons E-mails: Some ideas are &lt;a href=&#34;https://github.com/ViennaR/TextMiningVienna/tree/master/HillaryClinton&#34;&gt;there&lt;/a&gt; - focus will be on the newspapers project for now.&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://en.wikipedia.org/wiki/Fantasy_sport&#34;&gt;Fantasy Sports&lt;/a&gt;, looks abandoned by Daniel?&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&#34;talk&#34;&gt;Talk&lt;/h2&gt;

&lt;p&gt;Mateusz covered all important topics about &lt;em&gt;modern&lt;/em&gt; R package creation including the following points:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Setting up Github repo for &lt;a href=&#34;https://github.com/ViennaR/startupPackage&#34;&gt;startup package&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;Create R package skeleton&lt;/li&gt;
&lt;li&gt;Include test cases&lt;/li&gt;
&lt;li&gt;Include continous integration&lt;/li&gt;
&lt;li&gt;Include code coverage&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Materials are available at &lt;a href=&#34;http://zozlak.org/Setting_up_a_first_R_project.html&#34;&gt;http://zozlak.org/Setting_up_a_first_R_project.html&lt;/a&gt;.&lt;/p&gt;

&lt;h2 id=&#34;organisational&#34;&gt;Organisational&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;If you have any questions, ideas and want to share just drop a message on &lt;a href=&#34;http://www.meetup.com/ViennaR&#34;&gt;http://www.meetup.com/ViennaR&lt;/a&gt; or &lt;a href=&#34;mailto:host@viennar.com&#34;&gt;host@viennar.org&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;Also just let us know if you need access to a github repo at &lt;a href=&#34;https://github.com/ViennaR&#34;&gt;https://github.com/ViennaR&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;The next R-meetup will take place directly at the &lt;a href=&#34;http://www.artadvent.at&#34;&gt;Christkindlmarkt Karlsplatz&lt;/a&gt; on 22nd December, 7pm. Meeting point will be in front of &lt;del&gt;Karlskirche&lt;/del&gt; &lt;em&gt;Goldenes Lamm&lt;/em&gt;.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;See you next time for a punchR!&lt;/p&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 11-2015</title>
          <link>/blog/2015/11/22/meetup-11-2015/</link>
          <pubDate>Sun, 22 Nov 2015 21:00:00 &#43;0200</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2015/11/22/meetup-11-2015/</guid>
          <description>&lt;p&gt;The next R-meetup is around the corner on Tuesday 24th (currently 6 spots left) at &lt;em&gt;Goldenes Lamm&lt;/em&gt; with a presentation of a few project ideas and first results. Finally we will have a nice talk about setting up R-packages on Github by Mateusz Zoltak.&lt;/p&gt;

&lt;h2 id=&#34;projects&#34;&gt;Projects&lt;/h2&gt;

&lt;p&gt;The following project streams have been selected and coordinators assigned:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;https://github.com/ViennaR/OpenGovernmentVienna&#34;&gt;Open Government Data Vienna&lt;/a&gt;

&lt;ul&gt;
&lt;li&gt;Show &lt;a href=&#34;/blog/2015/10/30/population-densities-in-vienna&#34;&gt;Population Densities&lt;/a&gt; and &lt;a href=&#34;/blog/2015/11/08/bike-parking-lots-in-vienna&#34;&gt;Bike Parking Lots&lt;/a&gt; in Vienna with nice visualisations (Christian).&lt;/li&gt;
&lt;li&gt;Parliamentary voting results (input from Mateusz)

&lt;ul&gt;
&lt;li&gt;Harvesting data, e.g. &lt;a href=&#34;https://www.data.gv.at/katalog/dataset/1a3c44a8-7361-4dc7-ac0d-6ec19ddba191&#34;&gt;Beschlüsse des Nationalrats&lt;/a&gt; from &lt;a href=&#34;https://www.data.gv.at&#34;&gt;https://www.data.gv.at&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;Visualization&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://github.com/ViennaR/TextMiningVienna&#34;&gt;Text Mining Vienna&lt;/a&gt;

&lt;ul&gt;
&lt;li&gt;Text mining (Mario)

&lt;ul&gt;
&lt;li&gt;Public newspapers&lt;/li&gt;
&lt;li&gt;Most popular words in Newspapers&lt;/li&gt;
&lt;li&gt;HC Strache Posts on Facebook&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;Use Kaggle as an Inspiration (Walter, Mario)

&lt;ul&gt;
&lt;li&gt;Hillary Clintons E-mails&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;https://en.wikipedia.org/wiki/Fantasy_sport&#34;&gt;Fantasy Sports&lt;/a&gt;, e.g. from Yahoo (short intro/write&amp;ndash;up by Daniel)

&lt;ul&gt;
&lt;li&gt;Put team together based on statistics, e.g. Baseball&lt;/li&gt;
&lt;li&gt;References:

&lt;ul&gt;
&lt;li&gt;Book: &lt;a href=&#34;http://www.amazon.com/Analyzing-Baseball-Data-Chapman-Series/dp/1466570229&#34;&gt;Analyzing Baseball data in R&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;http://fantasyfootballanalytics.net/2014/06/scraping-fantasy-football-projections.html&#34;&gt;Scraping Fantasy Football Projections from the Web&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&#34;presentation-r-package-setup-on-github&#34;&gt;Presentation R-Package Setup on Github&lt;/h2&gt;

&lt;p&gt;Mateusz will give us an interesting introduction on Git(hub) and how to create and manage R-packages.
The talk will cover the following points:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;How to set up a git repository.&lt;/li&gt;
&lt;li&gt;How create a basic R-package (or &lt;code&gt;package.skeleton&lt;/code&gt;).&lt;/li&gt;
&lt;li&gt;Set up continous integration using platforms like &lt;a href=&#34;https://travis-ci.org/&#34;&gt;Travis&lt;/a&gt;, &lt;a href=&#34;http://www.appveyor.com/&#34;&gt;Appveyor&lt;/a&gt; or &lt;a href=&#34;http://wercker.com/&#34;&gt;Wercker&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;Set up code coverage using &lt;a href=&#34;https://codecov.io&#34;&gt;Codecov&lt;/a&gt;, &lt;a href=&#34;https://coveralls.io&#34;&gt;Coveralls&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;Explain all those colourful badges ;)&lt;/li&gt;
&lt;/ol&gt;

&lt;h2 id=&#34;organisational&#34;&gt;Organisational&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Determine next dates (with Matthias)&lt;/li&gt;
&lt;li&gt;Manage user access, contributions to Blog&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Like in previous Meetups we are happy to have some presentations of &lt;em&gt;ViennaR&lt;/em&gt; participants. Just drop us a mail that you are planning a presentation during one of the next meetups.
Any suggestions like fixed discussion topics, e.g. book presentations, new R packages around, etc. are highly welcome.
Just let us know!&lt;/p&gt;

&lt;p&gt;Best,&lt;/p&gt;

&lt;p&gt;-ViennaR&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Bike Parking Lots in Vienna</title>
          <link>/blog/2015/11/08/bike-parking-lots-in-vienna/</link>
          <pubDate>Sun, 08 Nov 2015 22:20:00 &#43;0200</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2015/11/08/bike-parking-lots-in-vienna/</guid>
          <description>&lt;p&gt;Another addition to the  &lt;a href=&#34;https://github.com/ViennaR/OpenGovernmentVienna&#34;&gt;OpenGovernmentVienna&lt;/a&gt; repo by Christian Brandstaetter showing the density of bike parking lots in Vienna.&lt;/p&gt;

&lt;p&gt;The script executes the following steps:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Download Vienna map including district boundaries and bike parking lots from &lt;a href=&#34;http://data.wien.gv.at&#34;&gt;http://data.wien.gv.at&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;Plot two Vienna maps showing bike parking lots and density per district.&lt;/li&gt;
&lt;/ol&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;library(rgdal) 
library(rgeos) 
library(XML)
library(RCurl)
library(ggplot2) 
library(plotrix) 
library(maptools)

download.vienna.shape &amp;lt;- function(shapename, outdir = &amp;quot;data&amp;quot;) {
	baseurl &amp;lt;- &amp;quot;http://data.wien.gv.at/daten/geo?service=WFS&amp;amp;request=GetFeature&amp;amp;version=1.1.0&amp;amp;typeName=ogdwien:&amp;quot;
	urlparam &amp;lt;- &amp;quot;&amp;amp;srsName=EPSG:4326&amp;amp;outputFormat=shape-zip&amp;quot;	
	url &amp;lt;- sprintf(&amp;quot;%s%s%s&amp;quot;, baseurl, shapename, urlparam)
	
	dir.create(outdir, showWarnings = FALSE)
	destfile = file.path(outdir, sprintf(&amp;quot;%s.zip&amp;quot;, shapename))
	download.file(url, destfile = destfile)
	unzip(destfile, exdir = file.path(outdir, shapename))
	invisible(file.remove(destfile))
}
&lt;/code&gt;&lt;/pre&gt;

&lt;h1 id=&#34;retrieve-data&#34;&gt;Retrieve Data&lt;/h1&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;## Read District Boundaries
download.vienna.shape(&amp;quot;BEZIRKSGRENZEOGD&amp;quot;)
wmap &amp;lt;- readOGR(&amp;quot;data/BEZIRKSGRENZEOGD&amp;quot;, layer=&amp;quot;BEZIRKSGRENZEOGDPolygon&amp;quot;) 
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code&gt;## OGR data source with driver: ESRI Shapefile 
## Source: &amp;quot;data/BEZIRKSGRENZEOGD&amp;quot;, layer: &amp;quot;BEZIRKSGRENZEOGDPolygon&amp;quot;
## with 23 features
## It has 15 fields
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;## Read Streets
download.vienna.shape(&amp;quot;STRASSENGRAPHOGD&amp;quot;)
smap &amp;lt;- readOGR(&amp;quot;data/STRASSENGRAPHOGD&amp;quot;, layer=&amp;quot;STRASSENGRAPHOGD&amp;quot;) 
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code&gt;## OGR data source with driver: ESRI Shapefile 
## Source: &amp;quot;data/STRASSENGRAPHOGD&amp;quot;, layer: &amp;quot;STRASSENGRAPHOGD&amp;quot;
## with 28293 features
## It has 18 fields
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;## Read bike parking lots
download.vienna.shape(&amp;quot;FAHRRADABSTELLANLAGEOGD&amp;quot;)
bmap &amp;lt;- readOGR(&amp;quot;data/FAHRRADABSTELLANLAGEOGD&amp;quot;, layer=&amp;quot;FAHRRADABSTELLANLAGEOGDPoint&amp;quot;) 
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code&gt;## OGR data source with driver: ESRI Shapefile 
## Source: &amp;quot;data/FAHRRADABSTELLANLAGEOGD&amp;quot;, layer: &amp;quot;FAHRRADABSTELLANLAGEOGDPoint&amp;quot;
## with 4061 features
## It has 5 fields
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;## Download Size of Each district
distsize &amp;lt;- download.vienna.bydistrict(&amp;quot;lebensraum/tabellen/nutzungsklassen-bez&amp;quot;, skip.row = 2)
distsizekm2 &amp;lt;- distsize[, 1] / 100
&lt;/code&gt;&lt;/pre&gt;

&lt;h1 id=&#34;calculate-stuff&#34;&gt;Calculate stuff&lt;/h1&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;## Number of bike parking lots per district
bikelotsperdistrict &amp;lt;- table(bmap$BEZIRK)
lotdensity &amp;lt;- bikelotsperdistrict / distsizekm2

## Normalization of Colour Scaling
normdichte &amp;lt;- round(lotdensity/max(lotdensity)*50,digits=0) 
colfunc &amp;lt;- colorRampPalette(c(&amp;quot;lightblue&amp;quot;, &amp;quot;darkgreen&amp;quot;)) 
colfunc100 &amp;lt;- colfunc(100) 
bezirksfarben &amp;lt;- colfunc100[normdichte]

# Transformations  111.1 
# km to degree (einfache Umrechnung) 
hundm &amp;lt;- (1/111.1)/10 
# 100 m for Radius of Punkte  
centroids &amp;lt;- gCentroid(wmap, byid=TRUE) # Mittelpunkt/Bezirk
&lt;/code&gt;&lt;/pre&gt;

&lt;h1 id=&#34;plots&#34;&gt;Plots&lt;/h1&gt;

&lt;h2 id=&#34;plot-1-vienna-map-with-bike-parking-lots-discrete-density&#34;&gt;Plot 1 - Vienna Map with Bike Parking Lots (Discrete Density)&lt;/h2&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;layout(1:2, heights=c(5,1)) 
par(mar=c(0.5,0.5,1,0.5), oma=rep(3, 4), las=1) 
plot(wmap, main=&amp;quot;Bike Parking Lots in Vienna&amp;quot;, col=bezirksfarben[wmap$BEZNR]) 

# add bike parking lots
plot(smap, add=TRUE, col = &amp;quot;grey&amp;quot;)
draw.circle(coordinates(bmap)[,1], coordinates(bmap)[,2],hundm,border=rgb(255,255,0,maxColorValue=255),col=rgb(255,255,0,maxColorValue=255))
text(as.character(wmap$BEZ_RZ), x = centroids@coords[,1], y = centroids@coords[,2], col=&amp;quot;orangered&amp;quot;,cex=0.8,font=2)

# Legend
par(mar=c(1,0.5,3,0.5)) 
colseq &amp;lt;- seq(range(lotdensity)[1],range(lotdensity)[2],20) 
image(x=colseq,y=1,z=matrix(seq_along(colseq)), col=colfunc(10), main=expression(paste(&amp;quot;Bike parking lot density per district km&amp;quot;^-2)),axes=F) 
axis(1)  
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/bikedensity-plot1-1.png&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; alt=&#34;Null&#34;&gt;&lt;/p&gt;

&lt;h1 id=&#34;plot-2-vienna-map-with-bike-parking-lots-continous-density&#34;&gt;Plot 2 - Vienna Map with Bike Parking Lots (Continous Density)&lt;/h1&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;# Transformation of bike parking lots shape to data.frame for ggplot2
RK &amp;lt;- data.frame(coordinates(bmap)) 
colnames(RK) &amp;lt;- c(&amp;quot;long&amp;quot;,&amp;quot;lat&amp;quot;) 
ws2 &amp;lt;- fortify(smap,region=&amp;quot;OBJECTID&amp;quot;)  
wmap2 &amp;lt;- wmap 
# assign ID
wmap2@data$id &amp;lt;- rownames(wmap2@data) 
# transform to data.frame  
test1 &amp;lt;- fortify(wmap2, region=&amp;quot;id&amp;quot;)   
# This may take a while...
ggplot(data=test1) + 
  aes(x=long,y=lat) + 
  geom_polygon(aes(group=group),col=&amp;quot;black&amp;quot;,fill=NA) + 
  geom_point(data=RK,aes(x=long,y=lat)) + 
  geom_line(data=ws2,aes(group=group))+ xlab(&amp;quot;longitude&amp;quot;)+ylab(&amp;quot;latitude&amp;quot;) + 
  stat_density2d(data=RK,aes(fill = ..level..),size=1,bins=200,alpha=0.1, geom=&amp;quot;polygon&amp;quot;,n=100) +  
  ggtitle(&amp;quot;Bike Parking Lots in Vienna 2015&amp;quot;)+ scale_fill_continuous(name = &amp;quot;Kernel Density&amp;quot;) + 
  theme_bw() 
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/bikedensity-plot2-1.png&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; alt=&#34;Null&#34;&gt;&lt;/p&gt;

&lt;p&gt;Comments/Pull Requests welcome!&lt;/p&gt;

&lt;p&gt;Authors: Christian Brandstaetter with minor modifications by Mario Annau&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Population Densities in Vienna</title>
          <link>/blog/2015/10/30/population-densities-in-vienna/</link>
          <pubDate>Fri, 30 Oct 2015 22:20:00 &#43;0200</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2015/10/30/population-densities-in-vienna/</guid>
          <description>&lt;p&gt;The new Github repository &lt;a href=&#34;https://github.com/ViennaR/OpenGovernmentVienna&#34;&gt;OpenGovernmentVienna&lt;/a&gt; has already been created including a very nice addition by Christian.&lt;/p&gt;

&lt;p&gt;The script executes the following steps:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Download Vienna map including district boundaries from &lt;a href=&#34;http://data.wien.gv.at&#34;&gt;http://data.wien.gv.at&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;Download population and district size from &lt;a href=&#34;https://www.wien.gv.at/statistik&#34;&gt;https://www.wien.gv.at/statistik&lt;/a&gt;, calculate population density.&lt;/li&gt;
&lt;li&gt;Plot Vienna map coloured by population density.&lt;/li&gt;
&lt;/ol&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;library(rgdal) 
library(rgeos) 
library(XML)
library(RCurl)

download.vienna.bydistrict &amp;lt;- function(tablename, skip.row = 3) {
	baseurl &amp;lt;- &amp;quot;https://www.wien.gv.at/statistik&amp;quot;
	popurl &amp;lt;- sprintf(&amp;quot;%s/%s.html&amp;quot;, baseurl, tablename)

	poptable &amp;lt;- readHTMLTable(getURL(popurl))[[1]]
	poptable &amp;lt;- poptable[-c(1:skip.row), ]
	poptable &amp;lt;- poptable[, -1]
	row.names(poptable) &amp;lt;- NULL
	poptable &amp;lt;- sapply(poptable, function(x) gsub(&amp;quot;.&amp;quot;, &amp;quot;&amp;quot;, x, fixed = TRUE))
	poptable &amp;lt;- gsub(&amp;quot;,&amp;quot;, &amp;quot;.&amp;quot;, poptable, fixed = TRUE)
	poptable &amp;lt;- matrix(as.numeric(poptable), nrow = nrow(poptable))
	poptable
}

# Download Data Shape Data District Boundaries
mapdata &amp;lt;- &amp;quot;http://data.wien.gv.at/daten/geo?service=WFS&amp;amp;request=GetFeature&amp;amp;version=1.1.0&amp;amp;typeName=ogdwien:BEZIRKSGRENZEOGD&amp;amp;srsName=EPSG:4326&amp;amp;outputFormat=shape-zip&amp;quot;

dir.create(&amp;quot;data&amp;quot;, showWarnings = FALSE)
destfile &amp;lt;- &amp;quot;data/BEZIRKSGRENZEOGD.zip&amp;quot;
download.file(mapdata, destfile = destfile)
unzip(destfile, exdir = &amp;quot;data/BEZIRKSGRENZEOGD&amp;quot;)
file.remove(destfile)
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code&gt;## [1] TRUE
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;# Read District Boundaries
wmap &amp;lt;- readOGR(&amp;quot;data/BEZIRKSGRENZEOGD&amp;quot;, layer=&amp;quot;BEZIRKSGRENZEOGDPolygon&amp;quot;) 
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code&gt;## OGR data source with driver: ESRI Shapefile 
## Source: &amp;quot;data/BEZIRKSGRENZEOGD&amp;quot;, layer: &amp;quot;BEZIRKSGRENZEOGDPolygon&amp;quot;
## with 23 features
## It has 15 fields
&lt;/code&gt;&lt;/pre&gt;

&lt;pre&gt;&lt;code class=&#34;language-r&#34;&gt;# Download Population per district
distpop &amp;lt;- download.vienna.bydistrict(&amp;quot;bevoelkerung/tabellen/bevoelkerung-alter-geschl-bez&amp;quot;)
distpopsum &amp;lt;- rowSums(as.matrix(distpop))

# Download Size of Each district
distsize &amp;lt;- download.vienna.bydistrict(&amp;quot;lebensraum/tabellen/nutzungsklassen-bez&amp;quot;, skip.row = 2)
distsizekm2 &amp;lt;- distsize[, 1] / 100

wd &amp;lt;- data.frame(distpopsum / distsizekm2)

centroids &amp;lt;- gCentroid(wmap, byid=TRUE) 
colfunc &amp;lt;- colorRampPalette(c(&amp;quot;lightblue&amp;quot;, &amp;quot;darkred&amp;quot;)) 
colnames(wd) &amp;lt;- &amp;quot;inh&amp;quot; 
wd$district &amp;lt;- seq(1,23) 
anstieg_pop &amp;lt;- wd$district[order(wd$inh)] 
colsort &amp;lt;- colfunc(23)[order(anstieg_pop)] 
layout(matrix(c(1,2), byrow = TRUE),height=c(1.3, 0.7)) 
par(mar=c(0,0,0,0)) 
plot(wmap,col=colsort[wmap$BEZNR]) 
text(as.character(wmap$BEZ_RZ), x = centroids@coords[,1], y = centroids@coords[,2],cex=0.8) 
par(mar=c(3,4,4,2),mgp=c(2,0.7,0)) 
barplot(wd$inh,main=&amp;quot;Population Density Vienna 2014&amp;quot;,yaxt=&amp;quot;n&amp;quot;,col=colsort,xlab=&amp;quot;District&amp;quot;,beside=T, ylab=expression(paste(&amp;quot;1.000 Inhabitants km&amp;quot;^-2)),names.arg=as.roman(wd$district),las=2) 
axis(2,labels=c(&amp;quot;0&amp;quot;,&amp;quot;5&amp;quot;,&amp;quot;10&amp;quot;,&amp;quot;15&amp;quot;,&amp;quot;20&amp;quot;),at=c(0,5000,10000,15000,20000),las=1) 
abline(h=c(5000,10000,15000,20000),lty=2) 
legend(&amp;quot;topright&amp;quot;,pch=c(15,15),col=c(&amp;quot;darkred&amp;quot;,&amp;quot;lightblue&amp;quot;),c(&amp;quot;high&amp;quot;,&amp;quot;low&amp;quot;),bg=&amp;quot;white&amp;quot;) 
&lt;/code&gt;&lt;/pre&gt;

&lt;p&gt;&lt;img src=&#34;/img/popdensity-1.png&#34; style=&#39;width:100%;&#39; border=&#34;0&#34; alt=&#34;Null&#34;&gt;&lt;/p&gt;

&lt;p&gt;Authors: Christian Brandstaetter with small additions by Mario Annau&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 10-2015 Minutes, Kick-Off!</title>
          <link>/blog/2015/10/21/meetup-10-2015-minutes-kick-off/</link>
          <pubDate>Wed, 21 Oct 2015 21:00:00 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2015/10/21/meetup-10-2015-minutes-kick-off/</guid>
          <description>&lt;p&gt;&lt;img src=&#34;/img/Kickoff-Charlie-Brown.gif&#34; alt=&#34;Kick-Off&#34; /&gt;&lt;/p&gt;

&lt;p&gt;After a fruitful discussion regarding the future direction of the R User Group just a little summary of the &lt;strong&gt;HOT&lt;/strong&gt; topics. The discussion started  with a very nice and detailed introduction of new &lt;em&gt;ViennaR&lt;/em&gt; participants and was followed by an interesting and&amp;mdash;as expected and encouraged&amp;mdash;controversial discussion about project goals, project management and topics.&lt;/p&gt;

&lt;h2 id=&#34;organisation&#34;&gt;Organisation&lt;/h2&gt;

&lt;p&gt;Our goal is to have monthly Meetups. Currently they are scheduled by the end of each month, however, this may change depending on the availability of the room. A big thank you to &lt;a href=&#34;http://www.statistik.tuwien.ac.at/public/templ&#34;&gt;Matthias&lt;/a&gt; who makes this Meetup at the Institute of Computational Statistics [1] possible!&lt;/p&gt;

&lt;p&gt;[1] At the legendary &lt;em&gt;Goldenes Lamm&lt;/em&gt;, birthplace at/of &lt;a href=&#34;https://cran.r-project.org/web/packages/e1071/index.html&#34;&gt;e1071&lt;/a&gt;.&lt;/p&gt;

&lt;h2 id=&#34;presentations&#34;&gt;Presentations&lt;/h2&gt;

&lt;p&gt;Like in previous Meetups we are happy to have some presentations of &lt;em&gt;ViennaR&lt;/em&gt; participants. Just drop us a mail that you are planning a presentation during one of the next meetups. We would also be happy to invite some guest speakers from time to time.&lt;/p&gt;

&lt;h2 id=&#34;projects&#34;&gt;Projects&lt;/h2&gt;

&lt;p&gt;To have an even more collaborative experience during &lt;em&gt;ViennaR&lt;/em&gt; Meetups plan to have different projects on our &lt;a href=&#34;https://github.com/viennar&#34;&gt;Github&lt;/a&gt; page. It should be manageable to have up to 5 people working on specific projects with an expected duration of about 3 Months.
The following project streams have been selected and coordinators assigned:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;&lt;a href=&#34;https://en.wikipedia.org/wiki/Fantasy_sport&#34;&gt;Fantasy Sports&lt;/a&gt;, e.g. from Yahoo (short intro/write&amp;ndash;up by Daniel)

&lt;ul&gt;
&lt;li&gt;Put team together based on statistics, e.g. Baseball&lt;/li&gt;
&lt;li&gt;References:

&lt;ul&gt;
&lt;li&gt;Book: &lt;a href=&#34;http://www.amazon.com/Analyzing-Baseball-Data-Chapman-Series/dp/1466570229&#34;&gt;Analyzing Baseball data in R&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href=&#34;http://fantasyfootballanalytics.net/2014/06/scraping-fantasy-football-projections.html&#34;&gt;Scraping Fantasy Football Projections from the Web&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;Parliamentary voting results (input from Mateusz)

&lt;ul&gt;
&lt;li&gt;Harvesting data, e.g. &lt;a href=&#34;https://www.data.gv.at/katalog/dataset/1a3c44a8-7361-4dc7-ac0d-6ec19ddba191&#34;&gt;Beschlüsse des Nationalrats&lt;/a&gt; from &lt;a href=&#34;https://www.data.gv.at&#34;&gt;https://www.data.gv.at&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;Visualization&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;Text mining (Mario)

&lt;ul&gt;
&lt;li&gt;Public newspapers&lt;/li&gt;
&lt;li&gt;Most popular words in Newspapers&lt;/li&gt;
&lt;li&gt;HC Strache Posts on Facebook&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;Use Kaggle as an Inspiration (Walter)

&lt;ul&gt;
&lt;li&gt;Hillary Clintons E-mails&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;We will use the &lt;a href=&#34;http://www.meetup.com/ViennaR/messages/boards/forum/2693272&#34;&gt;Meetup Discussion Forum&lt;/a&gt; to futher elaborate on &lt;a href=&#34;http://www.meetup.com/ViennaR/messages/boards/thread/49348746&#34;&gt;Project ideas&lt;/a&gt;.&lt;/p&gt;

&lt;h2 id=&#34;other-topics&#34;&gt;Other Topics&lt;/h2&gt;

&lt;p&gt;We would also be happy to have some other fixed discussion topics, e.g. book presentations, new R packages around, etc&amp;hellip;
Just let us know!&lt;/p&gt;

&lt;h2 id=&#34;next-steps&#34;&gt;Next Steps&lt;/h2&gt;

&lt;p&gt;For the next meetup we are planning a short Git(hub) and R package dev intro (Mario) and hope that each assigned coordinator can give us a little heads up of her topic.&lt;/p&gt;

&lt;p&gt;Looking forward to see you at the next meetup in November!&lt;/p&gt;</description>
        </item>
      
    
      
        <item>
          <title>Meetup 10-2015</title>
          <link>/blog/2015/10/20/meetup-10-2015/</link>
          <pubDate>Tue, 20 Oct 2015 21:31:57 CEST</pubDate>
          <author>Vienna-R</author>
          <guid>/blog/2015/10/20/meetup-10-2015/</guid>
          <description>&lt;p&gt;Main focus this time will be to discuss future directions/projects for ViennaR:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Our goal is to define R-package(s) organized as Github project(s) that can be realised within a timeframe of approx. one year resulting in CRAN release&lt;br /&gt;&lt;/li&gt;
&lt;li&gt;Agree on the main areas (project streams) - not more than 2, max. 3&lt;br /&gt;&lt;/li&gt;
&lt;li&gt;Brainstorm on potential project and/or package ideas&lt;/li&gt;

&lt;li&gt;&lt;p&gt;Decide on project leads, team members respectively form of collaboration (fork &amp;amp; pull vs. shared repo), type of license (GPL?), etc.&lt;/p&gt;&lt;/li&gt;

&lt;li&gt;&lt;p&gt;Evaluate interest in active preparation of short analytical topics  (monthly or bi-monthly) to be worked off individually or in groups (alternative: as a hackathon)&lt;br /&gt;
e.g. run a short analysis on some topic based on publicly available data;  present these @ meetup, compare approaches to project solution.&lt;/p&gt;&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Example topics could cover&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Open government data:

&lt;ul&gt;
&lt;li&gt;Text mine parliamentary votes (by time/topic) (source: &lt;a href=&#34;https://www.data.gv.at/katalog/dataset/1a3c44a8-7361-4dc7-ac0d-6ec19ddba191&#34;&gt;https://www.data.gv.at/katalog/dataset/1a3c44a8-7361-4dc7-ac0d-6ec19ddba191&lt;/a&gt;)&lt;br /&gt;&lt;/li&gt;
&lt;li&gt;Graphically display distribution of specific points of interest like voting booths, public toilets, etc. in Vienna based upon &lt;a href=&#34;https://open.wien.gv.at/site/open-data/)&#34;&gt;https://open.wien.gv.at/site/open-data/)&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;News-related data:

&lt;ul&gt;
&lt;li&gt;Analyze comments for specific topics of derstandard.at&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;li&gt;Price-data

&lt;ul&gt;
&lt;li&gt;e.g. correlate fuel prices in Austria with crude oil prices&lt;/li&gt;
&lt;/ul&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;We are looking forward to a fruitful/interesting/controversial discussion!&lt;/p&gt;</description>
        </item>
      
    

  </channel>
</rss>