french_medias.html

<style type="text/css" media="screen">
/*
.nodes-image {
	margin:-100;
}
*/	
@import url("//maxcdn.bootstrapcdn.com/font-awesome/4.4.0/css/font-awesome.min.css");

.imageblock .content img, .image img {max-width: 900px;max-height: 300px;}
.deck h3, .deck h4 {display: block !important;margin-bottom:8px;margin-top:5px;}
.listingblock {margin:8px;}
.pull-bottom {position:relative;bottom:1em;}
.admonitionblock td.icon [class^="fa icon-"]{font-size:2.5em;text-shadow:1px 1px 2px rgba(0,0,0,.5);cursor:default}
.admonitionblock td.icon .icon-note:before{content:"\f05a";color:#19407c}
.admonitionblock td.icon .icon-tip:before{content:"\f0eb";text-shadow:1px 1px 2px rgba(155,155,0,.8);color:#111}
.admonitionblock td.icon .icon-warning:before{content:"\f071";color:#bf6900}
.admonitionblock td.icon .icon-caution:before{content:"\f06d";color:#bf3400}
.admonitionblock td.icon .icon-important:before{content:"\f06a";color:#bf0000}
.admonitionblock.note.speaker { display:none; }
</style>
<style type="text/css" media="screen">
/* #editor.maximize-editor .CodeMirror-code { font-size:24px; line-height:26px; } */
</style>
<article class="guide" ng-controller="AdLibDataController">
  <carousel class="deck container-fluid">
    <!--slide class="row-fluid">
      <div class="col-sm-3">
        <h3>French/francophones medias</h3>
        <p class="lead">Information</p>
			<!dl>
				
				
				
				
				
			</dl>
		</div>
      <div class="col-sm-9">
        <figure>
          <img style="width:300px" src=""/>
        </figure>
      </div>
    </slide-->
    


<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>French/francophones medias</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>Jérôme BATON
v0.6, 2018-07-25
:toc: right
:imagesdir: images</p>
</div>
<div class="paragraph">
<p>This guide is written in asciidoc and converted to html.</p>
</div>
<div class="paragraph">
<p>Original URL is  <a href="https://wadael.github.io/presse_fr/french_medias.adoc" class="bare">https://wadael.github.io/presse_fr/french_medias.adoc</a></p>
</div>



   <h4>What is a Neo4j "GUIDE"</h4>
   <div class="paragraph">
<p>It is a web page to be read within the Neo4j browser (the builtin webapp accessible on port 7474). You certainly already know the movies guide that you ran with</p>
</div>
<div class="literalblock">
<div class="content">
<pre>:play movies</pre>
</div>
</div>



   <h4>Description</h4>
   <div class="paragraph">
<p>This Neo4j formatted guide&#8217;s goal is to show who owns what in the french media community by playing with some features of Neo4j.</p>
</div>
<div class="paragraph">
<p>We will see:</p>
</div>
<div class="ulist">
<ul>
<li>
<p>data loading</p>
</li>
<li>
<p>apply some algorithms</p>
</li>
<li>
<p>answer a few obvious questions</p>
</li>
</ul>
</div>
<div class="paragraph">
<p>Disclaimer: this guide has no journalistic pretention. I wrote it to share my reuse of existing data.</p>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Getting data</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>The data is published by the team of the best French weekly, Le Monde Diplomatique, at this URL <a href="https://github.com/mdiplo/Medias_francais" class="bare">https://github.com/mdiplo/Medias_francais</a>.</p>
</div>
<div class="paragraph">
<p>Thanks a lot guys</p>
</div>
<div class="paragraph">
<p>Get the two .tsv files and drop them into the import folder of your Neo4j server.</p>
</div>
<div class="paragraph">
<p>Tsv files are text files using tabulation as a field separator</p>
</div>
<div class="paragraph">
<p>(Do the same with other files used later)</p>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Modeling</h3>
    <br/>
    <div>
      <div class="paragraph">
<p><em>medias_francais.tsv</em> contains lines (Entities) which can be either contain contient (real) persons, companies or medias (tv, press,..).</p>
</div>
<div class="paragraph">
<p><em>relations_medias_francais.tsv</em> contains the relations between those entites.
The field 'valeur' gives the owning percentage of the 'source' entity on the 'cible' (target) entity.
The relation to be created between the entities will be 'OWNS'.</p>
</div>



   <h4>Labels</h4>
   <div class="paragraph">
<p>The labels to be used will be Les étiquettes que j&#8217;envisage sont: Entite, Person,and Media.</p>
</div>
<div class="paragraph">
<p>As the names will come from the content of the file which I dont want to change, and as this English version comes after the French version, I&#8217;ll keep the Fr names from the import and add equivalent EN labels thereafter</p>
</div>
<div class="paragraph">
<p>Because of the French content, in correct French, with spaces and accentuated e in the names, some data changes were to be made on import.
There are kept.</p>
</div>
<div class="paragraph">
<p>Please note that only one transformation can be made at import time on a field with plain Cypher.
Unless you decide to get something unreadable by chaining the transformations.
The best option for several modifications is to use a custom function.</p>
</div>
<div class="paragraph">
<p>In my book, I repeated not to forget to create constraints, for the checking and the automatic index creation.</p>
</div>
<div class="paragraph">
<p>So, even with known data of a small volume thus no need of an index, lets run this:</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->CREATE CONSTRAINT ON (e:Entity) ASSERT e.name IS UNIQUE<!--/code--></pre>
</div>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Data loading</h3>
    <br/>
    <div>
      


   <h4>Code</h4>
   <div class="paragraph">
<p>The codes proposed are</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->LOAD CSV WITH HEADERS FROM 'file:///medias_francais.tsv' AS line FIELDTERMINATOR '\t'
MERGE (node0:Entity{name: line.nom})
FOREACH(uselessVar IN CASE WHEN trim(line.`rangChallenges`) &lt;&gt; "" THEN [1] ELSE [] END | SET node0.challengesRanking= toInt(line.`rangChallenges`))
FOREACH(uselessVa IN CASE WHEN trim(line.`mediaType`) &lt;&gt; "" THEN [1] ELSE [] END | SET node0.mediaType = line.`mediaType`)
FOREACH(uselessV IN CASE WHEN trim(line.`mediaPeriodicite`) &lt;&gt; "" THEN [1] ELSE [] END | SET node0.mediaPeriodicity = line.`mediaPeriodicite`)
FOREACH(useless IN CASE WHEN trim(line.`mediaEchelle`) &lt;&gt; "" THEN [1] ELSE [] END | SET node0.mediaScale = line.`mediaEchelle`)
FOREACH(u IN CASE WHEN trim(line.commentaire) &lt;&gt; "" THEN [1] ELSE [] END | SET node0.comment = line.commentaire)

WITH node0,line
CALL apoc.create.addLabels(node0,[replace(line.`typeLibelle`," ","_")]) YIELD node
RETURN node<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>The field 'challengesRanking' is the ranking of the richest French nationals. It comes from the 'Challenges' periodical.</p>
</div>
<div class="paragraph">
<p>The code for the second file is:</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->LOAD CSV WITH HEADERS FROM 'file:///relations_medias_francais.tsv' AS ligne FIELDTERMINATOR '\t'
MERGE (nodeO:Entity{name: ligne.origine})
MERGE (nodeC:Entity{name: ligne.cible})
MERGE (nodeO)-[p:OWNS]-&gt;(nodeC)

FOREACH(u IN CASE WHEN trim(ligne.valeur) &lt;&gt; "" THEN [1] ELSE [] END | SET p.percentage = toInt(ligne.valeur))
FOREACH(u IN CASE WHEN trim(ligne.source) &lt;&gt; "" THEN [1] ELSE [] END | SET p.source = ligne.source)
FOREACH(u IN CASE WHEN trim(ligne.datePublication) &lt;&gt; "" THEN [1] ELSE [] END | SET p.publicationDate = ligne.datePublication)<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>To add the English labels and remove the French ones</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (pp:Personne_physique)
SET pp:Person
REMOVE pp:Personne_physique
WITH pp
MATCH (pm:Personne_morale)
SET pm:Company
REMOVE pm:Personne_morale
WITH pp, pm
MATCH (m:Média)
SET m:Media
REMOVE m:Média
RETURN pp,pm,m<!--/code--></pre>
</div>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Overview</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>Let&#8217;s see who owns what with a simple query</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH pom = (pp:Person)-[:OWNS*]-&gt;(m:Media)
RETURN pom<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>The star * character means that there can be more than one OWN relation in the path.</p>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Pagerank</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>Let&#8217;s launch the PageRank algorithm, just to see the most connected nodes.
This uses the APOC plugin.</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (node:Entity)
WITH collect(node) AS nodes
CALL apoc.algo.pageRank(nodes) YIELD node, score
RETURN node.name, score
ORDER BY score DESC<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>We get the most connected nodes first.
Surprisingly, the top 3 is</p>
</div>
<table class="table tableblock frame-all grid-all" style="width: 100%;">
<colgroup>
<col style="width: 50%;">
<col style="width: 50%;">
</colgroup>
<tbody>
<tr>
<td class="tableblock halign-left valign-top"><p class="tableblock">"La Tribune"</p></td>
<td class="tableblock halign-left valign-top"><p class="tableblock">0.69473</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top"><p class="tableblock">"Hima Groupe"</p></td>
<td class="tableblock halign-left valign-top"><p class="tableblock">0.64087</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top"><p class="tableblock">"Arte"</p></td>
<td class="tableblock halign-left valign-top"><p class="tableblock">0.62095</p></td>
</tr>
</tbody>
</table>
<div class="paragraph">
<p>Now, for the persons :</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (no:Person) WITH collect(no) AS nodes
CALL apoc.algo.pageRankWithConfig(nodes,{types:'OWNS'}) YIELD node, score
RETURN node.name, score
ORDER BY score DESC<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>Surprisingly, all the persons listed have a score of 0.15.
It means they all have the same number of connections. Weird coincidence.</p>
</div>
<div class="paragraph">
<p>Lets look for who owns the most medias, with a graph answer</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (pp:Person)-[p:OWNS]-&gt;(m:Media)
RETURN pp.name, collect(m.name) as titles, size(collect(m.name)) as nbTitles ORDER BY nbTitles DESC<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>Only direct possesions are given and they seem to only have one each. This explains that.</p>
</div>
<div class="paragraph">
<p>Let&#8217;s dig deeper.</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (pp:Person)-[p:OWNS*]-&gt;(m:Media)
RETURN pp.name, collect(m.name) as titles, size(collect(m.name)) as nbTitles ORDER BY nbTitles DESC<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>Look at the star. It gives a more credible result.</p>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Correlation medias count et Challenges ranking ?</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>Let&#8217;s see if the persons owning the most medias have the highest ranking.
What do you think ?</p>
</div>
<div class="paragraph">
<p>Try this Cypher query</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (pp:Person)-[p:OWNS*..]-&gt;(m:Media)
RETURN  coalesce(toInt(pp.challengesRanking),0) as rank, size(collect(m.name)) as nbTitles ORDER BY rank ASC<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>For a quick visual result, I use Spoon (which I believe was created by Michael Hunger) (:play spoon.html and follow instructions).
I got this chart as a result.</p>
</div>
<div class="imageblock">
<div class="content">
<img src="https://wadael.github.io/presse_fr/images/barchart_presse_01.png" alt="barchart presse 01">
</div>
</div>
<div class="paragraph">
<p>I attributed rank zero to the persons without a ChallengesRanking. That explains why the first column is high.</p>
</div>
<div class="paragraph">
<p>I expected a declining shape. Certainly it was true in the middle of the 20th century. Maybe.</p>
</div>
<div class="paragraph">
<p>Remember, extra data is about press which obviously puts aside tv.</p>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Data aggregation</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>A media list with owners is a good start but. The more data the better.
Its not uncommon to hear that there is many taxes in France.
It is also true there is a lot of governmental grants, and several for the press, on different criteria.</p>
</div>
<div class="paragraph">
<p>Amongst the questions a tax payer may have, I have:</p>
</div>
<div class="ulist">
<ul>
<li>
<p>what are the medias getting money ?</p>
</li>
<li>
<p>what are their diffusion ?</p>
</li>
<li>
<p>who owns them ?</p>
</li>
<li>
<p>are the owners filthy rich already ?</p>
</li>
<li>
<p>which owner has the most impact on the population ?</p>
</li>
</ul>
</div>



   <h4>Grants to press</h4>
   <div class="paragraph">
<p>I found some answers via the Internet</p>
</div>
<div class="paragraph">
<p><a href="http://archive.wikiwix.com/cache/?url=https%3A%2F%2Fwww.data.gouv.fr%2Ffr%2Fdatasets%2Faides-a-la-presse-classement-des-titres-de-presse-aides%2F">here</a>, <a href="https://www.data.gouv.fr/fr/datasets/aides-a-la-presse-classement-des-titres-de-presse-aides/">there</a> for years 2012 to 2016.</p>
</div>
<div class="paragraph">
<p>This is human-readable XLS format, with fancy presentation inside that forbids to convert it to csv and import dramalessly.</p>
</div>
<div class="paragraph">
<p>MATCHing on the name sounds obvious. But this file does not use the same naming than the first two.</p>
</div>
<div class="paragraph">
<p>For example "L&#8217;Humanité" is referenced as "Humanité (L')".
Same for "Le Figaro" et "Figaro (Le)"</p>
</div>
<div class="paragraph">
<p>And I&#8217;m writing this guide for fame, not for fortune so &#8230;&#8203; I can not use this file as only part of it would be used. Unless I rework the data.</p>
</div>
<div class="paragraph">
<p>The two biggest beneficiaries are:</p>
</div>
<table class="table tableblock frame-all grid-all" style="width: 100%;">
<colgroup>
<col style="width: 50%;">
<col style="width: 50%;">
</colgroup>
<tbody>
<tr>
<td class="tableblock halign-left valign-top"><p class="tableblock">SOCIETE COMMUNE POUR LES INFRASTRUCTURES DE LA DISTRIBUTION DE LA PRESSE</p></td>
<td class="tableblock halign-left valign-top"><p class="tableblock">4 419 926,00 €</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top"><p class="tableblock">IMPRIMERIE DE L&#8217;AVESNOIS</p></td>
<td class="tableblock halign-left valign-top"><p class="tableblock">2 503 123,00 €</p></td>
</tr>
</tbody>
</table>
<div class="paragraph">
<p>Unknown to me.</p>
</div>
<div class="paragraph">
<p>Other beneficiaries are listed as groups of websites like
 j7.agefi.fr; agefi.fr; L&#8217;Agefi Hebdo Magazine</p>
</div>
<div class="paragraph">
<p>for	159 708,00 €</p>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>One hit, two features</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>Luckily, I have found this  <a href="https://www.data.gouv.fr/fr/datasets/r/6f66e292-c4d9-439e-916d-26257107833a">URL</a> corresponding to a CSV file with both grants and diffusion informations.</p>
</div>
<div class="paragraph">
<p>Before to import and create nodes, I write a RETURN query to check the columns.</p>
</div>
<div class="paragraph">
<p>That helps to see that some newspapers are given with their website.
This will need to do some string splitting.</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->WITH "file:///aides-a-la-presse-classement-des-titres-de-presse-aides.csv" AS url
LOAD CSV WITH HEADERS FROM url AS ligne FIELDTERMINATOR ";"
RETURN ligne.`Bénéficiaires` as Beneficiary, split( ligne.`Bénéficiaires`,"/"), ligne.`Total des aides individuelles (en €)` as totalGrants,ligne.`Diffusion annuelle (en exemplaires` as yearlyDiffusion, ligne.`Année` as year LIMIT 5<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>This new content will be inserted in a new node labelled 'Statistic' linked to the  Media node with a 'DESCRIBE' relation.</p>
</div>
<div class="paragraph">
<p>First,</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (m:Media)
SET m.upName = toUpper(m.name)
RETURN count(m)<!--/code--></pre>
</div>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->WITH "file:///aides-a-la-presse-classement-des-titres-de-presse-aides.csv" AS url
LOAD CSV WITH HEADERS FROM url AS ligne FIELDTERMINATOR ";"

MERGE(m:Media {upName:toUpper( split( ligne.`Bénéficiaires`,"/")[0] ) })
SET m.name=split( ligne.`Bénéficiaires`,"/")[0]
CREATE (s:Statistic { year: toInt(ligne.`Année`), totalGrants: toInt(ligne.`Total des aides individuelles (en €)`), yearlyDiffusion: toInt(ligne.`Diffusion annuelle (en exemplaires`)  })

CREATE (m)&lt;-[d:DESCRIBE]-(s)
RETURN m,d,s<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>We notice that some press media do not have statistics for 2015 and 2016.</p>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Who gets the most euros ?</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>Who gets the most grant money</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (pp:Person)-[p:OWNS*..]-&gt;(m:Media)&lt;--(s:Statistic)
RETURN pp.name, coalesce(toInt(pp.challengesRanking),0) as challengesRanking, size(collect(m.name)) as nbTitles, sum(s.totalGrants) AS jackpot ORDER BY jackpot DESC<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>Interesting sums.
However, before to cry wolf, checking the percentage owned of each node between the person and the media is necessary.
It is not because a title of a billionaire gets millions that those millions ends in the billionaire&#8217;s (overseas tax-free) pockets.</p>
</div>
<div class="paragraph">
<p>This can be queried too :)</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH path=(p:Person)-[o:Owns*]-&gt;(m:Media)
RETURN p.name, m.name, reduce(prod=1.0, x in relationships(path) | prod*coalesce( x.value, 0.0 )/100 ) as possession
ORDER BY possession DESC<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>I thank Stefan Armbruster (Neo4j) who reminded me the existence of the  reduce function.</p>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Bonus - Some R</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>From time to time, I like to see other languages than those I am used to. Even more when they are specialized to do quickly a job I want to do.</p>
</div>
<div class="paragraph">
<p>The <a href="https://www.r-project.org/">R language</a> is the current best choice to get charts quickly, like those two next representations of who gets the most governmental money.</p>
</div>



   <h4>Neo4j result to R</h4>
   <div class="paragraph">
<p>I got the values as text in the Neo4j browser, then used a spreadsheet to convert from columns to lines.
A text editor with a column mode can be used too.</p>
</div>



   <h4>Pie chart</h4>
   <div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="r" lang="r"><!--code class="r language-r"-->group=c("Bernard Arnault","Famille Hutin","Famille Hurbain","Famille Lemoine","Famille Baylet","Xavier Niel","Matthieu Pigasse","Arnaud Lagardère","Bernard Tapie","Famille Mohn","Yves de Chaisemartin","Famille Saint-Cricq","François Pinault","Philippe Hersant","Édouard Coudurier")

value=c(18454766,7240740,2970948,2806813,2317190,1565190,1565190,1533157,1077799,550688	,250628	,243025	,219387,43018,11565)

pie(value, group)<!--/code--></pre>
</div>
</div>
<div class="paragraph">
<p>That creates an unreadable chart.</p>
</div>
<div class="imageblock">
<div class="content">
<img src="https://wadael.github.io/presse_fr/images/press_pie.png" alt="pie chart">
</div>
</div>



   <h4>Treemap</h4>
   <div class="paragraph">
<p>A treemap shares its total area between the results, according to their values</p>
</div>
<div class="imageblock">
<div class="content">
<img src="https://wadael.github.io/presse_fr/images/treemap_presse.png" alt="Treemap">
</div>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="r" lang="r"><!--code class="r language-r"-->library(treemap)

group=c("Bernard Arnault","Famille Hutin","Famille Hurbain","Famille Lemoine","Famille Baylet","Xavier Niel","Matthieu Pigasse","Arnaud Lagardère","Bernard Tapie","Famille Mohn","Yves de Chaisemartin","Famille Saint-Cricq","François Pinault","Philippe Hersant","Édouard Coudurier")

value=c(18454766,7240740,2970948,2806813,2317190,1565190,1565190,1533157,1077799,550688	,250628	,243025	,219387,43018,11565)

data=data.frame(group,value)

# treemap
treemap(data,index="group",vSize="value",type="index")<!--/code--></pre>
</div>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Who has the most impact ?</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>I say impact because in my opinion too many persons take what&#8217;s written for truth.</p>
</div>
<div class="paragraph">
<p>And some moguls are known to make print what serves them best.
Or not print what would disserve them.</p>
</div>
<div class="paragraph">
<p>So, imho, it is an interesting metric.
However, we know its TVs that makes opinions now, and fac*b*k.</p>
</div>
<div class="listingblock">
<div class="content">
<pre mode="cypher"  class="highlight pre-scrollable programlisting cm-s-neo code runnable standalone-example ng-binding" data-lang="cypher" lang="cypher"><!--code class="cypher language-cypher"-->MATCH (pp:Person)-[p:OWNS*..]-&gt;(m:Media)&lt;--(s:Statistic)
RETURN pp.name, coalesce(toInt(pp.challengesRanking),0) as challengesRanking, size(collect(m.name)) as nbTitles, apoc.number.format( sum(s.yearlyDiffusion)) AS targetSize ORDER BY targetSize DESC<!--/code--></pre>
</div>
</div>
	</div>
  </div>
</slide>



<slide class="row-fluid">
  <div class="col-sm-12">
    <h3>Self-promotion</h3>
    <br/>
    <div>
      <div class="paragraph">
<p>If you want to learn more about Neo4j, my book <a href="https://www.packtpub.com/big-data-and-business-intelligence/learning-neo4j-3x-second-edition?referrer=wadael">Learning Neo4j, Second Edition</a> is available on many websites.</p>
</div>
<div class="paragraph">
<p>My editor: Packt Publishing often does discount campaigns (like any book for 10$, 3 for 25$)</p>
</div>
<div class="imageblock">
<div class="content">
<img src="https://wadael.github.io/presse_fr/images/B05824_cover.png" alt="B05824 cover">
</div>
</div>
<div class="paragraph">
<p>Do you need a consultant ?
Contact me via <a href="https://www.linkedin.com/in/jeromebaton/">LinkedIn</a> or via email: firstnameLastname followed by .<a href="mailto:pro@gmail.com">pro@gmail.com</a></p>
</div>
<div class="paragraph">
<p>Thanks for reading the ad.</p>
</div>
<div class="paragraph">
<p>Your comments are welcome (pr, issues).
This guide may get more content based on your feedback</p>
</div>
<div class="paragraph">
<p>jerome baton</p>
</div>
	</div>
  </div>
</slide>
  </carousel>
</article>