Node-link polar layout with centrality (weighted eigenvector)

An improvement upon the previous experiment. This time, a variant of eigenvector centrality that takes into account the weights of the links is computed, yielding a very different diagram.

In order to do that, the network's adjacency matrix is populated with weights in place of binary (connected / not connected) information. Computing the centrality measure using this matrix is sufficient to obtain a result that takes topology and weights into account, as discussed in Newman 2004 (alongwith other analysis methods for weighted networks).

xxxxxxxxxx
 
<!DOCTYPE html>
<html>
    <head>
        <meta charset="utf-8">
        <meta name="description" content="Node-link polar layout with centrality (weighted eigenvector)" />
        <title>Node-link polar layout with centrality (weighted eigenvector)</title>
        <link type="text/css" href="index.css" rel="stylesheet"/>
        <script src="https://d3js.org/d3.v3.min.js"></script>
    </head>
    <body>
        <svg height="500" width="960"></svg>
        <script src="index.js"></script>
    </body>
</html>
    

Modified http://d3js.org/d3.v3.min.js to a secure url

https://d3js.org/d3.v3.min.js