Analyzing Twitter networks

For this demo we collected info about more than 100.000 Twitter accounts which :

  1. Could be identified as being located in Belgium
  2. Had the description field on their profile completed.

The aim of the project is to build a list of people with a specific interest and eventually find a way to contact them.


The Twitter accounts are collected by using different tools and were further processed by Google Refine. Next the data was loaded into our Semantic Analyses

In contrast to the RSS analysis the list of “top concepts” in this case already does reveal usefull information. What we see here is that many people are interested in music, consider themselves as a student and do something about marketing and media.


By scrolling down the list of top concepts we get a nice overview of relevant topics from which  we can retrieve more detailed information. For this demo we are looking for people who write as a journalist or as a blogger. 

As you can see in the table below 674 Belgian Twitter Users descripe themselves as such, and at the same time the system also tells us what the main topics of their blog is all about.


If we would have been looking for a consultant, the system would give us a overview of different kinds of consultants.


<H3>Bringing in the meta-data<H3>

Since we are looking for bloggers and journalists living in the main Belgian cities we listed the Twitter users according to these criteria.



And when we ask for details about – for example – bloggers van de city Leuven, we get a list of the URL’s to their website where we might find more information about how to contact them.