In this tutorial we present a method for topic modeling using text network analysis (TNA) and visualization. The approach we propose is based on identifying topical clusters in text based on co-occurrence of words. We will demonstrate how this approach can be used for topic modeling, how it compares to Latent Dirichlet Allocation (LDA), and how they can be used together to provide more relevant results.
Designing efficient content structures that guide readers through in ways that are efficient, maintaining diversity and narrative of the original.
In this article we demonstrate how you can visualize Google search results as a network of interrelated terms to identify the most relevant groups of keywords.
Mental disorders and various self-imposed psychological obstacles often arise because of the lack of choice. Gregory Bateson wrote about the double-bind origin of schizophrenic behavior where “crazy” reaction is simply a way of transcending a limiting situation, when the person sees no other choice. Various psychotherapeutic practices focus on expanding the range of possibilities for the patient, offering more choices than is currently available. Text network visualization can be a useful tool to aid this process. In this case study we are going to demonstrate how it works.
Every year in December Vladimir Putin addresses the Russian Federal Assembly. His address is considered to be a good indicator of the current political mood in the country. Putin’s 2014 speech is especially interesting, because it concerns not only Russia but the whole world. We at Nodus Labs together with Way to Russai analyzed Vladimir Putin’s address made on the 4th of December 2014 using text network analysis and below we present a report on what we discovered. We also made a comparison of the 2014 speech to the 6 speeches before from the years 2008-2013 to see how presidential rhetorics evolved over time.
