Tools based on visual representations of social networks are important to understand network data and convey the result of the analysis. Visualization facilitates qualitative interpretation of network data. This kind of tools helps us to identify key influencers, contents, and to focalize attention on interesting users optimizing our time.

Many experiments having to do with analysing tweets can be found on Kaggle and other websites that take on similar topics. I have tried to aggregate the results of these experiments in a simple web application (Exploring Tweets) that shows how different views of the data can be unified in order to get an overall picture that points out interesting connections at a glance.

screenshot-tweets-explorer

I have also added an interactive map (using D3.js) that is able to highlight relationships among users. By clicking on a node on the map, you can highlight that node with its connections, placing all of the others in the background. The edge thickness depends on how many communications have been made. The node colors represent three user categories (green: only senders, orange: only receivers, blue: senders and receivers).

You can try this web application by following the link: Exploring Tweets. It is based on an interesting public dataset with 17K+ tweets from 100+ users that is widely used on Kaggle. I got this dataset in June of 2016 so it could be not so updated but this is not the objective.

This web application is just an exercise and is an harmonized mixture of many different algorithms. From the point of view of the execution time, the most demanding algorithm is the Latent Dirichlet allocation (LDA) used to extract the top 6 topics/concepts.

Enjoy and let me know what you would add.

Posted by lorenzo

Full-time engineer. I like to write about data science and artificial intelligence.

Vuoi commentare?