Twitter Data-Analyse: What can I do with the data?

I retrieve data to a specific topic from Twitter and did my sentiment analysis on it. I never did anything in NLP, etc. So what else can I do with that? Main goal would be to find out if the Twitter community is against this topic or not.

I am also struggling with cleaning the data and I mean by that, that I am unsure how much should I clean on that Tweet.

I would be also glad to get any advise on books, articles, communities, videos...

Topic data-science-model twitter topic-model nlp data-cleaning

Category Data Science


Community analysis implies graph analysis.

here is a short list of things you can work on:

  1. People often reshares tweets among a certain social group. Minimum-cut method, Girvan–Newman and Modularity maximization are someof the starting algorithms to extract these type of substructures.
  2. You can try and find different hierarchies among the groups sharing a particular topics
  3. You can try and analyse the lifetime of tweets for particular topics (survival analysis)

Analysing tweets is closer to graph analytics rather than NLP. Here is a great overview on community analysis. For coding and algorithms, please check graphX Spark library. If your data is not too large, networkX is easier. For survival analysis, lifeline is one of the easier options.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.