How are tweets analysed in Social?

In this article, you will learn how tweets are tagged with topics in order to generate data in the Social and Materiality modules.

Datamaran uses Natural Language Processing (NLP) to analyze the text of tweets. The tool analyzes the latest 24 weeks of data.

Our NLP algorithms tag tweets with social topics to generate three different metrics:

  • Total tweets - the number of tweets, including retweets (i.e. a re-posting of a tweet) that mention one or more topics.
  • Unique tweets - the number of unique tweets, excluding retweets, that mention one or more topics.
  • Total contributors - the number of accounts that mention one or more topics, including retweets.

For each topic, our data scientists create a list of inclusive and exclusive terms, and rules that will allow our AI to determine if a tweet mentions at least one of the topics we track.