Analyzing Text Data

Topics

This week’s assignments will guide you through the following topics:

• Know the essential components of text analysis
• Analyze text on Wikipedia

Reading

Please read these two papers:

• Sergio Martinez-Ortuno, Deepak Menghani, and Lars Roemheld. “Sentiment as a Predictor of Wikipedia Editor Activity” [Link]
• Chan Young Park, Xinru Yan, Anjalie Field, and Yulia Tsvetkov. “Multilingual Contextual Affective Analysis of LGBT People Portrayals in Wikipedia” [Link]

Tasks

Complete the following tasks:

• Continue to compute/optimize the M-Statistic for all English Wikipedia Articles
• Compare the text generated by human vs. generated by bots
  • Pick some highly-edited pages
  • Consider the text generated by bots vs human (can consider only the parts of the text being changed)
  • Run topic model (ex. LDA) OR sentiment analysis on two sets of edits
  • Report what you think you find
• You might find these links helpful:
  • Topic Modeling in Python: Latent Dirichlet Allocation (LDA)
  • A Beginner’s Guide to Sentiment Analysis with Python

Weekly Questions

Answer the standard participation questions