Analysing Russian Troll Tweets data with Python
summary
Analysis uses the FiveThirtyEight/Clemson dataset (about 2,98 million tweets from 2,848 handles) with Python for cleaning, UMAP clustering and time-series plots, producing clear groupings: right/left/newsfeed/hashtaggamer/fearmonger accounts and peaks in Oct 2016. Claim is coordinated, state‑sponsored agenda building that weaponizes social platforms; evidence includes language spread across 30+ languages, activity from Feb 2012 to May 2018, and geographic hits beyond the US (Canada, UK, Italy, Spain, Israel).
AI-written summary · about
chapters (16)
- 0:00Introduction
- 0:35Background
- 5:31Online trolls and bots
- 9:13Analyzing Tweets Datasets with Python
- 12:30Difference between troll account types
- 14:44Left Troll
- 16:21NewsFeed Troll
- 17:06HashtagGamer Troll
- 18:01Fearmonger Troll
- 19:19Clustering Troll Tweets with UMAP (Dimensionality reduction)
- 20:04Russian Trolls Activity over Time
- 26:20Russian Troll Activities beyond the U.S.
- 27:22Importance of problem research
- 30:15Challenges
- 34:20End of talk
- 34:50Questions
related talks
30mAnalyzing COVID-19 Protest Movements: A Multidimensional Approach Using Geo-Social Media Data
26mSocial Networks and Protest Participation: Evidence from 130 Million Twitter Users
17mWhat Moves Players? Visual Data Exploration of Twitter and Gameplay Data
01h 05mSocial Text, Sentiment, and Tone Analysis
41mDemystifying Social Network Analysis (SNA)
35mAnalyzing Public Conversation using LDA and Topic Modeling, in Python
49mWhy You Should Do Text Analysis in Python (Even if You Don't Want to)
29mInsights into Social Media Data using Entropy Theory