COVID-19 Political Analysis from 387K Scrapped Twitter
Overview
We developed a comprehensive analysis system to track and analyze the spread of COVID-19 related information across social media platforms. Our team processed 387,000 tweets, analyzed 25 news portals, and tracked 35 trusted health officials to understand how misinformation impacts public sentiment during the pandemic. The project aimed to provide data-driven insights for policymakers and raise awareness about information consumption patterns.
Technical Details
We implemented a sophisticated data pipeline that collected and analyzed social media data using Twitter’s API. Our system processed tweets with geolocation data to understand regional patterns of information spread. We developed custom NLP models for sentiment analysis and bias detection, categorizing content sources into political alignments (Left, Right, Center) and credibility levels (High, Mixed, Low). The analysis included tracking conspiracy theories, measuring public reaction to key events, and quantifying the impact of influential personalities.
Implementation
Using Python and advanced NLP libraries, we created a robust analysis framework that processed massive amounts of social media data. Our implementation included automated tweet scraping, sentiment scoring, political bias classification, and conspiracy theory detection. We developed interactive visualizations using Plotly to represent complex patterns in information spread and public sentiment changes over time. The system tracked specific events like Trump’s mask-wearing stance and various COVID-19 related conspiracy theories.
Key Achievements
Our analysis revealed significant patterns in how different political segments influence public opinion. We successfully mapped the spread of both factual information and misinformation across different demographic segments. The project demonstrated how media bias affects public sentiment during a health crisis, with quantifiable metrics showing the impact of different information sources on public perception.
Impact
This research provides valuable insights for policymakers and health officials in understanding how information spreads during public health crises. Our findings help identify effective strategies for combating misinformation and highlight the importance of responsible information sharing by influential personalities and media outlets. The project contributes to the broader discussion of managing “infodemics” during global health emergencies.
Future Directions
We plan to enhance the system with real-time analysis capabilities, develop more sophisticated bias detection algorithms, and create an early warning system for emerging misinformation trends. Future work will focus on expanding the analysis to other social media platforms and developing automated fact-checking mechanisms.
Project Links Live Demo: Project Website Source Code: GitHub Repository