Projects
Built this: Social media sentiment and trend analytics on MongoDB
A Big Data Analytics course project: 2,944,814 real tweets in MongoDB, cleaned and enriched with sentiment, topics and keywords, served by FastAPI to a React dashboard, with Spark and sharding experiments.
- The README is upfront about the data: real tweets, but all from state-linked troll accounts, with synthetic engagement counts that are clearly flagged.
- Experiments measure ingestion, indexing, aggregation and NLP at 100K, 500K, 1M and 2.95M documents.