This repo contains tools to pull data from Twitter and analyse tweets according to the metadata contained with them. We have put together tools which have enabled a pipeline to execute NLP sentiment analysis on the data obtained Twitter, as well as providing visualization amongst out findings, this includes user group classification, topic modelling & wordclouds.
- Scraper:
- Scraper script is set up to automate the continuous pull of twitter data containing specified hashtags; limited to the last 7 days
- Scraper can be extended to pull tweets containing any tweet attribute
- User config file must be provided to authenticate against the Twitter API
- Cleaner:
- Clean twitter metadata
- FeX:
- Feature Extraction
- WordCloudGen:
- Word Cloud Generator
- Classifier:
- User Classification
- GeoTagger
- Used to determine geo-spatial sentiment analysis / region based