This project is being open sourced while we work on other projects.
Project owners:
@shanahanjrs
@tysongg
@aglovaile
analyzer/
Sentiment Analysis Engine
scraper/
Scrapes Reddit, Twitter, and News for specific search terms
analysis_utils/
Contains general-use util functionality for all internal microservices
lavadash-ui/
Contains all the Django code for the Lavadash website
ci/
Continuous Integration
Build everything
./build.sh
Build a single service
docker-compose build <service name>
./run.sh
- Home page: http://localhost:8000/
- Chronograf: http://localhost:8888/sources/0/admin-influxdb/databases
- Grafana: http://localhost:3000/?orgId=1
- Add a search term if Redis is empty on start: http://127.0.0.1:5100/search/terms/military
- Every N second we pull all the terms we want to scrape from Redis (Redis.Set.scraperTerms) and Publish them to a pub/sub queue (toScrape) for the scraper service.
- TODO: Add the SPAM FILTERING step here...
- When the scraper service picks up terms from the toScrape queue we will scrape Twitter, Reddit, and top news headlines then Publish the results in a pub/sub queue (toAnalyze) for the analysis service.
- When the analysis service picks up scrape-results from the toAnalyze queue it will perform the standard Sentiment Analysis + others (in the future) and store the results in a Time Series DB.