social-software-engineering

Live demo link: https://zachd.github.io/social-software-engineering/

Social Software Engineering project for CS7009 '17. This project crawls from a specified User or Organisation on GitHub, gets information about the associated repos, then crawls further to get the contributors to these repos.

This proof-of-concept aims to answer the question "Who is the most influential user in an organisation?". Test results are shown for Facebook, Airbnb and Netflix on the linked demo page. The most influential user can be inferred from the amount of relationships connecting them to other nodes.

A crawl is started by accessing the desired endpoint with an auth token, like /user/<name>/<token> or /org/<name>/<token>. Logs for crawling are shown in stdout. The crawler is limited to 25 results per Organisation/Repo/User for speed during large requests.

Elements

servant-crawler

Crawls GitHub for organisations, users and repos.

yesod-web

Frontend for obtaining Github auth and viewing results.

Requirements

• Haskell (GHCI)
• Neo4J

Running

• cd yesod-web; stack build; stack exec yesod-web
• cd servant-crawler; stack build; stack exec servant-crawler
• Visit localhost:3000 for yesod-web
• Visit localhost:8080 for servant-crawler