hackferenceindia / submissions

Hackference Hackathon Submissions
7 stars 14 forks source link

deadlock - GitLit #28

Open vchrombie opened 5 years ago

vchrombie commented 5 years ago

DEADLOCK - GITLIT

ℹ️ Project information

🔥 Your Pitch

Open Source is, by large, hard to adjust into. Whenever one enters the world of Open Source, he/she is, in most of the cases, greeted by large organizations where new contributors may find difficult to contribute something worthy, it can be their lack of patience or simply the lack of enough skills or it might be what we think it is - simply the lack of finding the correct projects.

GitLit is a web app which matches projects available on GitHub by rating the user by analyzing the skills and interests and then recommend projects for them, after analyzing the projects by activity, complexity, etc to get you the perfect project to contribute to which is available for your skills.

Furthermore, even for the contributors, discovering interesting and contributable communities, and then to find workable issues and cool features to take up, is, in fact, a complicated and time-taking task, often leading many beginners to drop out of Open Source as they don't know where and how to get started.

GitLit is a web app that rates repositories and contributors through a logical algorithm which takes into consideration not only a few standard metrics like Stars, Forks, and PRs, but also accounting for factors like Activity in the Community (eg: Issue Resolution, Feature Addition), backgrounds of the contributors etc.

Target Customers

General Metrics

User Analysis:

Repo Analysis:

🔦 Any other specific thing you want to highlight?

We plan to scale up by adding more advanced metrics by using dedicated processing by Google Big Query since we are largely dependent on the API's of GitHub, there are large limitations for now.

We also have started working on an ML algorithm which will efficiently match the developers with the most relevant projects, backing on our metrics.

Further, we would like to integrate NLP for identifying the quality of code, and community engagement.

✅ Checklist

Before you post the issue: