https://github.com/magnexis/feed-ranking-algorithm
This is a simple backend feed-ranking prototype I built as one of my first learning projects. It uses basic engagement signals, session awareness, and topic rotation, exposed through a small FastAPI API. There’s no frontend included—feel free to use, modify, or build on it however you like.
https://github.com/magnexis/feed-ranking-algorithm
algorithms fastapi python
Last synced: 4 days ago
JSON representation
This is a simple backend feed-ranking prototype I built as one of my first learning projects. It uses basic engagement signals, session awareness, and topic rotation, exposed through a small FastAPI API. There’s no frontend included—feel free to use, modify, or build on it however you like.
- Host: GitHub
- URL: https://github.com/magnexis/feed-ranking-algorithm
- Owner: magnexis
- Created: 2026-01-20T23:19:32.000Z (5 months ago)
- Default Branch: main
- Last Pushed: 2026-01-20T23:28:48.000Z (5 months ago)
- Last Synced: 2026-06-13T00:23:16.375Z (17 days ago)
- Topics: algorithms, fastapi, python
- Language: Python
- Homepage:
- Size: 5.86 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
\# Feed Ranking Algorithm Prototype
This project is an educational prototype of a social-media-style feed ranking system.
\## Features
\- Engagement-based ranking
\- Session-level signals (fatigue detection)
\- Topic rotation to prevent repetition
\- Event-driven interaction modeling
\- FastAPI web API interface
\## Tech Stack
\- Python 3
\- FastAPI
\- Uvicorn
\## How It Works
1\. User interaction events are loaded from JSON
2\. A session object is constructed
3\. Content is ranked using multiple weighted signals
4\. Topic repetition is softly penalized
5\. Results are exposed via a REST API
\## Run Locally
```bash
pip install -r requirements.txt
uvicorn api:app --reload
Visit:
http://127.0.0.1:8000/feed
http://127.0.0.1:8000/docs