TasteEngine / README.md
Abdallah4z's picture
Production readiness: port 7860, Dockerfile, HF Spaces config
f9ade9b
metadata
title: TasteEngine
emoji: 🎯
colorFrom: indigo
colorTo: pink
sdk: docker
app_port: 7860
pinned: false
license: mit

TasteEngine — Multi-Approach Recommender System

An interactive recommender system comparing Collaborative Filtering, Content-Based, and Knowledge-Based approaches side by side.

Features

  • 5 Collaborative Filtering methods: User-Based, Item-Based, SVD, KNN, Slope One
  • 2 Content-Based methods: TF-IDF Similarity, Feature Matching
  • 3 Knowledge-Based methods: Constraint, Rule, Utility
  • Evaluation Dashboard: Compare all methods across 6 metrics (RMSE, MAE, Precision@k, Recall@k, F1@k, Coverage)
  • Interactive UI: Built with Alpine.js, HTMX, and Chart.js
  • Incremental Evaluation: Results appear as each method finishes
  • Dark/Light Mode: Toggleable theme

Tech Stack

  • Backend: Python, Flask, NumPy, Pandas, scikit-learn
  • Frontend: Alpine.js, HTMX, Chart.js, Lucide icons
  • Deployment: Docker on Hugging Face Spaces