Sho Da Igram 🎮

"What game should I play?" - A work-in-progress game recommendation system

Ever spend more time browsing game libraries than actually playing? This project aims to fix that by building a smart recommendation engine that actually understands what you like.

🎯 Features

• Similar Games Discovery: Find games similar to your favorites using TF-IDF content-based recommendations
• Natural Language Search: Search for games by describing what you want (e.g., "space shooter with crafting")
• Tag-Based Filtering: Filter games by genres, themes, platforms, and more
• Explainable Recommendations: See why games were recommended with match reason badges

🏗️ Architecture

Backend

• Framework: Spring Boot 3.5.6 (Kotlin)
• Database: PostgreSQL with full-text search
• Algorithms: TF-IDF (Apache Lucene), BM25 (PostgreSQL)
• Data: ~7,500 games from IGDB and RAWG APIs

Frontend

• Framework: React 18 + TypeScript + Vite
• UI Library: Material-UI (MUI) v5
• State Management: TanStack Query (React Query)
• Styling: Styled Components + MUI theming

Data Engineering

• Language: Python 3.12
• APIs: IGDB, RAWG
• Processing: ETL pipeline for game data ingestion

🚀 Getting Started

Prerequisites

• Java 21
• Node.js 20+
• PostgreSQL 15+
• Yarn
• Python 3.12 (for data engineering)

Backend Setup

cd backend
./gradlew bootRun

API runs on http://localhost:8080

Frontend Setup

cd frontend
yarn install
yarn dev

App runs on http://localhost:5173

Data Engineering

cd data-engineering
uv sync
make fetch  # Fetch game data from APIs

📁 Project Structure

ShoDaIgram/
├── backend/           # Spring Boot API
├── frontend/          # React TypeScript app
└── data-engineering/  # Python ETL scripts