Art Emotion Recommender

An AI-powered web application that suggests artworks based on emotions and situations using LangChain and OpenAI. The system provides personalized art recommendations that resonate with your current emotional state.

Overview

This project creates an intelligent art recommendation system that:

• Takes user input about their emotional state or situation
• Analyzes the emotional context using AI
• Recommends relevant artworks from a curated collection
• Provides explanations for why each artwork matches their current state

Features

• 🎨 AI-powered art recommendations based on emotional input
• 🤖 LangChain + OpenAI for natural language processing
• ⚡ Real-time recommendations with loading states
• 🌐 Modern tech stack: Next.js + FastAPI
• ♿ Accessibility with ARIA labels
• 🎯 Responsive, minimalist design

Tech Stack

Frontend

• Next.js 14 (App Router)
• TypeScript
• Tailwind CSS
• Loading skeletons for better UX

Backend

• FastAPI + Uvicorn
• LangChain for AI orchestration
• OpenAI API integration
• Python dotenv for configuration

How It Works

1. Input Processing: Enter your emotional state or situation (e.g., "I'm feeling peaceful", "I had a stressful day")

2. Emotional Analysis: Using LangChain and OpenAI, the system analyzes the input to identify core emotions and context.

3. Art Matching: The system searches through a curated collection of artworks, each tagged with emotional attributes.

4. Recommendation Generation: You'll receive up to 3 artworks that either:

   • Match and validate your current emotional state
   • Offer a contrasting perspective that might help shift your mood

5. Explanation: Each recommendation includes a detailed explanation of why the artwork was chosen.

Art Collection

The system includes a curated collection of famous artworks with emotional attributes, including:

• The Starry Night (Vincent van Gogh)
• Water Lilies (Claude Monet)
• The Scream (Edvard Munch)
• The Persistence of Memory (Salvador Dalí)
• Sunflowers (Vincent van Gogh)
• And more...