Welcome to the AI-Driven Fashion Shopping Platform project! This is an innovative fashion e-commerce site that harnesses the power of Artificial Intelligence (AI) to provide a personalized, smart, and intuitive shopping experience. Please use this URL to checkout out website https://smartwardrobe.store/
This platform revolutionizes online shopping by allowing users to:
- Search for products using natural language queries like "Show me eco-friendly, sleeveless jumpsuits for a summer festival under 75 euros, with reviews mentioning comfort and durability."
- Upload images to find visually similar fashion items.
- Get personalized recommendations based on user behavior, previous purchases, and preferences.
- Enjoy AI-powered outfit matching and automated review summaries to enhance the shopping experience.
The project focuses on delivering a fast, seamless, and enjoyable experience for fashion lovers, transforming the way users discover and purchase their favorite items.
- Natural Language Queries: Search using phrases like "Find casual jackets for winter." or "Show me trendy summer shoes."
- Visual Search: Upload images to discover visually similar items using AI-powered image recognition.
- Content-Based Filtering: Suggest products based on attributes like color, style, or brand.
- Collaborative Filtering: Recommend items based on the preferences and behavior of similar users.
- AI-powered suggestions for complementary items to complete a look.
- NLP-driven sentiment analysis summarizes customer reviews into easy-to-read highlights, e.g., "Highly rated for comfort."
- Kaggle H&M Personalized Fashion Recommendations Dataset
This dataset includes transaction logs, product metadata, and customer interaction data, all crucial for personalized recommendations and insights. - External Fashion Trends & Metadata: Additional fashion trend data, seasonality, and social signals.
Smartwardrobe-backend is an advanced e-commerce backend repository built with clean code architecture using the Nest.js framework. It is designed to handle a wide range of features for modern e-commerce platforms. The backend leverages AWS RDS PostgreSQL as its database.
To set up the project, create a .env file under the src folder and populate it with relevant configurations from the env/sample.env file. Replace the placeholders with your actual credentials and settings.
Key Features:
- User Authentication and Management:
- Secure user authentication using Passport.js.
- Password hashing implemented with Bcrypt for added security.
- Product Management:
- CRUD operations for products, categories, and subcategories.
- Cart Functionalities:
- Add, update, and manage cart items.
- Collaborative Chat Room:
- Support for user-to-user and group messaging.
- Attachment handling and message typing.
- AI-Powered Similar Items API:
- Integration with OpenAI CLIP Model for finding visually and contextually similar products.
- Uses Faiss Search library to recommend similar products.
- Friendship and Social Features:
- Friend requests, friendship management, and group creation.
- Notifications for real-time updates.
The database schema supports a variety of e-commerce modules, including users, products, categories, inventory, carts, chats, social features, and more. Below is a high-level depiction:
$ npm install
# development
$ npm run start
# watch mode
$ npm run start:debug
# production mode
$ npm run start:prod
Modules Overview
User Module
- Handles user registration, login, and profile management.
- Includes password reset and secure token generation. Product Module
- Supports hierarchical categorization with categories and subcategories.
- Real-time inventory management through triggers and procedures. Cart
- Seamless cart operations for adding/removing items and tracking sizes.
- Chat and Notifications
- Real-time group and individual chats.
- In-app and email notifications for friend requests, updates, and more.
- Advanced Similar Items
- FAISS-based search implementation in Nest.js for fast similar item recommendations.
This backend is a robust solution for modern e-commerce platforms, blending scalable architecture with innovative features.