VeriCart AI

AI-Powered E-Commerce Trust Platform

VeriCart AI is an intelligent e-commerce platform that leverages multiple LLM providers (DeepSeek, Qwen, Hunyuan) to deliver AI-driven product reviews analysis, authenticity verification, and smart shopping recommendations — building trust between buyers and sellers.

---

Tech Stack

Layer	Technology
Backend	Spring Boot 3.2.0, Java 17, MyBatis 3.0.3, Spring Security + JWT
Frontend	Vue 3.4, Vite 5, Pinia, Vue Router, Element Plus, Axios
Database	MySQL 8.0 (production), H2 (testing)
AI	DeepSeek (deepseek-chat), Qwen (qwen-plus), Hunyuan (hunyuan-lite)
Infra	Docker Compose

---

Project Structure

VeriCart AI/
├── backend/                  # Spring Boot backend
│   ├── src/main/java/com/vericart/
│   │   ├── config/           # Security, AI, CORS configuration
│   │   ├── controller/       # REST API controllers
│   │   ├── dto/              # Data transfer objects
│   │   ├── entity/           # MyBatis entity classes
│   │   ├── mapper/           # MyBatis mapper interfaces
│   │   ├── service/          # Business logic & AI gateway
│   │   └── util/             # Utility classes
│   ├── src/main/resources/
│   │   ├── mapper/           # MyBatis XML SQL mappings
│   │   ├── schema.sql        # Database DDL (10 tables)
│   │   └── application.yml   # Main configuration
│   └── src/test/             # JUnit tests (H2 in-memory)
├── frontend/                 # Vue 3 SPA frontend
│   ├── src/
│   │   ├── components/       # Reusable Vue components
│   │   ├── views/            # Page-level views
│   │   ├── router/           # Vue Router config
│   │   ├── stores/           # Pinia state stores
│   │   └── api/              # Axios API clients
│   └── vite.config.js
├── docs/                     # Project documentation & reports
├── PRD/                      # Product requirements documents
├── docker-compose.yml        # Full-stack Docker deployment
└── README.md

---

Quick Start

Prerequisites

• Java 17 + Maven 3.8+
• Node.js 18+ + npm
• MySQL 8.0 (local or Docker)
• Docker & Docker Compose (optional)

1. Clone & Environment

git clone https://github.com/simul49/VeriCart-AI.git
cd VeriCart-AI

Set your API keys and database password via environment variables (or create a .env file):

export DB_PASSWORD=your_mysql_password
export DEEPSEEK_API_KEY=sk-xxxxxxxxxxxxxxxx
export QWEN_API_KEY=sk-xxxxxxxxxxxxxxxx
export HUNYUAN_API_KEY=sk-xxxxxxxxxxxxxxxx
export JWT_SECRET=your_jwt_secret

2. Start with Docker (Recommended)

docker-compose up -d

This starts MySQL, the Spring Boot backend (port 8080), and the Vue frontend (port 80) in one command.

3. Manual Start

Database — Create the MySQL database:

CREATE DATABASE IF NOT EXISTS vericart
  CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;

Tables are auto-created by schema.sql on startup.

Backend (port 8080):

cd backend
mvn spring-boot:run

Frontend (port 5173):

cd frontend
npm install
npm run dev

Open http://localhost:5173 in your browser.

---

Database

10 MySQL tables managing the full e-commerce trust lifecycle:

Table	Purpose
user	User accounts (buyers & sellers)
product	Product listings
category	Product categories
cart_item	Shopping cart items
wishlist	User wishlists
orders	Purchase orders
order_item	Line items within orders
review	Product reviews & ratings
ai_analysis_log	AI review analysis results
notification	System notifications

Testing uses H2 in-memory with MySQL compatibility mode (MODE=MySQL).

---

API Endpoints

Base URL: http://localhost:8080/api

Method	Endpoint	Description
POST	/auth/register	User registration
POST	/auth/login	User login (JWT)
GET	/products	List products
GET	/products/{id}	Product detail
GET	/products/{id}/reviews	Product reviews
POST	/reviews/analyze	AI review analysis
GET	/cart	View cart
POST	/cart/add	Add to cart
GET	/wishlist	View wishlist
POST	/orders	Place order
GET	/orders/{id}	Order detail

---

AI Features

The platform integrates 3 LLM providers through a unified gateway:

• DeepSeek — Core engine for review sentiment analysis and authenticity scoring
• Qwen (Alibaba) — Product description generation and multilingual support
• Hunyuan (Tencent) — Lightweight fallback for basic analysis tasks

Each provider is configured in application.yml with environment-variable API keys.

---

Running Tests

Backend (JUnit + H2 in-memory):

cd backend
mvn test

Frontend (Vitest):

cd frontend
npm test

---

License

This project is developed for academic/learning purposes as part of an internship training program.