Fraud Detection App

Description

The Fraud Detection App is a web-based RAG (Retrieval-augmented generation) application designed to identify and classify fraudulent messages. The application uses a combination of machine learning models and keyword-based analysis to detect and flag messages related to phishing and financial fraud.

Features

• Fraudulent Message Detection: Classify messages as 'fraudulent' or 'legitimate'.
• Phishing and Financial Fraud Identification: The app identifies fraud through keyword matching and uses Gemini API for more advanced analysis.
• User-friendly Interface: Provides an easy-to-navigate web interface for users to report and check fraud.

Tech Stack

• Frontend: React, Material UI
• Backend: FastAPI
• Machine Learning: Gemini API, Cosine Similarity and Vector Search
• Database: MongoDB Atlas (for storing messages and related data)