FRANKFRANK5/FRAUD-DETECTION-EAST-AFRICA

🛡️ FRAUD_DETECTION_IN_EAST_AFRICA — Real-time fraud detection and risk scoring engine for East Africa's short-rental market. Built for Pumzika Hackathon 2026 (Track #04). # 🛡️ Fraud Detection & Trust Scoring System ## East African Rental Market Intelligence Platform ## ⚡ Executive Summary A real-time fraud detection and trust scoring system for East African rental listings that evaluates risk using multi-factor analysis to prevent rental scams before payment occurs. ## 🧾 Submission Info - Participant: Frank Karani - Country: Tanzania - Challenge: #04 Fraud Detection & Trust Scoring - Institution: Institute of Accountancy Arusha (IAA) - Year: First Year (2025/2026) --- ## 👨‍💻 Author **Frank Karani** Cyber Security Student | Institute of Accountancy Arusha (IAA) 🔗 LinkedIn: Frank Karani **GitHub:** github.com 🚀 Live Demo: Fraud Detection & Trust Scoring System ## 📌 Overview This system detects fraudulent rental listings in the East African market (Tanzania, Kenya, Uganda) using a multi-factor risk analysis engine. It is designed to help users identify suspicious listings before making financial commitments, improving trust and safety in digital property marketplaces. 🎯 Live Demo & System Validation The system has been tested with real-world scenarios from the East African rental market. Below is visual proof of the Fraud Detection Engine in action, demonstrating both approved listings and blocked fraudulent attempts. ### 1. User Input Interface The system provides a clean, intuitive form for landlords to submit property details for real-time risk analysis. All critical data points including price, location, images, and user history are captured for the ML model. ### 2. Legitimate Listing Approved - Trust Score: 100% This demonstrates the system's ability to accurately identify and approve legitimate properties. The ML model found no suspicious patterns, resulting in a Fraud Score: 0% and Risk Level: LOW. The listing is automatically approved and published. ### 3. Fraudulent Listing Blocked - Price Anomaly Detected This shows the fraud detection engine successfully identi …