This is the solution/ script for the Findefender Fraud Detection System. It is implemented above the Open Finance platform so as to allow users to report potential fraud by submitting images and descriptions. The system processes these reports, extracts text from images, and analyzes the fraud percentage.
The User Portal allows individuals to report potential fraud by submitting images and descriptions. The system:
- Extracts text from images.
- Compares the description text with the extracted image text.
- Calculates fraud percentages and provides a summary.
If the fraud percentage exceeds a threshold (70%), the report is flagged as fraud.
- Frontend: NextJS
- Backend (optional):
- Database: Supabase
- Others: OpenAI, Gemini
- Upload images and add descriptions to reports.
- Receive fraud analysis with percentage results.
- View alerts based on fraud percentage.
Click above to watch a demo showcasing the FinDefender Solution in action.
- Clone the repository:
git clone https://github.com/SiowYenChong/FinDefender.git