Breast Cancer Predictor is an application that combines Machine Learning and Web Development technologies. The project focuses on detecting breast cancer using Support Vector Machines (SVM).
🔗 Live Demo: https://breastcancerpredictor.vercel.app/
- Goal: To predict whether breast cancer tumors are benign or malignant.
- Machine Learning:
- Dataset: Sklearn Breast Cancer Dataset.
- Model: SVC (Support Vector Classifier) with an linear kernel.
- Technologies Used:
- Backend: Flask API and ASP.NET Core API.
- Frontend: React (TypeScript) + TailwindCSS.
- Database: PostgreSQL.
- Docker: Used to containerize all components.
- Dataset: Sklearn Breast Cancer Dataset.
- Libraries:
- Scikit-learn (SVC model).
- Flask (API).
- Pandas, NumPy.
- Joblib (Model saving/loading).
- JWT Authentication: Ensures secure access to API endpoints.
- Layered Architecture:
- Controller → Service → Repository.
- Database: PostgreSQL.
- Tech Stack: React + TypeScript + Vite.
- Styling: TailwindCSS.
- Pre-Built Components: Shadcn/UI.
- Docker Compose orchestrates all services (Flask API, .NET API, PostgreSQL) as containers.
- Prediction API:
- A
/predict
endpoint exposed via Flask. - Accepts 30 features to predict tumor type (Benign or Malignant).
- Previous predicts for users.
- A
- Secure JWT Authentication:
- Ensures secure user authentication and authorization.
- Hashed password and encrypted email for database.
- User-Friendly React Interface:
- Clean and modern UI design.
- CI/CD Deployment:
- Deployed using Vercel with GitHub Actions.
git clone https://github.com/ykdid/BreastCancerPredictor.git
cd BreastCancerPredictor
docker compose up --build
cd react-ts-vite
pnpm install
pnpm run dev
https://breastcancerpredictor.vercel.app/
Yusuf Kaya
📧 Email: yusufkaya.yjk@gmail.com