Breast Cancer Prediction

Breast Cancer Prediction App (KNN-Based)

This project is a web application built with Streamlit that predicts whether a breast tumor is benign or malignant using a machine learning model trained on the Wisconsin Breast Cancer Dataset.

The backend model is an optimized K-Nearest Neighbors (KNN) classifier, tuned with GridSearchCV for best performance and saved using joblib for fast loading and deployment.

Key Features:

Interactive UI for manual data input or loading sample/test data

Predicts diagnosis with confidence score

Pie chart visualization of prediction probability

Highlights the top 5 features contributing to the prediction

Includes a feature explorer tab with medical explanations

Model and scaler are pre-trained and loaded automatically (optimized_knn_model.pkl, feature_scaler.pkl)