Breast Cancer Detection

Overview

Breast Cancer is a leading form of cancer in woman often going undetected to advanced stages, thus a detection model is built using 32 feautures. It aims to efficiently detect Breast Cancer using various ML models and demonstrates basic concepts for Machine Learning

Feautures

• Pandas & numpy library of Python is used for basic mathematical operations on Dataset
• Sklearn is used for pre-processing data using One Hot Encoding, Variance Threshold & CHI sq
• Cross Validation is performed during train-test split
• Sklearn is further used for deploying Linear & Logistic Regression, Decision Tree, Random Forest, Support Vector Machine & K-nearest neighbour
• Tensorflow is used for Artificial Neural Networks
• Matplotlib & Seaborn is used to visualise models and their Confusion Matrix for all algorithms used.

Tech Stack

Setup & Installation

• Clone the repo in local Github.
• Use Google Colab or Install Jupyter
• Run the .ipnyb file and ensure to give correct path for CSV
• Now download dataset from this repository and upload it to Google Colaboratory or set appropriate path in Jupyter
• If using Jupyter/Anaconda download dataset from: https://www.kaggle.com/datasets/uciml/breast-cancer-wisconsin-data
• Install all libraries in terminal/Command Prompt (using cmd command on Windows) if using Jupyter/Anaconda

  pip install numpy
  

   pip install pandas
  

  pip install matplotlib
  

  pip install seaborn
  

  pip install tensorflow
  

  pip install sklearn
  

• Ensure dataset is present in same directory and specify the correct path
• For Google Colab, libraries are downloaded and imported during runtime
• Run the .ipnyb file

Accuracy

• Linear Regression: 78.2%
• Logistic Regression: 98.8%
• K-nearest neighbour: 95.9%
• Decision Tree: 95.34%
• Random Forest: 95.1%
• Artificial Neural Networks (ANN) : 96.49%

License

This project is licensed under License

If you find my repository helpful, please star⭐ it 🌟.