Breast Cancer Wisconsin (Diagnostic) Classification Project

This project uses the Breast Cancer Wisconsin (Diagnostic) dataset from the UCI Machine Learning Repository to build and evaluate machine learning models and a neural network for the classification of breast cancer as either malignant or benign.

Dataset Overview

• Name: Breast Cancer Wisconsin (Diagnostic)
• UCI Repository ID: 17
• Number of Instances: 569
• Number of Features: 30 (excluding the ID and target variables)
• Target Variable: Diagnosis (M = Malignant, B = Benign)
• Feature Types: Real-valued features computed from digitized images of breast mass cell nuclei

Project Structure

Data Preprocessing

• Scaling: Features are scaled using StandardScaler.
• Target Encoding: The target variable (Diagnosis) is label-encoded, converting 'M' to 1 (Malignant) and 'B' to 0 (Benign).

Model Evaluation

1. Machine Learning Models:

The following models are evaluated for classification:

• Logistic Regression
• K-Nearest Neighbors (KNN)
• Support Vector Machine (SVM)
• Decision Tree
• Random Forest

Evaluation Metrics:
The models are evaluated based on the following metrics:

• Accuracy
• Confusion Matrix
• Classification Report (Precision, Recall, F1-Score)

Results Summary

Model	Accuracy	Precision	Recall	F1-Score
Logistic Regression	1.0000	1.00	1.00	1.00
K-Nearest Neighbors	0.9737	0.97	0.97	0.97
Support Vector Machine	1.0000	1.00	1.00	1.00
Decision Tree	0.9912	0.99	0.99	0.99
Random Forest	1.0000	1.00	1.00	1.00

Best Model:

• Logistic Regression, Support Vector Machine and Random Forest achieved perfect accuracy (1.0000) on the test set.

2. Neural Network Model:

A neural network model is built for classification, with the following architecture:

• Input Layer: 128 neurons, Tanh activation
• Hidden Layers: 64 neurons, Tanh activation; 16 neurons, Tanh activation
• Output Layer: 1 neuron, Sigmoid activation

The model is trained using:

• Early stopping to prevent overfitting
• Learning rate reduction based on validation loss
• Epochs: 8
• Batch Size: 16

Evaluation Metrics:

• Validation Loss: 0.0297

• Validation Accuracy: 1.0000

• Training Loss: 0.0538

• Training Accuracy: 0.9890

Dependencies

• Python 3.x
• Pandas
• NumPy
• Matplotlib
• Seaborn
• Scikit-learn
• TensorFlow
• UCIMLRepo

Installation

To run the project, ensure that you have the necessary dependencies installed. You can install them using pip:

pip install pandas numpy matplotlib seaborn scikit-learn tensorflow ucimlrepo