Data-Science-Project-Breast_Cancer_Classification_ML

Breast Cancer Classification Using Machine Learning 🎗️

This repository contains an ML-based breast cancer classification system using the Breast Cancer Wisconsin (Original) Dataset.
The project evaluates and compares six machine learning models to determine the most effective algorithm for early breast cancer detection.

---

📑 Table of Contents

• Introduction
• Dataset
• Machine Learning Models
• Data Preprocessing
• Model Evaluation
• Results and Performance Comparison
• How to Use This Repository
• Future Work
• License

---

📌 Introduction

Breast cancer is one of the most common causes of cancer-related deaths worldwide. Early detection significantly improves survival rates, but traditional diagnostic methods can be slow and error-prone.

This project explores machine learning algorithms to classify breast cancer tumors as benign (non-cancerous) or malignant (cancerous) using medical attributes.

📌 Why Machine Learning for Cancer Detection?

✅ Automates diagnosis for faster results
✅ Reduces human error in medical screening
✅ Assists doctors in clinical decision-making
✅ Can be deployed in hospitals and low-resource settings

---

📊 Dataset

📌 Breast Cancer Wisconsin (Original) Dataset

• 📂 Source: UCI Machine Learning Repository
• 🏥 Collected by: Dr. William H. Wolberg (University of Wisconsin Hospitals, 1992)
• 🔢 Total Instances: 699 cases
• ⚕️ Attributes:
  • Clump Thickness
  • Uniform Cell Size
  • Uniform Cell Shape
  • Marginal Adhesion
  • Bland Chromatin
  • Bare Nuclei
  • Normal Nucleoli
  • Mitoses
  • Single Epithelial Cell Size
• 🎯 Target Variable:
  • 2 = Benign
  • 4 = Malignant

---

🛠️ Machine Learning Models

This project evaluates the classification performance of six models:

Algorithm	Type	Key Strengths
Logistic Regression	Linear Model	Simple & interpretable
Decision Tree	Tree-Based	High interpretability
Random Forest	Ensemble	Reduces overfitting
Support Vector Machine	Kernel-Based	Effective for high-dimensional data
Artificial Neural Network	Deep Learning	Captures complex patterns
XGBoost	Boosting	Highly optimized, scalable

---

🔧 Data Preprocessing

Before training the models, we performed the following steps:

1️⃣ Handling Missing Values

• The Bare Nuclei column had missing values, which were replaced with the column median.

2️⃣ Binary Transformation of Target Variable

• The target variable was converted to 0 = Benign and 1 = Malignant for compatibility with ML models.

3️⃣ Data Splitting

• 80% for training and 20% for testing, maintaining class distribution (stratified sampling).

4️⃣ Feature Normalization

• Standardized data using StandardScaler (mean = 0, std = 1) to improve model performance.

---

📈 Model Evaluation

Each model was evaluated based on the following metrics:

✅ Accuracy: Measures overall correctness
✅ Precision: Proportion of correctly predicted malignant cases
✅ Recall (Sensitivity): Ability to detect malignant cases
✅ F1-Score: Balances precision and recall
✅ ROC-AUC Score: Measures classification capability

---

📊 Results and Performance Comparison

Here’s a summary of model performance:

Model	Accuracy (%)	Precision (%)	Recall (%)	F1-Score (%)
Logistic Regression	89.2%	87%	90%	88.5%
Decision Tree	92.1%	91%	93%	92%
Random Forest	95.7%	96%	94%	95%
Support Vector Machine	94.3%	95%	92%	93.5%
Artificial Neural Network	96.3%	97%	96%	96.5%
XGBoost	97.1%	98%	97%	97.5%

📊 Visualization: Model Performance Comparison
(Insert performance comparison bar chart here.)

---

⚙️ How to Use This Repository

1️⃣ Clone the Repository

git clone https://github.com/your_username/Breast_Cancer_Classification_ML.git
cd Breast_Cancer_Classification_ML