Breast Cancer Detection

Overview

This Jupyter Notebook file contains code for detecting breast cancer using machine learning algorithms. The dataset used in this project is data.csv.

Data Preprocessing

Import Libraries and Dataset

• Necessary libraries like NumPy, Matplotlib, Pandas, Seaborn, and scikit-learn are imported.
• The dataset is imported using Pandas' read_csv function.

Data Exploration

• The shape and information of the dataset are examined.
• Categorical and numerical columns are identified.
• Statistical summary of the dataset is generated.

Dealing with Missing Values

• Missing values are checked and dropped from the dataset.

Dealing with Categorical Data

• Categorical columns are identified and converted into dummy variables.
• Boolean values are replaced with 0s and 1s.

Data Visualization

• Countplot is used to visualize the distribution of classes (Benign and Malignant).
• Correlation matrix and heatmap are plotted to explore relationships between variables.

Model Building

Two models are built:

1. Logistic Regression
2. Random Forest Classifier

Logistic Regression

• The model is trained and evaluated.
• Cross-validation is performed to assess model performance.
• Randomized search is conducted to find the best parameters.
• Final model with optimized parameters is trained and evaluated.
• Prediction function is defined to predict the class of new data.

Random Forest Classifier

• The model is trained and evaluated.
• Cross-validation is performed to assess model performance.

Prediction

• A sample observation is provided to test the prediction function.