Predicting Malignant Breast Cancer: A Data Science Project Analysis

This project focuses on predicting malignant breast cancer using a dataset of 569 cases with 30 continuous variables. The data was split into training and testing sets. Descriptive statistics, including tables and plots, were generated to analyze variables like radius, texture, perimeter, and compactness. Hypothesis testing was conducted, revealing significant differences in texture and compactness between benign and malignant cases. A logistic regression model was developed using radius mean, worst concave points, and worst area as key predictors, achieving high significance. The model's performance was evaluated using an ROC curve. This project demonstrates the application of statistical techniques and data analysis in predicting breast cancer diagnosis, offering insights into key variables for differentiating between benign and malignant cases. The analysis was conducted using R programming.