Breast Cancer Classification Project Using Machine Learning

I am pleased to share with you the Breast Cancer Classification Project using Machine Learning, where a set of models and techniques were used to analyze medical data and classify the tumor into malignant or benign based on a set of clinical characteristics.

Models used:

Several machine learning models were applied and compared to ensure the highest possible accuracy, including:

Logistic Regression: A simple and fast model based on linearity in class separation.

Support Vector Machine (SVC): Tested with multiple types of kernels (linear, RBF, and polynomial) to improve performance.

Random Forest: Used to build multiple decision trees and combine their results to reduce errors.

Gradual Boosting: A robust model that gradually improves performance.

Kest Neighbors (KNN): A model based on the geographical proximity of samples to classify them.

XGBoost: An advanced model that enhances performance quickly and accurately, especially with big data.

Libraries used:

Several libraries were relied upon to develop the project and achieve its goals, including:

NumPy and Pandas: For data analysis and processing.

Matplotlib and Seaborn: To create graphs and visualize data.

Skit-learn: To apply models, measure data, and use grid search techniques such as GridSearchCV and RandomizedSearchCV.

Imbalanced-learn: To handle data imbalance using SMOTE.

XGBoost: To apply advanced boosting in classification.

Data Analysis and Graphical Visualizations:

Exploratory Data Analysis (EDA) was used to clarify important relationships and features. The analyses included:

Distribution plotting using Seaborn to analyze the distribution of independent features.

Correlation matrix to identify relationships between features