Breast-cancer-prediction

Table of Contents

• Dataset
• Model
• Results

---

Dataset

The Breast Cancer Wisconsin (Diagnostic) Dataset contains data from digitized images of breast mass cell nuclei in fine needle aspirates. The task is to classify the cells as either malignant (cancerous) or benign (non-cancerous) based on the features provided.

The dataset was fetched from the UC Irvine Machine Learning Repository.

---

Model

The model uses XGBoost (XGBClassifier)

1. Data Preprocessing

• Label Encoding: The target variable (Malignant and Benign) is encoded into numerical values (1 and 0, respectively).
• Train-Test Split: The data is split into a training set (80%) and a test set (20%).
• Correlation Handling: Highly correlated features (correlation > 0.9) are dropped to avoid multicollinearity.
• SMOTE (Synthetic Minority Over-sampling Technique): Used to handle class imbalance by oversampling the minority class in the training set.

2. Feature Selection

• RFE (Recursive Feature Elimination): A feature selection method used to select the most important 20 features that contribute the most to the model's performance.

3. Model Tuning

• GridSearchCV: Hyperparameter tuning is performed using GridSearchCV to find the best parameters for the XGBoost classifier. The grid includes:
  • learning_rate, n_estimators, max_depth, gamma, reg_alpha, reg_lambda, and scale_pos_weight.

4. Evaluation

• Classification Report: The model's performance is evaluated using precision, recall, F1-score, and support.
• ROC-AUC Score: The area under the ROC curve is calculated to assess the model's performance in distinguishing between the two classes.

---

Results

The best model, after hyperparameter tuning, is evaluated on the test set. Below is the classification report for the Breast Cancer Dataset model:

Class	Precision	Recall	F1-Score	Support
0 (Benign)	0.99	0.96	0.97	71
1 (Malignant)	0.93	0.98	0.95	43
Accuracy			0.96	114
Macro avg	0.96	0.97	0.96	114
Weighted avg	0.97	0.96	0.97	114

ROC-AUC Score:

The ROC-AUC score of the model is 0.9931, indicating an excellent ability to distinguish between malignant and benign tumors with high confidence.