/BreastCancerPrediction

Breast Cancer Prediction using Machine Learning

Primary LanguageJupyter NotebookMIT LicenseMIT

Overview

Machine Learning classification task to predict if a breast cancer is benign or malignant.

In the notebook you can:

  • get a first insight of the dataset
  • see all the steps needed for preparing the data for the algorithms
  • look at the training of different models
  • observe a final comparison between them

Dataset

https://www.kaggle.com/uciml/breast-cancer-wisconsin-data

Goals and metrics

In the notebook and in the /res/plots/ folder it's possible to observe more of results/metrics of the project; below are reported some of these.

N.B.: it's possible to find some suffix in the model names:

  • "scaling" means that features have been scaled for that model
  • "grid" means that a grid search has been performed on that model

Accuracy

Accuracy plot

Recall

Recall plot

Precision

Precision plot

ROC_AUC

ROC_AUC plot

Confusion matrices

Confusion matrices plot