Prediksi Diagnosis dan Prognosis Breast Cancer menggunakan Machine Learning
Abstract
Breast cancer is one of the most common health problems affecting women around the world. Breast cancer diagnostic involves the identification and assessment of tumors in the breast tissue to determine the malignant or benign nature of the cancer. Meanwhile, breast cancer prognostic aims to identify disease progression after treatment and predict the likelihood of recurrence. This research aims to analyze the latest developments for predicting breast cancer diagnosis and prognosis using machine learning with k-nearest neighbors and logistic regression models and deep learning using artificial neural network models with sequential models. in this research, things are done such as: conducting data exploration, preprocessing data, oversampling for unbalanced datasets and training models. The results show that deep learning and machine learning predictions are suitable for predicting breast cancer diagnosis while prediction for breast cancer prognosis is suitable using machine learning. All results were compared using the evaluation metrics used in this study such as accuracy, precision, recall and F1-Scores. The best-performing model for the diagnosis dataset is logistic regression, while for the prognosis dataset, the best-performing model is the deep learning model using oversampling. The best-performing model for the diagnosis dataset is logistic regression, while for the prognosis dataset, the best-performing model is the deep learning model using oversampling.