Breast Cancer Prediction Using Long Short-Term Memory Algorithm

Abstract

Breast cancer (BC) isconsidered the second leading cause of death in both developed and developing countries, with 8% of women being diagnosed with the disease at some time in their life. So, it's more crucial to identify BC and the damaged breast region. In today's world, Machine Learning (ML) algorithms are frequently employed in the classification of breast cancer datasets. These algorithms have quite a significant level of classification accuracy and diagnostic capability. Because a specific classifier may or may not perform well enough for such datasets, a comparison examination of classifiers is necessary in order to get maximum performance in such significant breast cancer predictions. Deep learning is the branch of machine learning with architecture and functions inspired by the human brain. It's especially effective for classifying enormous data sets because the findings are fast and accurate. In this paper, we have used five different machine learning algorithms: K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and long short-term memory (LSTM) on BC dataset. The outcomes produced by KNN, SVM, RF, and DT classifier will all be compared to the LSTM classifier on the basis of confusion matrix, precision, F1 score, Recall, and accuracy. This study's main aim is to diagnose the best machine-learning algorithm for breast cancer prediction. It is observed that the LSTM algorithmoutperforms all other discussed algorithms with 96% accuracy.