Classification Models for Breast Cancer Detection
B. Varsha, P. Sneka, A. Tanuja, J. Shana · 2023
In the modern era, breast cancer is one of the most common cancers worldwide. In 2020, nearly 2.3 million women were diagnosed with breast cancer (one in every eight women). The goal of this study is to compare three machine learning models, namely, logistic regression, decision tree, and random forest classifier which have been implemented for breast cancer. The patient's dataset was collected; the dataset contained 569 rows of data, that is 569 patients’ data and 33 columns which are the features based on the classification. The dataset consists of attributes of the nuclei measurements which consist of texture, radius, perimeter, area, concavity, etc. In this breast cancer classification, the cancer is mainly classified based on the type of cancer cells, that is either benign or malignant (already given in the dataset as patients having cancer or not). Benign are non-cancerous tumour cells which do not invade neighbouring cells, whereas malignant are the cancerous cells causing tumours which invade neighbouring cells. The split data consists of 25% of testing data and 75% of training data. Machine learning models such as logistic regression, random forest classifier and decision tree classifier models are implemented. The results showed that random forest classifier has a comparatively higher accuracy with 98%.