Extraction Of Features From Mammographic Images For Screening And Classification Of Breast Cancer And Logistic Regression Algorithm.
Abstract
This study was performed on the extraction of mammographic imaging features for the detection and classification of breast cancer and the logistic regression algorithm. This study employed an ML model for the classification and identification of breast cancer. To achieve this, we utilized logistic regression (LR) and also compared the performance of our model with other existing ML models, namely Support Vector Machine (SVM), Naïve Bayes (NB) and Multilayer Perceptron (MLP)). original Wisconsin Breast Cancer Diagnostic (WDBC) dataset. Our performance review was conducted in two phases, namely phase 1: when the WBCD is scaled (feature scale) and phase 2: when the data set is not scaled. All models, excluding MLP, performed well when there was no scaling of the data set characteristics with f1 scores of (LR = 97%, SVM = 97%, NB = 95%, MLP = 52%). However, when the characteristics scale is applied to the data set, all four models have f1 scores greater than 90% (SVM = 98%, LR = 97%, NB = 97%, MLP = 97%). Remarkably, the f1 score for LR in both cases has not changed, therefore, as far as we know, we have concluded that LR, given its simplicity and low time complexity, is a good model to use for Binomial Classification.