Comparative Analysis of Accuracy of Supervised Learning Classifier for breast cancer classification
Vrinda Sachdeva, Vasudha Arora · 2022
Breast Cancer is widely spreading disease among women all over the world. This cancer is considered as one of the deadly disease among women. Data mining algorithms play a vital role to predict the early stage cancer .The research problem is that there are lots of classifier with different level of accuracy. An approach for improving the performance and accuracy of three different classifiers, Decision Tree, Logistic Regression, and SVM, is proposed in this study. We also compare the classifier on Wisconsin breast cancer dataset. Accuracy of classifier depends on the shape of data. Imbalanced data is a big problem for the classification phase. Each classifier's efficiency is accessed in terms of confusion matrix accuracy, AUC, and ROC.There are five sections to this study. First, we'll load the dataset and applied correlation to remove some features. For scaling and to filter data, preprocessing algorithm is applied then dataset is splited into training and testing data set and then applied CART algorithms for classification purpose. Then generate some graph to visualize the data. Based on the WBC dataset using Jupyter notebook software, the goal of this paper is to discover the most accurate classifier. The results have been compared with the results of previous papers and reveal that they are extremely precise. The SVM classifier model was discovered to be the best classifier of all the types. It also removes the over fitting of data.