Breast Cancerous Tumor Detection Using Supervised Machine Learning Techniques

Abstract

Around the globe the leading cause of cancer mortality in women is Breast Cancer; BCs, have been under many studies and several analyses have shown that many abnormal conditions and risk of BCs can be diagnosed with the help of fast and perfect judgment of the clinical physicians of this domain. This kind of intelligent decision making is not the thing where every person is good at so here, we have kept Artificial intelligence to use with supporting steps like Machine Learning Classifiers from Wisconsin Breast Cancer Dataset; WBCD. In the Combination of Human and Computer Intelligence, we have worked our way around the hybrid Mechanism where we can make the use of both entities and detect as well as prevent the risks of Very Early Stages of BCs. Our Mechanism includes the use of Principle Component Analysis; PCA, at the first stage of pre-processing the decreasing features in the pre-processed data the second stage of pre-processing was more beneficial for accessing the multi pre-processed data for analysis and diagnosis. When the Multi Pre-processed data is accessed by the Support Vector Machine; SVM, the old Accuracy, Specificity and Significance level was ranging from 95 to 100% but the new levels after the multi pre-processing were decreased to 5% and the improvement was considerable and confirmed that the proposed Algorithm has achieved the optimized results then RF, KNN, DT. It means the Proposed algorithm can assist the human factor with faster decision making and can categorize data (High Risk) malignant and (Low Risk) Benign because to avoid Human errors AI is useful and can achieve optimal results.

A novel SVM Kernel Classifier Technique using Support Vector Machine for Breast Cancer Classification

“…Color-level co-occurrence matrix (GLCM), multi-level discrete wavelet transforms (MWT), or Principal Component Analysis (PCA) are some of the techniques used to extract texture from photos (PCA). Only 5% of the previous data's [7] Accuracy, Specificity, and Significance levels remained after the multi pre-processing, indicating that the proposed approach outperformed the other algorithms. A comparison of data mining classification algorithms is therefore presented [8], which focuses on data mining using the learning approach.…”