Breast Cancer Detection Using K-Nearest Neighbors, Logistic Regression and Ensemble Learning
Ram MurtiRawat, Shivam Panchal, Vivek Kumar Singh, Yash Panchal
Abstract
Breast Cancer is one of the most severe diseases that is faced by women leading nowhere other than increased death rates in society and it is considered to be one of the most intense disease in the history of medical science. Looking at the number of deaths caused by Breast cancer, it is considered to be a major threat but today's advancement in medical science has the capability to cure such threat completely if detected at its early stages without causing any harm to the patient. The major challenge arise during the detection of cancer and differentiating between the diagnosis that affirms whether the patient has a benign or malignant type of cancer. Machine Learning Algorithms like K-Nearest Neighbors, Support Vector Machine (SVM) and Artificial Neural Network (ANN) helps us solve this problem by achieving results with high precision and accuracy. The following paper helps in diagnosis of breast cancer using Logistic Regression (LR), K-Nearest Neighbors (KNN) and Ensemble Learning with Principal Component Analysis (PCA) and a comparative study is also made with other papers on the basis of accuracy. The models used here are trained and tested on Wisconsin breast cancer diagnosis data set which is taken from UCI machine learning repository. Pre processing of data was performed followed by feature extraction of data set using Principal Component Analysis (PCA). Various Machine Learning techniques were proposed in the paper which helped us achieve an accuracy of 98.60% using K-Nearest Neighbors, 97.90 using Logistic Regression and 99.30% using Ensemble Learning.