Abstract

Abstract: The oldest recognised cancer type in humans is Breast Cancer. In approximately 1600 BC, Egypt recorded the oldest cancer detection and definition. Although this problem has subsequently been evaluated and investigated for the purpose of avoiding serious consequences, it remains one of the deadliest diseases ever, with breast cancer mortality in the U.S alone exceeding 40,000 in 2012. In most cancer-affected women, cancer of breast is the leading cause of mortality. With a low mortality rate, mammography is one of the most successful early detection and diagnostic methods for cancer of breast . Breast x-rays, or mammograms, are used to identify early symptoms of breast cancer.These X-ray images reduce human cyst detection errors, reduce diagnostic time, and increase diagnosis accuracy.This study provides an overview of strategies machin learningn for diagnosis the breast cancer & classification and may be divided into three primary stages: preprocessing, feature extraction and classification. There are many innovative methods and approaches for timely identification of breast cancer in modern medical science. Most of these approaches involve state-of-the-art technologies like medical image processing. Early detection is useful for doctors, which considerably increases cancer patients' survival rates. This research examines three of the most used ML approaches for the diagnosis & direction of breast cancer. The techniques are the SVM, Random Forestry (RF) and the Naive Bayes (NB). For each of the above strategies, the probability will be computed and the most accurate algorithm results will be obtained.Keywords: Medical Diagnosis, Breast Cancer, Classification, Machine Learning .