Published Paper Details:

Currently the most common form of cancer diagnosed in women all over the world is breast cancer. One of the leading causes of death for women, it arises in breast tissue. If this cancer is discovered in its early stages, it can be treated. There are two types of tumors that can occur in breast cancer patients: malignant and benign. Malignant tumors are deadly because its growth rate is much higher than the benign one. Therefore, early identification of tumor type is essential for adequate treatment of a patient with breast cancer. In this work, the Wisconsin breast cancer data set was used, which was collected from the UCI repository. The goal is to analyze the data set and evaluate the performance of various machine learning Breast cancer prediction algorithms, here support vector Machine, logistic regression, K-nearest neighbors, Gradient Boosting Classifier and Random Forest classifiers have been implemented to classify tumors into benign and malignant. To determine the most appropriate algorithm, the accuracy of each is calculated and compared. Accordingly, Logistic Regression classifier has the highest accuracy of 99.12%. These classifiers can be used to build an automatic diagnostic system for preliminary breast diagnosis cancer.