A Comprehensive Examination, Execution, and Evaluation of Machine Learning Algorithms Applied to a Breast Cancer Dataset Utilizing the WEKA Software Tool

In the contemporary world, cancer ranks as one of the most critical diseases. The factors contributing to cancer include modern lifestyle choices, environmental influences, and genetic predispositions. It has emerged as the leading cause of mortality in developed nations. Among various types of cancer, Breast Cancer is the most prevalent among women and is a significant contributor to female mortality. Cancer screening methods are typically experimental or natural in their approach. Machine Learning algorithms play a crucial role in identifying significant patterns within data, particularly in the context of disease prediction. In this research, a new Machine Learning approach is developed in predicting breast cancer at an early stage. The objective is to assess the effectiveness of the ML algorithm based on metrics such as precision, specificity, sensitivity, and accuracy. A real-time Breast Cancer dataset was collected from the UCI Machine Learning repository for performing research. The data was firstly preprocessed to identify the missing values of the features in the dataset and replaced with the mean of the relevant variable. Then the dataset was projected on popular Machine Learning algorithms, including Sequential Minimal Optimization (SMO), Naive Bayes (NB), J48 (C4.5 decision tree), K-Means, and K-Nearest Neighbours (k-NN). From the analysis, it is observed that the SMO Machine Learning algorithm provides 96.90% accuracy which is very high when compared to the accuracy of existing approaches. The results indicate that SMO outperforms the others in terms of accuracy and low error rate.