Early disease detection and prevention play a very significant role in reducing deaths as well as the cost of healthcare. It was found that 8% of women were diagnosed with Breast Cancer (BC) throughout their life. BC is characterized by gene mutation, constant pain, as well as changes in size, color (redness), and breast skin texture. Machine Learning (ML) technologies play an important role in diagnosing and predicting the prognosis of BC. Also, it helps in recognizing people with BC, distinguish benign from malignant tumors using classification techniques. In the current study, we apply four various classifier algorithms: K-star, Na ̈ıve Bayes (NB), Clonal Selection Algorithm (CLONALG), and Artificial Immune Recognition System (AIRS) for BC classification model. The two algorithms were evaluated through a series of experiments over real datasets. We chose five metrics to evaluate performance of the applied algorithms, i.e. accuracy, precision, sensitivity, specificity, and Area Under the ROC Curve (AUC). The results showed that the K-star algorithm has better results than the old ones. Also, experiments indicated that the K-star algorithm provides the highest accuracy, sensitivity, specificity, precision, and AUC with 97.142%, 100.00%, 95.24%, 93.3%, and 0.998, respectively.
Digital Object Identifier (DOI)
Sakr, Mohamed; Saber, Abeer; M. Abo-Seida, Osama; and Keshk, Arabi
"Machine Learning for Breast Cancer Classification Using K-Star Algorithm,"
Applied Mathematics & Information Sciences: Vol. 14:
5, Article 13.
Available at: https://digitalcommons.aaru.edu.jo/amis/vol14/iss5/13