Association Rule Mining Based on Ethnic Groups and Classification using Super Learning¹

Cancer is one of the most devastating diseases worldwide. It affects nearly every household, although the prevalence of cancer types varies by geographical regions. One example is breast cancer, which is the most common type of cancer in women worldwide. Therefore, prevention strategies are needed to reduce morbidity and mortality. Identifying risk factors of breast cancer is crucial since it allows physicians to acquaint them with the risks. Accordingly, physicians can recommend precautionary actions. We will first look at the discovery of significant rules for breast cancer patients focusing on different ethnic groups. Predicting the risk of breast cancer occurrence is an essential issue for clinical oncologists. A reliable prediction will help oncologists and other clinicians in their decision-making process and allow clinicians to choose the most reliable and evidence-based treatment. We will next describe the use of a super learner or stacked ensemble technique to a breast cancer data set obtained from the Breast Cancer Surveillance Consortium (BCSC) database. We conducted a comparison of the performance of the proposed super learner and individual base learners. The results of the first part of this study (rule extraction from distinct ethnic groups of breast cancer patients) identified well-known ethnic disparities in cancer prevalence. These experimental results revealed that the generated rules hold the highest confidence level. We also interpreted crucial rules that can be easily understood. Physicians or primary care providers can make decisions by analyzing these rules. A prevention plan or process based on ethnic backgrounds can begin in the primary stage of disease or cancer progression. The second study showed that super learning for risk factor data provides a decent predictive performance compared to the individual three machine learning algorithms that were employed as the base learners for this study.