Performance analysis of regression methods

This master's thesis conducted a performance analysis of various regression methods on heart attack data. Methods such as linear regression, logistic regression, ridge regression, and support vector regression were employed to predict and analyze the likelihood of a heart attack. The aim of the study was to evaluate the performance of each regression method to identify the optimal one that most accurately predicts factors associated with heart attacks. In the first section, a comprehensive review of the existing literature on medical data analysis and heart attack prediction was presented. The second section provided detailed information about the heart attack dataset used and explained the preprocessing steps taken to ensure data integrity. The third section focused on the theoretical foundations of the regression methods and their application to the heart attack dataset. The fourth section explained the performance metrics used to evaluate the regression methods, such as accuracy, precision, and recall. The fifth section presented the results obtained from applying different regression methods to the heart attack data, highlighting the strengths and weaknesses of each method. In the sixth section, the findings were critically analyzed, discussing the potential applications of the identified optimal regression method in real-world scenarios. The final section summarized the main findings, outlined the limitations of the study, and provided suggestions for future research. This study aims to enhance the applicability of regression analysis in the medical field and provide valuable insights to identify the most effective regression method for heart attack prediction.