Abstract: Cardiovascular disease (CVD) continues to be a leading cause of death worldwide. Early detection is critical for timely intervention and reducing mortality. Although vast medical data is generated daily, effective utilization of this data for accurate prediction remains a challenge. This study applies data mining techniques to multiple publicly available datasets, including the Cleveland Heart Disease dataset, Framingham dataset, and UCI Heart Disease dataset, to develop predictive models for heart disease detection. Using the Knowledge Discovery in Databases (KDD) methodology, three classifiers—J48 Decision Tree, Naïve Bayes, and Artificial Neural Network (ANN)—were trained and evaluated. Results indicate high classification accuracy across all datasets, with J48 achieving the highest average performance (accuracy ~94.8%). The study demonstrates that data mining can efficiently predict heart disease and offers decision support tools for clinicians to enhance diagnostic consistency.

Keywords: KDD, Data Mining, Heart Disease, Decision Tree, Neural Network, Naïve Bayes, Cleveland Heart Disease Dataset, Framingham Dataset.

Downloads: | DOI: 10.17148/IARJSET.2025.12923