Help via Camal

In the context of heart disease risk prediction, data mining plays a crucial role in extracting insights from complex datasets. This dataset encapsulates a collection of information about various factors impacting heart health. The HELP via CAMaL: Heart Evaluation and Lifestyle Prediction via Classifier Algorithm of Machine Learning aims to unveil patterns that might be hidden and systematically uncover complex relationships by employing data mining, especially through classification algorithms. The source of data was obtained from Behavioral Risk Factor Surveillance System for the year 2021. The original dataset contains a total of 308,855 instances, but only 20,000 instances were selected to apply in the model. The recognition of significant features performed through data cleaning, data preparation including data transformation and data reduction processes. Prediction models were then developed using different classifier algorithms: Decision Tree, Naive Bayes, Random Forest, Support Vector Machine, and Vote. The results reveal the prediction model developed using the Vote (i.e. hybrid of Random Forest and Logistic Regression) algorithms with highest accuracy of 90% effectively predict the risk factors of heart diseases. The summary of results displayed using dashboard for data visualization, providing a clearer understanding of the dataset's insights.