DESIGN AND DEVELOPMENT OF A MACHINE LEARNING BASED HYBRID MODEL UISNG BPNN AND ASSOCIATIVE CLASSIFIER FOR HEART DISEASE PREDICTION

Abstract

Heart disease is a major public health concern with millions of reported deaths annually. Data mining techniques have received attention in recent years as a tool aiding diagnosis and prediction of heart disease cases. Also, Back Propagation Neural Network (BPNN) provides excellent results in binary decision-making applications. As the size of the data under consideration is huge in number, the exploitation of Artificial Neural Network (ANN) in combination with data mining techniques will further enhance the effectiveness of the prediction process.

Rules were extracted for feature selection using Associative classifier as the data mining technique adopted, followed by BPNN training. Model performance was evaluated on heart disease datasets collected from the online dataset repository of Cleveland Heart Disease dataset (https://archive.ics.uci.edu/ml/datasets/heart+disease)

The proposed hybrid model showed improved accuracy compared to standalone BPNN and Associative Classification models. It achieved higher precision, recall, and F1-score, demonstrating better predictive performance on heart disease datasets.

Thus the developed model effectively improved heart disease prediction accuracy, leveraging interpretability from rules and complex pattern recognition from neural networks, offering a balanced approach for clinical applications.