ONTOLOGY-BASED DECISION TREE MODEL FOR PREDICTION OF CARDIOVASCULAR DISEASE
Hakim El Massari, Noreddine Gherabi, Sajida Mhammedi, Zineb Sabouri, Hamza Ghandi
Abstract
Nowadays, cardiovascular diseases (CVD) are one of the most critical reasons for death. Thus, CVD prediction is a crucial challenge in the field of clinical data analysis. Researchers are using a variety of statistical and machine learning methods to assess immense amounts of complex medical data, to help doctors predict heart disease. In this paper, we proposed a new approach to predict CVD using ML techniques and Ontology to build an efficient ontology-based model able to predict accurately the presence of cardiac disease and establish an early diagnosis. the approach consists of extracting rules from the Decision Tree algorithm that differentiate the patients with or without cardiovascular disease then implementing these rules in the ontology reasoner using Semantic Web Rule Language (SWRL). The ontology model result reach high classification accuracy of 75% compared to the decision tree model. The approach can be employed in the medical field for the prediction of cardiovascular diseases.