A Semantic Approach for Extracting the Medical Association Rules

Abstract

Healthcare data analytics has indispensable responsibility in advancing remedial decision-making and healthcare practices. In this paper, a novel semantic approach is introduced for extracting medical association rules, designed at discovering expressive interactions between medical things. Our methodology incorporates medical ontologies, and improved semantic measures to boost the accuracy and interpretability of the obtained rules. Our method delivers clinically pertinent information into patient perspectives on mental health, the accessibility of mental health resources, and the influence of physical health on mental well-being by mapping medical notions to ontological items and examining semantic relationships. Experimental validation on a case study of mental health dataset proves the dominance of our method over traditional association rule mining methods. The proposed semantic method presents valuable support to healthcare data analysis and decision-making processes.