An intelligent AAA++ approach to predict high blood pressure using PARP classifier

Abstract

ObjectiveHigh Blood Pressure (HBP) is a major health challenge of many around the world. Existing research covers extensively how to treat HBP, but predicting HBP in advance based on biological and psychological parameters of a person is not covered in the literature. The objective of this paper is to predict HBP based on Bio-Psychological factors of a person. MethodsWe proposed an intelligent Rule-based classifier to predict HBP. The proposed model can be used to prevent HBP rather than using medication. In our approach, we considered AAA++ (Age, Anger level, Anxiety level, Obesity level (+), Cholesterol level (+)) of a person for experimental study. The proposed approach uses priority-based apriori rule pruning (PARP) classifier, which works in 3 stages. Stage 1: generate association rules using apriori. Stage 2: it uses the priority of an attribute to prune the association rules generated in stage 1. Step 3: Rules extracted in stage 2 are used to build a rule-based classifier to predict the class label of test instances. The Results of the proposed model are compared with JRip, PART, OneR and, ZeroR. ResultsExperimentation is done on real-time data set using 10 fold cross-validations. In each fold, 90% data is used to train the model and 10% is used to test the model. The proposed approach has shown improved accuracy (86.4%) and reduced mean length of a rule (1.7) compared to existing rule-based algorithms. Although JRip is good at accuracy (86.9%), but the proposed model has outperformed at the mean length of the rule (1.7). ConclusionThe extracted rules after experimentation are understandable and informative to the technical and nontechnical community to predict HBP.