Multi-objective Association Analysis of Parkinson Disease with Intelligent Optimization Algorithms

Abstract

Parkinson's disease is a neurological disorder that has significant social and economic impacts affecting the patient's quality of life. Combined Parkinson's disease assessment scale is used together with clinical observations and evaluations in the diagnosis of the disease. However, this method may be insufficient especially at the beginning of the disease. Using data mining for analyzing of Parkinson's disease data can lead to the identification of similar patients, with the aim to assist the clinicians to respond more promptly and in a more personalized fashion to the changes of the patients' status. The majority of the studies within Parkinson disease are related to the classification task of data mining. However, association rules mining is one of the most common data mining problems used to find interesting and valuable associations that often occur in large data sets. There are very few studies on association analysis of Parkinson disease. To the best of our knowledge, there is not any study about multi-objective optimization for association analysis in real Parkinson data sets. In this study, multi-objective association analysis of Parkinson disease with MOPNAR and QAR-CIP-NSGAII that aims to automatically find the association rules with related intervals by optimizing many conflicting objectives such as support, confidence, lift, certainty factor, netconf, yulesQ, and coverage simultaneously. © 2019 IEEE.