Feature Selection for Identifying Parkinson’s Disease Using Binary Grey Wolf Optimization
11 Pages Posted: 7 Mar 2018
Date Written: November 15, 2017
Parkinson’s disease is a neurological disorder that directly affects human gait. It leads to slowness of movement, cannot speak properly and muscle rigidity. In this paper we perform a comparative analysis of various bio-inspired algorithms to select optimal feature. The proposed Binary Grey Wolf Optimization method is compared with four common optimizers namely Particle Swarm Optimization (PSO),Genetic Algorithm (GA), Binary Bat Algorithm (BBA) and Modified Cuckoo Search Algorithm (MCS). Accuracy and number of features selected are used to evaluate and compared with speech dataset from the UCI repository. Results indicate that the proposed binary version of grey wolf optimization (BGWO) has searched the feature space for optimal feature combinations with good accuracy.
Keywords: Parkinson’s disease, Feature selection, Binary grey wolf optimization algorithm
Suggested Citation: Suggested Citation