Enhancing Feature Selection Accuracy using Butterfly and Lion Optimization Algorithm with Specific Reference to Psychiatric Disorder Detection & Diagnosis

Abstract

As the complexity of medical computing increases the use of intelligent methods based on methods of soft computing also increases. During current decade this intelligent computing involves various meta-heuristic algorithms for Optimization. Many new meta-heuristic algorithms are proposed in last few years. The dimension of this data has also wide. Feature selection processes play an important role in these types of wide data. In intelligent computation feature selection is important phase after the pre-processing phase. The success of any model depends on how better optimization algorithms is used. Sometime single optimization algorithms are not enough in order to produce better result. In this paper meta-heuristic algorithm like butterfly optimization algorithm and enhanced lion optimization algorithm are used to show better accuracy in feature selection. The study focuses on nature based integrated meta-heuristic algorithm like Butterfly Optimization and lion-based optimization. Also, in this paper various other Optimization algorithms are analyzed. The study shows how integrated methods are useful to enhance the accuracy of any computing model to solve Complex problems. Here experimental result has shown by proposing and hybrid model for two major psychiatric disorders one is known as autism spectrum and second one is Parkinson's disease.