Diagnosis of Human Psychological Disorders using Supervised Learning and Nature-Inspired Computing Techniques: A Meta-Analysis

Abstract

A psychological disorder is a mutilation state of the body that intervenes the imperative functioning of the mind or brain. In the last few years, the number of psychological disorders patients has been significantly raised. This paper presents a comprehensive review of some of the major human psychological disorders (stress, depression, autism, anxiety, Attention-deficit hyperactivity disorder (ADHD), Alzheimer, Parkinson, insomnia, schizophrenia and mood disorder) mined using different supervised and nature-inspired computing techniques. A systematic review methodology based on three-dimensional search space i.e. disease diagnosis, psychological disorders and classification techniques has been employed. This study reviews the discipline, models, and methodologies used to diagnose different psychological disorders. Initially, different types of human psychological disorders along with their biological and behavioural symptoms have been presented. The racial effects on these human disorders have been briefly explored. The morbidity rate of psychological disordered Indian patients has also been depicted. The significance of using different supervised learning and nature-inspired computing techniques in the diagnosis of different psychological disorders has been extensively examined and the publication trend of the related articles has also been comprehensively accessed. The brief details of the datasets used in mining these human disorders have also been shown. In addition, the effect of using feature selection on the predictive rate of accuracy of these human disorders is also presented in this study. Finally, the research gaps have been identified that witnessed that there is a full scope for diagnosis of mania, insomnia, mood disorder using emerging nature-inspired computing techniques. Moreover, there is a need to explore the use of a binary or chaotic variant of different nature-inspired computing techniques in the diagnosis of different human psychological disorders. This study will serve as a roadmap to guide the researchers who want to pursue their research work in the mining of different psychological disorders.

Keywords: Accuracy; Classification; Nature-inspired computing techniques; Psychological disorders; Supervised learning techniques.