ENSEMBLE-OF-CLASSIFIERS APPROACH FOR DIAGNOSIS OF PERVASIVE DEVELOPMENTAL DISORDERS USING PSYCHO-METRIC PROFILES OF CHILDREN

Main Article Content

Dr. Poorna B
Sumathi M.R.

Abstract

Maintaining good health, mentally and physically, is important at every stage of life from childhood to adulthood to live a long and healthy life. But, due to various factors mental health problems have become common today than cancer, diabetes or heart disease. The onset of mental illness starts typically in the early stages of life i.e. infancy or childhood. The diagnosis of the onset of mental illness by the professionals is a complicated task as many factors are involved. An attempt has been made in this research to diagnose the early onset of pervasive developmental disorders from the psycho-metric profiles of children maintained by the psychologist. An ensemble-of-classifiers approach was used for the diagnosis. Individual classifier’s balanced accuracies are used for weighing the classifiers and a final decision was made by averaging their predictions. This ensemble approach provided accuracy from 91% to 97%. Hence, this classifier may be used as an additional tool by the psychologists to diagnose the pervasive developmental disorder.

Downloads

Download data is not yet available.

Article Details

Section
Articles