Early Prediction of Mental Health Problems of Children Using Fuzzy Clustered Ensemble Classifier

Sumathi M R
2017 International Journal for Research in Applied Science and Engineering Technology  
Recent surveys show that 12% of children between 4 to 16 years suffer from psychiatric disorders. Early diagnosis and effective treatment are needed to improve the quality of life of children and to avoid complicated problems at a later stage. Mental health decision support systems are generally based on single classifier model. In this research article, anensemble classifier is proposed for predictingmental health problems of children. The proposed ensemble classifier ensembles Random Forest
more » ... les Random Forest classifiers with different features clustered upon the mutual information among them. The data set was collected from a clinical psychologist. The study investigated the effectiveness of the ensemble classifier with different feature sets and by comparing the results with single classifier-based models. K-fold cross-validation technique is used to assess the performance of the ensemble-classifier. The experimental results show that the proposedfuzzy clustered ensemble classifier achieved high diagnostic accuracy for language and communication problem and for attention & academic problem.
doi:10.22214/ijraset.2017.10160 fatcat:6bhlhu6dqjcwnfaqlmbiipxsrm