Prediction of Basic Mental Health Problems of Children (Age <=16) using Fuzzy Clustered Ensemble-Classifier



Mental health decision support systems are generally based on single classifier or a combination of these models. In this paper, we propose a novel ensemble-based classifier that predicts basic mental health problems (BMHP) of children (age <=16 years) with relatively higher accuracy. Recent surveys show that 12% of children between 4 to 16 years suffer from psychiatric disorders. Early diagnosis and effective treatment improves the quality of life of children and avoids complicated problems at a later stage. For predicting BMHP, our classifier ensembles Random Forest classifier with different features clustered upon the mutual information among the features. The data set for predicting BMHP was collected from a clinical psychologist. We investigate the effectiveness of the ensemble classifier with different feature sets by comparing the results with single classifier-based models. K-fold cross-validation tests were used to assess the performance the ensemble-classifier. The experimental results show that our fuzzy clustered ensemble classifier achieves high diagnostic accuracy of 98% for language and communication problem and 88% accuracy for attention & academic problem. 


Mental Health Problem, Ensemble-Classifier, Random Forest, Fuzzy Clustering, Prediction

Full Text:



  • There are currently no refbacks.

Subscribe to Print Journals

 IJAIKD is currently Indexed By  Journal Seek