International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 45 - Number 7 |
Year of Publication: 2012 |
Authors: Mabruk. Ali Fekihal, Jabar H. Yousif |
10.5120/6793-9120 |
Mabruk. Ali Fekihal, Jabar H. Yousif . Self-Organizing Map Approach for Identifying Mental Disorders. International Journal of Computer Applications. 45, 7 ( May 2012), 25-30. DOI=10.5120/6793-9120
Classifications of mental illness such as schizophrenia are very broad; therefore, the proposed approach attains at practical and task-relevant diagnostic categories by use of clustering techniques. A Self-Organizing Feature Map (SOFM) approach was design and implemented for classifying transcribed speech samples and determines mental disorders. An unsupervised Artificial Neural Network was implemented using the NeuroSolution. The proposed classification system is used to determine whether a text or speech sample is generated by a person has mental illness or not. The proposed approach shows clearly that all the categories are identified and classified appropriately, with the proposed SOFM achieving a high accuracy of (97) in the classification phase for predicting the desired output.