A Survey on Data Mining Algorithms in Prediction of Psychiatric Disorders

The medical diagnosis and automated decision making process by computer algorithms based on data from our behaviors is fundamental to the digital economy. Researchers have been using various data mining techniques and statistical analysis to improve the disease diagnosis accuracy in medical healthcare. Psychiatric disorder is a highly prevalent condition associated with many adverse health problems. Several data mining algorithms with better classification accuracy will provide more sufficient information to identify the severe psychiatric disorders. The objective of the proposed research is the comparison of different data mining algorithms and to predict the acute psychiatric disordered patients more accurately. After feature analysis, models by five algorithms including C5.0, Neural Network, Support Vector Machine (SVM), K - Nearest Neighbor (KNN) and Na?ve bayes developed and validated. C5.0 Decision tree has been able to build a model with greatest accuracy 90.77%, KNN, SVM, Neural Network and Na?ve Bayes have been 73.85%, 83.08%, 77.93 and 90.71% respectively.