Prediction of generalized anxiety disorder using particle swarm optimization

Abstract

Diseases can be predicted by using historical patient information stored in clinical databases. Large data is required to ensure the accuracy of prediction. However, processing and extracting valuable information from huge data is a challenging and time-consuming task. Missing and incomplete data may easily cause the data to be ignored and not fully utilized in the prediction. In this paper we focus our study on generalized anxiety disorder. Prediction of generalized anxiety disorders is carried out using feature selection and classification approach. This research focuses on studying and implementing Particle Swarm Optimization algorithm and Fuzzy Rough Set in the classification of generalized anxiety disorders. Performance of classifier model is evaluated respectively based on the accuracy, sensitivity and specificity of results produced. It is found that the proposed hybrid approach in feature selection has different results in performance depending on the selection of classification technique.