No abstract

No abstract

No abstract

No abstract

A PSO based SVM method has been implemented in diagnosing Parkinson's disease. This hybrid method produces parameter optimization and it helps to predict the gene expression pattern of the patient affected from Parkinson's disease. Implementing a computational tool on the PD data set alleviates the symptoms to predict accurately the occurrence of the disease. In data classification, there may arise some incomplete or missing data during pre-processing in the probabilistic model. In order to overcome this, an Expectation Maximization (EM) algorithm is implemented. The proposed Particle Swarm Optimization (PSO) based Support Vector Machine (SVM) technique is also compared with the Bayesian network model that outperforms in prediction accuracy.