Power Allocation RBF-type Neural Network Regression Method
Abstract
A novel method, the Power Allocation Radial Basis Function (PA-RBF) is developed in this study. It is based on Decision Tree (DT) classification and RBF neural network regression. In PA-RBF, the DT algorithm is applied in order to separate the whole dataset into segments, after which smaller RBF models are applied on the separated datasets. Then trial and error is utilized in order to find the optimum Division Percent (DP) and Minimum Parent Size (MPS) characteristics of the DT algorithm, as well as the RBF models’ numbers of hidden neurons and spread amounts. The regression performance of the RBF and PA-RBF methods is compared by using a case study of Parkinson’s disease tracking. The results show that the PA-RBF model with PA-RBF with RMSE of 7.14 performs nearly 22 % more accurately compared with the RBF model with RMSE of 8.75.
Short Biography
Dr. Hossein Bonakdari, PhD, P.Eng., is a professor of Department of Soils and Agri‐Food Engineering, Laval University, Québec, Canada, earned his PhD in Civil Engineering at the University of Caen-France. An innovative contribution of his work has been the establishment of a new perspective on application of artificial intelligence techniques for solving engineering problems. He has developed different classes of soft computing methods to deal with the large amounts of water resources data he has encountered during research projects and to address a shortage of existing empirical, semi-empirical, and simplified equations. Results obtained from his researches have been published in more than 180 papers in international journals (h-index=26). He has also more than 150 presentations in national and international conference. He published two books.