1 AIT Asian Institute of Technology

Bezier neural networks

AuthorDhargye, Trinley
Call NumberAIT Thesis no. CS-99-19
Subject(s)Neural networks (Computer science)
NoteA thesis submitted in partial fulfillment of the requirements for the Degree of Master of Engineering, School of Engineering and Technology
PublisherAsian Institute of Technology
AbstractA time series could be considered as a collection of observations made sequentially over time at almost equal time intervals. In this study, various types of neural networks were developed using Bernstein polynomials and Ball basis functions to build the most suitable network model for forecasting and filtering river discharge and some economic data. Two types of Bezier network models were developed for the said networks. The first model was created by a combination of univariate Bernstein polynomials and the analytical method, the second model by taking the tensor product of univariate Bernstein polynomials. The analytical method and instantaneous learning rule were used to find the weight vector, which keeps the information of the network under construction. Applications of these models to several data sets show that they perform well in a number of cases, regardless of their simple structure.
Year1999
TypeThesis
SchoolSchool of Engineering and Technology (SET)
DepartmentDepartment of Information and Communications Technologies (DICT)
Academic Program/FoSComputer Science (CS)
Chairperson(s)Huynh Ngoc Phien
Examination Committee(s)Kanchana Kanchanasut;Hoang Le Tien
Scholarship Donor(s)Finnish Government
DegreeThesis (M.Eng.) - Asian Institute of Technology, 1999


Usage Metrics
View Detail0
Read PDF0
Download PDF0