A case study in the use of knowledge-based artificial neural networks for product selection decision support | |
| Author | Silva, W. P. C. Duminda de |
| Call Number | AIT Thesis no.CS-02-01 |
| Subject(s) | Expert systems (Computer science) Decision support systems Neural networks (Computer science) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, School of Advanced Technologies |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. CS-02-01 |
| Abstract | In this study, we explore the use of Knowledge Based Altificial Neural Networks (KBANN), pioneered by Shavlik and Towell, 1994 [25], to learn user preferences under ce1tainty. We start by describing the problem of choosing a used motorcar, where it is reasonable to make several assumptions about preferential independence and monotonicity. We then show how to represent these assumptions as Horn-Clause theories that can be encoded in a !<BANN. We then empirically compare the KBANN with simple back propagation Artificial Neural Network (ANN) in terms of learning rate and accuracy. We also examine the performance of KBANN in learning preferences using examples generated from a number of value functions that violate the assumptions to various degrees and compare the results with that of ANN. |
| Year | 2002 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. CS-02-01 |
| Type | Thesis |
| School | School of Advanced Technologies (SAT) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Haddawy, Peter; |
| Examination Committee(s) | Phan Minh Dung ;Dentcho N. Batanov; |
| Scholarship Donor(s) | H.E. the President's Scholarship Government of Sri Lanka ;Asian Institute of Technology; |
| Degree | Thesis (M.Sc.) - Asian Institute of Technology, 2002 |
 | |