1
Cascade correlation algorithm for forecasting traffic congestion | |
Author | Suhendra, Adang |
Call Number | AIT Thesis no.CS-94-39 |
Subject(s) | Traffic estimation Algorithms |
Note | A thesis submitted in partial fulfillment of the requirements for degree of the Master of Science, School of Engineering and Technology |
Publisher | Asian Institute of Technology |
Abstract | Many applications have been made by utilizing the Artificial Neural Networks. One of those applications is managing the traffic by forecasting the traffic congestion involving such parameter as capacity at intersection. It is necessary to have short-term forecasting congestion at busy intersection for dynamic decision making. Therefore a fast forecasting algorithm is i·equired to predict the traffic capacity. Research on forecasting the traffic congestion has been done by utilizing Backpropagation Algorithm. The disadvantage of Backpropagation Algorithm is the low training speed .. This thesis investigates the Cascade Correlation Algorithm for forecasting the traffic capacity. Cascade Correlation technique trains the networks faster than Backpropagation. Cascade Correlation is supervised learning algorithm that dynamically builds the network while training the input output patterns. The experiments of this thesis investigate the performance of Cascade Correlation for forecasting the traffic capacity in Bangkok, and compare the results with the Backpropagation Algorithm. |
Year | 1994 |
Type | Thesis |
School | School of Engineering and Technology (SET) |
Department | Other Field of Studies (No Department) |
Academic Program/FoS | Computer Science (CS) |
Chairperson(s) | Sadananda, Ramakoti |
Examination Committee(s) | Huynh, Ngoc Phien ; Kanchana Kanchanasut |
Scholarship Donor(s) | Yayasan Gunadarma Jakarta, Indonesia |
Degree | Thesis (M.Sc.) - Asian Institute of Technology, 1994 |