1 AIT Asian Institute of Technology

Concept generalization in adaptive resonance theory

AuthorEi Ei Khin
Call NumberAIT Thesis no. CS-96-9
Subject(s)Neural networks (Computer science)
NoteA thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering.
PublisherAsian Institute of Technology
Series StatementThesis ; no. CS-96-9
AbstractThis literature concerns with Adaptive Resonance Theory network (ART) which widens the path to implementing human brain performance to computers. This literature proposes and characterizes the new similarity measure which agrees with the general belief Increase in vigilance value should increase the number of clusters formed. The proposed similarity measure is consistent with it for one list presentation of input patterns as well as repeated list presentations needed for stable categorization. This measure also possesses the property of requiring lowest list presentations needed for self stabilization of network comparing with other similarity measures. It is an important and useful characteristic. The author also encompasses the papers related to ART and experimentally compares the similarity measures before her thesis. Analysis of Carpenter and Grossberg' similarity measure and the proposed similarity measure under noisy environment is also made.
Year1996
Corresponding Series Added EntryAsian Institute of Technology. Thesis ; no. CS-96-9
TypeThesis
SchoolSchool of Engineering and Technology (SET)
DepartmentDepartment of Information and Communications Technologies (DICT)
Academic Program/FoSComputer Science (CS)
Chairperson(s)Sadananda, R.
Examination Committee(s)Phien, H. N.;Shrestha, Amarottam
Scholarship Donor(s)DAAD
DegreeThesis (M.Eng.) - Asian Institute of Technology, 1996


Usage Metrics
View Detail0
Read PDF0
Download PDF0