File(s) under permanent embargo
Robust modular artmap for multi-class shape recognition
conference contribution
posted on 2008-01-01, 00:00 authored by C Tan, C Loy, W Lai, Chee Peng LimChee Peng LimThis paper presents a fuzzy ARTMAP (FAM) based modular architecture for multi-class pattern recognition known as modular adaptive resonance theory map (MARTMAP). The prediction of class membership is made collectively by combining outputs from multiple novelty detectors. Distance-based familiarity discrimination is introduced to improve the robustness of MARTMAP in the presence of noise. The effectiveness of the proposed architecture is analyzed and compared with ARTMAP-FD network, FAM network, and One-Against-One Support Vector Machine (OAO-SVM). Experimental results show that MARTMAP is able to retain effective familiarity discrimination in noisy environment, and yet less sensitive to class imbalance problem as compared to its counterparts.
History
Event
IEEE International Joint Conference on Neural Networks (2008 : Hong Kong)Pagination
2405 - 2412Publisher
IEEELocation
Hong KongPlace of publication
Piscataway, N. J.Start date
2008-06-01End date
2008-06-08ISSN
1098-7576ISBN-13
9781424418206ISBN-10
1424418208Language
engPublication classification
E1.1 Full written paper - refereedTitle of proceedings
IJCNN 2008 : Proceedings of the IEEE International Joint Conference on Neural NetworksUsage metrics
Categories
No categories selectedKeywords
adaptive resonance theoriesclass imbalance problemsfuzzy ARTMAPmodular architecturesnoisy environmentsproposed architecturesshape recognitionssupport vectorsScience & TechnologyTechnologyComputer Science, Artificial IntelligenceComputer Science, CyberneticsEngineering, Electrical & ElectronicComputer ScienceEngineering
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC