Adaptive Case Based Reasoning for Fault Diagnosis

A hybrid system of Case Based Reasoning (CBR) with Fuzzy ARTMAP (FAM) has been proposed to perform fault diagnosis for actuator system in DAMADICS benchmark. The hybrid system of CBR and FAM is for undertaking the stability plasticity dilemma for the incremental learning problem in CBR. At the same...

Full description

Bibliographic Details
Main Authors: Pang, Shen Yee, Loo, Chu Kiong, Lim, Way Soong
Format: Conference or Workshop Item
Published: 2009
Subjects:
Online Access:http://shdl.mmu.edu.my/1931/
_version_ 1848789917322182656
author Pang, Shen Yee
Loo, Chu Kiong
Lim, Way Soong
author_facet Pang, Shen Yee
Loo, Chu Kiong
Lim, Way Soong
author_sort Pang, Shen Yee
building MMU Institutional Repository
collection Online Access
description A hybrid system of Case Based Reasoning (CBR) with Fuzzy ARTMAP (FAM) has been proposed to perform fault diagnosis for actuator system in DAMADICS benchmark. The hybrid system of CBR and FAM is for undertaking the stability plasticity dilemma for the incremental learning problem in CBR. At the same time, FAM can overcome the difficulty of indexing and retrieval in CBR as well as adaption of cases. FAM is used to make hypotheses and to guide the search of similar cases in the library, while CBR is used to select the most similar match for a given problem, supporting a particular hypothesis. A CBR system supports problem solving based on past experience with similar decision problems. The main strength lies in the fact that it enables directly reusing concrete examples in history and consequently eases the knowledge acquisition bottleneck.
first_indexed 2025-11-14T18:04:20Z
format Conference or Workshop Item
id mmu-1931
institution Multimedia University
institution_category Local University
last_indexed 2025-11-14T18:04:20Z
publishDate 2009
recordtype eprints
repository_type Digital Repository
spelling mmu-19312020-12-29T18:57:46Z http://shdl.mmu.edu.my/1931/ Adaptive Case Based Reasoning for Fault Diagnosis Pang, Shen Yee Loo, Chu Kiong Lim, Way Soong T Technology (General) QA75.5-76.95 Electronic computers. Computer science A hybrid system of Case Based Reasoning (CBR) with Fuzzy ARTMAP (FAM) has been proposed to perform fault diagnosis for actuator system in DAMADICS benchmark. The hybrid system of CBR and FAM is for undertaking the stability plasticity dilemma for the incremental learning problem in CBR. At the same time, FAM can overcome the difficulty of indexing and retrieval in CBR as well as adaption of cases. FAM is used to make hypotheses and to guide the search of similar cases in the library, while CBR is used to select the most similar match for a given problem, supporting a particular hypothesis. A CBR system supports problem solving based on past experience with similar decision problems. The main strength lies in the fact that it enables directly reusing concrete examples in history and consequently eases the knowledge acquisition bottleneck. 2009-12 Conference or Workshop Item NonPeerReviewed Pang, Shen Yee and Loo, Chu Kiong and Lim, Way Soong (2009) Adaptive Case Based Reasoning for Fault Diagnosis. In: International Conference of Soft Computing and Pattern Recognition, 04-07 Dec 2009, Malacca, Malaysia. http://dx.doi.org/10.1109/SoCPaR.2009.135 doi:10.1109/SoCPaR.2009.135 doi:10.1109/SoCPaR.2009.135
spellingShingle T Technology (General)
QA75.5-76.95 Electronic computers. Computer science
Pang, Shen Yee
Loo, Chu Kiong
Lim, Way Soong
Adaptive Case Based Reasoning for Fault Diagnosis
title Adaptive Case Based Reasoning for Fault Diagnosis
title_full Adaptive Case Based Reasoning for Fault Diagnosis
title_fullStr Adaptive Case Based Reasoning for Fault Diagnosis
title_full_unstemmed Adaptive Case Based Reasoning for Fault Diagnosis
title_short Adaptive Case Based Reasoning for Fault Diagnosis
title_sort adaptive case based reasoning for fault diagnosis
topic T Technology (General)
QA75.5-76.95 Electronic computers. Computer science
url http://shdl.mmu.edu.my/1931/
http://shdl.mmu.edu.my/1931/
http://shdl.mmu.edu.my/1931/