Fuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem
碩士 === 東海大學 === 工業工程研究所 === 86 === In current complex environment with information continuously updating, knowledge acquisition has become a bottleneck for the expert systems. A symbolic artificial intelligent (AI) system, though, may possess reasoning knowledge with interpretability,...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
1998
|
Online Access: | http://ndltd.ncl.edu.tw/handle/78315688675261939971 |
id |
ndltd-TW-086THU00030011 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-086THU000300112015-10-13T17:34:41Z http://ndltd.ncl.edu.tw/handle/78315688675261939971 Fuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem 模糊訊息需求分類者系統-以IC測試機台故障診斷問題為例 Liu,Chia-Hong 劉嘉宏 碩士 東海大學 工業工程研究所 86 In current complex environment with information continuously updating, knowledge acquisition has become a bottleneck for the expert systems. A symbolic artificial intelligent (AI) system, though, may possess reasoning knowledge with interpretability, but lack of learning capability-especially as environmental status and dynamic trend change, input condition habit switches. Numerical artificial intelligent systems, on the other hand, though, possess relearning capability, but lack of interpretability-such Holland first proposed the classifier system in 1975. Classifier systems are genetic-algorithm based systems, incorporate the reinforcement learning, and constitute one of the theories of machine learning. It has used simple binary-variables (1, 0, #) for knowledge representation, and expedites the system responsive speed. Nevertheless, if encountered with environmental uncertainty/fuzziness, continuous input variables, and fuzzy feedback, there is a limitation. Based on these reasons described above, this research proposed a fuzzy message requirement classifier system (FMRCS), which integrated learning and reasoning. The FMRCS can be divided into two systems: message requirement classifier system (MRCS) and fuzzy classifier system (FCS). First, the MRCS simulates the internal reasoning process of human being, and finds the key factor of decision problem, under the limited known of messages of problem domain. Then, according to the key factor created by the MRCS, t-1 -aFuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem Chang,Ping-Teng 張炳騰 1998 學位論文 ; thesis 115 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 東海大學 === 工業工程研究所 === 86 === In current complex environment with information continuously updating, knowledge acquisition has become a bottleneck for the expert systems. A symbolic artificial intelligent (AI) system, though, may possess reasoning knowledge with interpretability, but lack of learning capability-especially as environmental status and dynamic trend change, input condition habit switches. Numerical artificial intelligent systems, on the other hand, though, possess relearning capability, but lack of interpretability-such Holland first proposed the classifier system in 1975. Classifier systems are genetic-algorithm based systems, incorporate the reinforcement learning, and constitute one of the theories of machine learning. It has used simple binary-variables (1, 0, #) for knowledge representation, and expedites the system responsive speed. Nevertheless, if encountered with environmental uncertainty/fuzziness, continuous input variables, and fuzzy feedback, there is a limitation. Based on these reasons described above, this research proposed a fuzzy message requirement classifier system (FMRCS), which integrated learning and reasoning. The FMRCS can be divided into two systems: message requirement classifier system (MRCS) and fuzzy classifier system (FCS). First, the MRCS simulates the internal reasoning process of human being, and finds the key factor of decision problem, under the limited known of messages of problem domain. Then, according to the key factor created by the MRCS, t-1 -aFuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem
|
author2 |
Chang,Ping-Teng |
author_facet |
Chang,Ping-Teng Liu,Chia-Hong 劉嘉宏 |
author |
Liu,Chia-Hong 劉嘉宏 |
spellingShingle |
Liu,Chia-Hong 劉嘉宏 Fuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem |
author_sort |
Liu,Chia-Hong |
title |
Fuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem |
title_short |
Fuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem |
title_full |
Fuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem |
title_fullStr |
Fuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem |
title_full_unstemmed |
Fuzzy Message Requirement Classifier System:A Case Study of IC Test Machine Diagnosis Problem |
title_sort |
fuzzy message requirement classifier system:a case study of ic test machine diagnosis problem |
publishDate |
1998 |
url |
http://ndltd.ncl.edu.tw/handle/78315688675261939971 |
work_keys_str_mv |
AT liuchiahong fuzzymessagerequirementclassifiersystemacasestudyofictestmachinediagnosisproblem AT liújiāhóng fuzzymessagerequirementclassifiersystemacasestudyofictestmachinediagnosisproblem AT liuchiahong móhúxùnxīxūqiúfēnlèizhěxìtǒngyǐiccèshìjītáigùzhàngzhěnduànwèntíwèilì AT liújiāhóng móhúxùnxīxūqiúfēnlèizhěxìtǒngyǐiccèshìjītáigùzhàngzhěnduànwèntíwèilì |
_version_ |
1717780878861533184 |