Constructing Fuzzy Automata by Knowware System
This article proposes constructing fuzzy automata by intelligent components in Knowware System (KWS). The fuzzy,automata under discussion are fuzzified deterministic finitestate automata (DFA). The fuzzification appears at two levels:each step of state transition is associated with a predefined membership degree;the classification of input word is under a fuzzy basis,i.e.:whether a specific word belonging to a predeflned category is a matter of degree.A fuzzy automaton is designed based on a fuzzy grammar formalized from a domain model provided by application developer.Each node in such fuzzy automaton is an extension from its original concept in DFA, and realized by intelligent component in KWS with support of domain knowledge.The recognition processing of a fuzzy language consists of multiple chains of state transition in the corresponding fuzzy automaton under the control of KWS engine, it offers the flexibility of reconstructing a fuzzy automaton when receiving updated fuzzy grammar. Algorithms are proposed for converting a domain model to fuzzy automaton,and constructing the fuzzy automaton using KWS components. Examples are provided to show its application potential.
fuzzy automata fuzzy grammars fuzzy languages knowledge-based systems knowware system
Liya Ding Sio-Long Lo
Faculty of Information Technology Macau University of Science and Technology Macau SAR. China
国际会议
The 10th International Conference on Intelligent Technologies(第十届智慧科技国际会议 InTech09)
桂林
英文
534-541
2009-12-12（万方平台首次上网日期，不代表论文的发表时间）