Design expert system tool with hybrid knowledge representation and reasoning
Zhan Jiang, Yi Wu, Li Ping Chen, Nian Meng Luo
Abstract
Zhan Jiang, Yi Wu, Li Ping Chen, Nian Meng Luo
Abstract
According to the characteristics and requirements for design expert systems, this paper describes a Design Expert System Tool (DEST) with hybrid knowledge representation and reasoning. The main goal is to provide an effective and convenient development environment of expert system for design domains. It adopts the technique of object oriented knowledge representation, which integrates frame, rule and procedure. For each knowledge representation, different inference engines are designed. Inductive reasoning and Case-Based Reasoning (CBR) are combined to form a hybrid inference engine. For improving the quality of case retrieval, it presents a new method of attribute similarity calculation and weight assigning. Moreover, it provides a visualisation knowledge management platform for expert and designer to build and apply expert systems.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
According to the characteristics and requirements for design expert systems, this paper describes a Design Expert System Tool (DEST) with hybrid knowledge representation and reasoning. The main goal is to provide an effective and convenient development environment of expert system for design domains. It adopts the technique of object oriented knowledge representation, which integrates frame, rule and procedure. For each knowledge representation, different inference engines are designed. Inductive reasoning and Case-Based Reasoning (CBR) are combined to form a hybrid inference engine. For improving the quality of case retrieval, it presents a new method of attribute similarity calculation and weight assigning. Moreover, it provides a visualisation knowledge management platform for expert and designer to build and apply expert systems.
Key concepts: Expert system, Inference engine, Computer science, Knowledge representation and reasoning, Inference, Legal expert system, Model-based reasoning, Representation (politics)