This paper presents a two-agent framework to build a natural langua g e query interface for IC information system, focusing more on scope queries in a single English sentence. The first agent, parsing agent, syntact...This paper presents a two-agent framework to build a natural langua g e query interface for IC information system, focusing more on scope queries in a single English sentence. The first agent, parsing agent, syntactically p rocesses and semantically interprets natural language sentence to construct a fu zzy structured query language (SQL) statement. The second agent, defuzzif ying agent, defuzzifies the imprecise part of the fuzzy SQL statement into its e quivalent executable precise SQL statement based on fuzzy rules. The first agent can also actively ask the user some necessary questions when it manages to disa mbiguate the vague retrieval requirements. The adaptive defuzzification approach employed in the defuzzifying agent is discussed in detail. A prototype interface has been implemented to demonstrate the effectiveness.展开更多
文摘This paper presents a two-agent framework to build a natural langua g e query interface for IC information system, focusing more on scope queries in a single English sentence. The first agent, parsing agent, syntactically p rocesses and semantically interprets natural language sentence to construct a fu zzy structured query language (SQL) statement. The second agent, defuzzif ying agent, defuzzifies the imprecise part of the fuzzy SQL statement into its e quivalent executable precise SQL statement based on fuzzy rules. The first agent can also actively ask the user some necessary questions when it manages to disa mbiguate the vague retrieval requirements. The adaptive defuzzification approach employed in the defuzzifying agent is discussed in detail. A prototype interface has been implemented to demonstrate the effectiveness.
