Application of fuzzy theory to natural language processing in interactive fuzzy decision making support system.
スポンサーリンク
概要
- 論文の詳細を見る
This paper is concerned with a natural language interface in an interactive fuzzy decision-making support system (FDSS). A natural language interface in an interactive system is very desirable because the users of the system are not required much preliminary knowledge.FDSS partly uses a natural language interface of japanese language in the case where users input their evaluated objectives for a multi-objective decision problem. When the user inputs an evaluated objective by natural language expression, FDSS can extract a keyword representing its concept from the expression. The natural language interface is designed by the following concepts: 1) the dictionary for japanese language processing is as little a scale as possible, 2) the processing for appearance of unregistered words in the dictionary is developed in the basis of fuzzy theory. The reason of this is that since the natural language interface is only one part of FDSS, the processing for it cannot be a largescale in consideration of operationality and responsibility on FDSS. Consequently unregisterd words frequently appear.An algorithm for treating the unregistered words is presented by partial character matching, similarity relation among registration words, and approximate reasoning. The algorithm was evaluated with regard to 158 real examples, which 70 subjects inputted to FDSS by free natural language; it was verified that the algorithm was very useful.
- 公益社団法人 計測自動制御学会の論文
公益社団法人 計測自動制御学会 | 論文
- Self-Excited Oscillation of Relay-Type Sampled-Data Feedback Control System
- タイトル無し
- Mold Level Control for a Continuous Casting Machine Using an Electrode-Type Mold-Level Detector
- Assessment and Control of Noise:Pollution by Noise from General Sources
- Information network system and home automation.