Generality of Spoken Dialogue System using SeGA-IL for Different Languages

K. Araki and M. Kuroda (Japan)

Keywords

Genetic algorithm, Inductive Learning, spoken dialogue processing, sexual selection, and generality

Abstract

This paper presents a Spoken Dialogue Method Using Inductive Learning Method Based on Genetic Algorithm with Sexual Selection (SeGA-ILSD) along with analysis of its generality for different languages: Japanese and English. This method achieves high generality using a learning function. Therefore, this method can show equivalent performance for different languages. To confirm it, we developed an experimental system of this method and carried out an experiment of performance evaluation of this method. These results show that this method has equivalent performance for two languages: Japanese and English. Results verify that this proposed method has high generality for these languages.

Important Links:



Go Back