Language generation for conversational agent by evolution of plan trees with genetic programming

As dialogue systems are widely demanded the research on natural language generation in dialogue has raised interest. Contrary to conventional dialogue systems that reply to the user with a set of predefined answers, a newly developed dialogue system generates them dynamically and trains answers to support more flexible and customized dialogues with humans. The paper proposes an evolutionary method for generating sentences using interactive genetic programming. Sentence plan trees, which stand for the sentence structure, are adopted as the representation of genetic programming. With interactive evolution process with the user, a set of customized sentence structures is obtained. The proposed method applies to a dialogue-based travel planning system and the usability test demonstrates the usefulness of the proposed method.