著者

柳井孝介 やないこうすけ Kohsuke Yanai

巻号頁・発行日

vol.1, pp.31-38, 2014-11-12

---

This paper discusses architecture for artificial intelligence (AI) that can debate problemswith humans. The proposed architecture is composed of (1) offline system which createstext annotations over large amount of text data in advance, and (2) online system which runsvarious algorithms asynchronously in real time. We built a prototype of debating AI system basedon the architecture and evaluated it over 50 discussion topics extracted from a popular debatewebsite. The AI system outputs three opinions per a discussion topic; each opinion is composedof 7 sentences. Thus, it outputs 150 opinions for 50 topics in total. We found that in 61 opinionsout of 150 the argumentations are understandable. We plan to develop a feature of combinatorialoptimization of algorithms onto the architecture as a future work.