Computer-Generated Conversation Based on Newspaper Headline Interpretation
While some robot-related research has been done on conversional text, based on greetings and other general types of conversation and utilizing extensive conversation knowledge bases; to date, very little work on conversation involving topical newspaper articles has been done. If robots could engage in conversation on subjects pertaining to current affairs, it is likely they would seem much more appealing and approachable. However, if a robot were to merely read out news mechanically it would lack the conversational flexibility to be an effective communication tool, and the approachability of the robot would be significantly limited. In light of this, the present study investigates the automatic generation of highly expandable conversation sentences using cues taken from news stories.
Unable to display preview. Download preview PDF.
- 1.Sekine, S., Inui, K., Torisawa, K.: Corpus Based Knowledge Engineering, http://nlp.cs.nyu.edu/sekine/CBKE/
- 2.Kurohashi, S., Nagao, M.: A method of case structure analysis for Japanese sentences based on examples in case frame dictionary. IEICE Transactions on Information and Systems E77-D(2), 227–239 (1994)Google Scholar
- 3.Dohsaka, K., Shimaze, A.: A model for incremental utterance production in task-oriented dialogues. Transactions of Information Processing Society of Japan 37(12), 2190–2200 (1996)Google Scholar
- 4.Mainichi.jp – Mainichi Newspaper, http://www.mainichi.jp/
- 5.Kawahara, D., Kurohashi, S.: Case Frame Compilation from the Web using High-Performance Computing. In: Proceedings of the 5th International Conference on Language Resources and Evaluation, pp. 1344–1347 (2006)Google Scholar
- 6.Watabe, H., Kawaoka, T.: The Degree of Association between Concepts using the Chain of Concepts. In: Proc. of SMC 2001, pp. 877–881 (2001)Google Scholar