Paper: Learning To Predict Problematic Situations In A Spoken Dialogue System: Experiments With How May I Help You?

ACL ID A00-2028
Title Learning To Predict Problematic Situations In A Spoken Dialogue System: Experiments With How May I Help You?
Venue Annual Conference of the North American Chapter of the Association for Computational Linguistics
Session Main Conference
Year 2000
Authors

Current spoken dialogue systems are deficient in their strategies for preventing, identifying and re- pairing problems that arise in the conversation. This paper reports results on learning to automatically identify and predict problematic human-computer dialogues in a corpus of 4774 dialogues collected with the How May I Help You spoken dialogue system. Our expectation is that the ability to predict prob- lematic dialogues will allow the system's dialogue manager to modify its behavior to repair problems, and even perhaps, to prevent them. We train a problematic dialogue classifier using automatically- obtainable features that can identify problematic dialogues significantly better (23%) than the base- line. A classifier trained with only automatic fea- tures from the first exchange in th...