Interpretation of Implicit Conditions in Database Search Dialogues

Proceedings of the 27th International Conference on Computational Linguistics, pages 477–486 Santa Fe, New Mexico, USA, August 20-26, 2018. 477 Interpretation of Implicit Conditions in Database Search Dialogues Shun-ya Fukunaga1 Hitoshi Nishikawa1 Takenobu Tokunaga1 Hikaru Yokono2 Tetsuro Takahashi2 1Tokyo Institute of Technology 2Fujitsu Laboratories Ltd. {fukunaga.s.ab@m, hitoshi@c, take@c}.titech.ac.jp {yokono.hikaru, takahashi.tet}@jp.fujitsu.com Abstract Targeting the database search dialogue, we propose to utilise information in the user utterances that do not directly mention the database (DB) field of the backend database system but are useful for constructing database queries. We call this kind of information implicit conditions. Interpret- ing the implicit conditions enables the dialogue system more natural and efficient in communi- cating with humans. We formalised the interpretation of the implicit conditions as classifying user utterances into the related DB field while identifying the evidence for that classification at the same time. Introducing this new task is one of the contributions of this paper. We imple- mented two models for this task: an SVM-based model and an RCNN-based model. Through the evaluation using a corpus of simulated dialogues between a real estate agent and a customer, we found that the SVM-based model showed better performance than the RCNN-based model. 1 Introduction The information that a dialogue system needs to extract from user utterances highly depends on its backend application. When being used as a natural language interface of a database system, the dialogue system should be able to extract pieces of information corresponding to the record fields of the database (DB) for constructing a query. There have been several attempts to extract this type of information from user utterances in database search dialogues. For instance, several studies (Raymond and Riccardi, 2007; Mesnil et al., 2015; Liu and Lane, 2016b) tried to extract values for the database field defined in the ATIS (The Air Travel Information System) corpus (Hemphill et al., 1990; Dahl et al., 1994) from the user utterances. The ATIS corpus includes a set of dialogues between users and an air travel system that were collected through the Wizard-of-Oz method. The tags that correspond to the backend database fields, e.g. departure city, arrival date, are annotated to the expressions in the user utterances. However, the utterances in real dialogues include information that does not always directly correspond to database fields but provides useful information for constructing database queries. It will be natural and more efficient if the dialogue system can utilise this type of information for retrieving the database. For instance, in real estate search dialogues, which is our target domain, the number of family members provides useful information for deciding the size of a house, but it is rarely a field of the real estate database as the number of family members is an attribute of the customer rather than that of houses. We call this type of information implicit conditions (Fukunaga et al., 2018). To realise a dialogue system that can utilise the implicit conditions for the database search, we need to tackle the following two tasks. (1) extracting …

Content truncated to 500 words