Knowledge-Driven Slot Constraints for Goal-Oriented Dialogue Systems

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 3407–3419 June 6–11, 2021. ©2021 Association for Computational Linguistics 3407 Knowledge-Driven Slot Constraints for Goal-Oriented Dialogue Systems Piyawat Lertvittayakumjorn∗ Imperial College London pl1515@ic.ac.uk Daniele Bonadiman Amazon AI dbonadim@amazon.com Saab Mansour Amazon AI saabm@amazon.com Abstract In goal-oriented dialogue systems, users pro- vide information through slot values to achieve specific goals. Practically, some combinations of slot values can be invalid according to exter- nal knowledge. For example, a combination of “cheese pizza” (a menu item) and “oreo cook- ies” (a topping) from an input utterance “Can I order a cheese pizza with oreo cookies on top?” exemplifies such invalid combinations according to the menu of a restaurant busi- ness. Traditional dialogue systems allow exe- cution of validation rules as a post-processing step after slots have been filled which can lead to error accumulation. In this paper, we for- malize knowledge-driven slot constraints and present a new task of constraint violation de- tection accompanied with benchmarking data. Then, we propose methods to integrate the ex- ternal knowledge into the system and model constraint violation detection as an end-to-end classification task and compare it to the tra- ditional rule-based pipeline approach. Exper- iments on two domains of the MultiDoGO dataset reveal challenges of constraint viola- tion detection and sets the stage for future work and improvements. 1 Introduction Natural language understanding (NLU) is an impor- tant component of goal-oriented dialogue systems. The function of NLU is to construct a semantic frame for a user utterance by performing two tasks – intent classification (IC) and slot labelling (SL) (Chen et al., 2017). The former task aims to iden- tify the intent of the user (i.e., an activity or a trans- action that the user wants to accomplish), while the latter task extracts attributes of the intent. For example, given an input utterance “Please add one XL fries to my order” in Figure 1(A), IC classifies that the user intent is “AddToOrder” (Adding a new menu item to the order), while SL detects “one”, ∗ Work performed while at Amazon AI “XL”, and “fries” as Quantity, MenuItemSize, and MenuItem, respectively. These two tasks, IC & SL, could be performed either independently (Zhao and Wu, 2016; Haffner et al., 2003; Kurata et al., 2016) or jointly (Xu and Sarikaya, 2013; Li et al., 2018; Gupta et al., 2019) although recent research shows that training jointly generally leads to better results (Hakkani-Tür et al., 2016; Goo et al., 2018). To make the recognition of intents and slots more reliable, NLU models require the list of all possible intents and the slots associated to each intent. For instance, the intent show_flights has airline, departure_city, arrival_city, departure_date, and departure_time as its associated slots. Prac- tically, each slot has its own type. Some types are domain-agnostic such as DATE for the depar- ture_date, while other types are domain-specific, such as AIRLINE for the slot airline. We also re- fer to the …

Content truncated to 500 words