SAIL-GRS: Grammar Induction for Spoken Dialogue Systems using CF-IRF Rule Similarity

Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), pages 508–511, Dublin, Ireland, August 23-24, 2014. SAIL-GRS: Grammar Induction for Spoken Dialogue Systems using CF-IRF Rule Similarity Kalliopi Zervanou, Nikolaos Malandrakis and Shrikanth Narayanan Signal Analysis and Interpretation Laboratory (SAIL), University of Southern California, Los Angeles, CA 90089, USA kzervanou@gmail.com, malandra@usc.edu , shri@sipi.usc.edu Abstract The SAIL-GRS system is based on a widely used approach originating from in- formation retrieval and document index- ing, the TF-IDF measure. In this im- plementation for spoken dialogue system grammar induction, rule constituent fre- quency and inverse rule frequency mea- sures are used for estimating lexical and semantic similarity of candidate grammar rules to a seed set of rule pattern instances. The performance of the system is evalu- ated for the English language in three dif- ferent domains, travel, tourism and finance and in the travel domain, for Greek. The simplicity of our approach makes it quite easy and fast to implement irrespective of language and domain. The results show that the SAIL-GRS system performs quite well in all three domains and in both lan- guages. 1 Introduction Spoken dialogue systems typically rely on gram- mars which define the semantic frames and re- spective fillers in dialogue scenarios (Chen et al., 2013). Such systems are tailored for specific domains for which the respective grammars are mostly manually developed (Ward, 1990; Seneff, 1992). In order to address this issue, numerous current approaches attempt to infer these grammar rules automatically (Pargellis et al., 2001; Meng and Siu, 2002; Yoshino et al., 2011; Chen et al., 2013). The acquisition of grammar rules for spoken language systems is defined as a task comprising This work is licensed under a Creative Commons Attribution 4.0 International Licence. Page numbers and proceedings footer are added by the organisers. Licence details: http: //creativecommons.org/licenses/by/4.0/ of two subtasks (Meng and Siu, 2002; Iosif and Potamianos, 2007), the acquisition of: (i) Low-level rules These are rules defining domain-specific entities, such as names of lo- cations, hotels, airports, e.g. CountryName: “USA”, Date: “July 15th, 2014”, CardType: “VISA” and other common domain multi-word ex- pressions, e.g. DoYouKnowQ: “do you know”. (ii) High-level rules These are larger, frame-like rule patterns which contain as semantic slot fillers multi-word entities identified by low-level rules. For exam- ple: DirectionsQ: “<DoYouKnowQ> <where> the <MuseumName> is lo- cated”, ExpressionCardProblem: “my <CardType> has expired”. The shared task of Grammar Induction for Spo- ken Dialogue Systems, where our system partic- ipated, focused on the induction of high-level grammar rules and in particular on the identifica- tion and semantic classification of new rule pat- terns based on their semantic similarity to known rule instances. Within this research framework, the work de- scribed in this paper proposes a methodology for estimating rule semantic similarity using a varia- tion of the well-known measure of TF-IDF as rule constituent frequency vs. inverse rule fre- quency, henceforth CF-IRF. In the remainder of this paper, we start in Sec- tion 2 by a detailed description of our system. Sub- sequently, …

Content truncated to 500 words