Sentence Rewriting for Semantic Parsing

Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, pages 766–777, Berlin, Germany, August 7-12, 2016. c⃝2016 Association for Computational Linguistics Sentence Rewriting for Semantic Parsing Bo Chen Le Sun Xianpei Han Bo An State Key Laboratory of Computer Sciences Institute of Software, Chinese Academy of Sciences, China. {chenbo, sunle, xianpei, anbo}@nfs.iscas.ac.cn Abstract A major challenge of semantic parsing is the vocabulary mismatch problem be- tween natural language and target ontol- ogy. In this paper, we propose a sen- tence rewriting based semantic parsing method, which can effectively resolve the mismatch problem by rewriting a sentence into a new form which has the same struc- ture with its target logical form. Specifi- cally, we propose two sentence-rewriting methods for two common types of mis- match: a dictionary-based method for 1- N mismatch and a template-based method for N-1 mismatch. We evaluate our sen- tence rewriting based semantic parser on the benchmark semantic parsing dataset – WEBQUESTIONS. Experimental results show that our system outperforms the base system with a 3.4% gain in F1, and gen- erates logical forms more accurately and parses sentences more robustly. 1 Introduction Semantic parsing is the task of mapping natu- ral language sentences into logical forms which can be executed on a knowledge base (Zelle and Mooney, 1996; Zettlemoyer and Collins, 2005; Kate and Mooney, 2006; Wong and Mooney, 2007; Lu et al., 2008; Kwiatkowksi et al., 2010). Figure 1 shows an example of semantic parsing. Semantic parsing is a fundamental technique of natural language understanding, and has been used in many applications, such as question answering (Liang et al., 2011; He et al., 2014; Zhang et al., 2016) and information extraction (Krishnamurthy and Mitchell, 2012; Choi et al., 2015; Parikh et al., 2015). Semantic parsing, however, is a challenging Sentence: What is the capital of Germany? Logical form: λx.capital(Germany,x) Result: {Berlin} (Semantic parsing) KB (Execution) Figure 1: An example of semantic parsing. task. Due to the variety of natural language ex- pressions, the same meaning can be expressed us- ing different sentences. Furthermore, because log- ical forms depend on the vocabulary of target- ontology, a sentence will be parsed into different logical forms when using different ontologies. For example, in below the two sentences s1 and s2 express the same meaning, and they both can be parsed into the two different logical forms lf1 and lf2 using different ontologies. s1 What is the population of Berlin? s2 How many people live in Berlin? lf1 λx.population(Berlin,x) lf2 count(λx.person(x)∧live(x,Berlin)) Based on the above observations, one major challenge of semantic parsing is the structural mis- match between a natural language sentence and its target logical form, which are mainly raised by the vocabulary mismatch between natural lan- guage and ontologies. Intuitively, if a sentence has the same structure with its target logical form, it is easy to get the correct parse, e.g., a semantic parser can easily parse s1 into lf1 and s2 into lf2. On the contrary, it is difficult to parse a sentence …

Content truncated to 500 words