Sparse Single-Hidden Layer Feedforward Network for Mapping Natural Language Questions to SQL Queries
Mapping natural language (NL) statements into SQL queries allows users to interact with systems through everyday language. Semantic parsing has seen a growing interest over the past decades. In this paper, we extend single hidden layer feedforward network (SLFN) by adding the Kullback-Liebler (KL) divergence parameter to its objective function. We refer to this algorithm as Sparse SLFN (S-SLFN) which can learn whether an SQL query answers a particular NL question. With Bag of Words (BoW) representing the questions and the queries, the algorithm, by enforcing sparsity, is meant to retain robust features representing informative relationships and structure of the data. Experimental results show that S-SLFN outperforms SLFN and other algorithms for the GeoQueries dataset by a respectable margin.
KeywordsSingle-hidden Layer Feedforward Network (SLFN) Sparsity Semantic Parsing
Unable to display preview. Download preview PDF.
- 1.Jones, B.K., Johnson, M., Goldwater, S.: Semantic parsing with bayesian tree transducers. In: Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Long Papers, vol. 1, pp. 488–496. Association for Computational Linguistics (2012)Google Scholar
- 2.Jones, B., Andreas, J., Bauer, D., Hermann, K.M., Knight, K.: Semantics-based machine translation with hyperedge replacement grammars. In: COLING, pp. 1359–1376 (2012)Google Scholar
- 3.Poon, H., Domingos, P.: Deep learning for semantic parsingGoogle Scholar
- 5.Ng, A.: Sparse autoencoderGoogle Scholar
- 7.Hosmer Jr., D.W., Lemeshow, S., Sturdivant, R.X.: Applied logistic regression. Wiley. com (2013)Google Scholar
- 9.Wallach, H.M.: Topic modeling: beyond bag-of-words. In: Proceedings of the 23rd International Conference on Machine Learning, pp. 977–984. ACM (2006)Google Scholar