Learning to Map Sentences to Logical Form: Structured Classification with Probabilistic Categorial Grammars

This paper addresses the problem of mapping natural language sentences to-calculus encodings of their meaning. We describe a learning algorithm takes as input a training set of sentences labeled with expressions in the calculus. The algorithm induces a grammar for the problem, along with a-linear model that represents a distribution over syntactic and semantic conditioned on the input sentence. We apply the method to the task of natural language interfaces to databases and show that the learned outperform previous methods in two benchmark database domains.