Semantics-based Question Generation and Implementation

This paper presents a question generation system based on the approach of semantic rewriting. State-of-the-art deep linguistic parsing and generation tools are employed to map natural language sentences into their meaning representations in the form of Minimal Recursion Semantics (mrs) and vice versa. By carefully operating on the semantic structures, we obtain a principled way of generating questions which avoids ad-hoc manipulation of syntactic structures. Based on the (partial) understanding of the sentence meaning, the system generates questions that are semantically grounded and purposeful. As the generator uses a deep linguistic grammar, the grammaticality of the generation results is licensed by the grammar. With a specialized ranking model, the linguistic realizations from the general purpose generation model are further refined for the question generation task. The evaluation results from QGSTEC2010 show promising results for the proposed approach.