Automatic Generation of High Quality CCGbanks for Parser Domain Adaptation
This addresses domain adaptation for CCG parsing, enabling better performance in specialized domains like speech and math, but it is incremental as it builds on existing parser architectures.
The paper tackles domain adaptation for CCG parsing by automatically generating CCG corpora from dependency trees, resulting in significant performance gains for an off-the-shelf parser, improving from 90.7% to 96.6% on speech conversation and from 88.5% to 96.8% on math problems.
We propose a new domain adaptation method for Combinatory Categorial Grammar (CCG) parsing, based on the idea of automatic generation of CCG corpora exploiting cheaper resources of dependency trees. Our solution is conceptually simple, and not relying on a specific parser architecture, making it applicable to the current best-performing parsers. We conduct extensive parsing experiments with detailed discussion; on top of existing benchmark datasets on (1) biomedical texts and (2) question sentences, we create experimental datasets of (3) speech conversation and (4) math problems. When applied to the proposed method, an off-the-shelf CCG parser shows significant performance gains, improving from 90.7% to 96.6% on speech conversation, and from 88.5% to 96.8% on math problems.