On Parsing as Tagging
This work provides a systematic analysis for researchers in natural language processing, but it is incremental as it builds on and compares existing tagging methods.
The paper tackles the problem of reducing constituency parsing to tagging by unifying existing approaches into a three-step pipeline and evaluating it empirically. It finds that the linearization and alignment of derivation trees with input sequences are the most critical factors for accuracy, with results tested on English and 8 diverse languages.
There have been many proposals to reduce constituency parsing to tagging in the literature. To better understand what these approaches have in common, we cast several existing proposals into a unifying pipeline consisting of three steps: linearization, learning, and decoding. In particular, we show how to reduce tetratagging, a state-of-the-art constituency tagger, to shift--reduce parsing by performing a right-corner transformation on the grammar and making a specific independence assumption. Furthermore, we empirically evaluate our taxonomy of tagging pipelines with different choices of linearizers, learners, and decoders. Based on the results in English and a set of 8 typologically diverse languages, we conclude that the linearization of the derivation tree and its alignment with the input sequence is the most critical factor in achieving accurate taggers.