Syntactic Dependency Representations in Neural Relation Classification
This work addresses the problem of selecting effective syntactic representations for relation classification in NLP, but it is incremental as it compares existing schemes rather than introducing new methods.
The paper compared three syntactic dependency representations (CoNLL, Stanford Basic, Universal Dependencies) in neural relation classification, finding that Universal Dependencies performed best with a 2.1% F1-score improvement over syntax-agnostic baselines.
We investigate the use of different syntactic dependency representations in a neural relation classification task and compare the CoNLL, Stanford Basic and Universal Dependencies schemes. We further compare with a syntax-agnostic approach and perform an error analysis in order to gain a better understanding of the results.