Improving a Strong Neural Parser with Conjunction-Specific Features
This work addresses a specific bottleneck in dependency parsing for natural language processing, but it is incremental as it builds on an existing strong parser.
The paper tackled the problem of poor performance in coordination construction for dependency parsers by extending a state-of-the-art parser with conjunction-specific features, resulting in improved 'conj' attachment and overall dependency parsing accuracy on the Stanford dependency conversion of the Penn TreeBank.
While dependency parsers reach very high overall accuracy, some dependency relations are much harder than others. In particular, dependency parsers perform poorly in coordination construction (i.e., correctly attaching the "conj" relation). We extend a state-of-the-art dependency parser with conjunction-specific features, focusing on the similarity between the conjuncts head words. Training the extended parser yields an improvement in "conj" attachment as well as in overall dependency parsing accuracy on the Stanford dependency conversion of the Penn TreeBank.