A Benchmark of Rule-Based and Neural Coreference Resolution in Dutch Novels and News
This work provides insights into domain-specific coreference resolution for Dutch, addressing challenges in NLP applications for literature and news, but it is incremental as it applies existing methods to new data.
The paper evaluated rule-based and neural coreference resolution systems on Dutch datasets from novels and news/Wikipedia, finding that the neural system performed best on news/Wikipedia text while the rule-based system excelled on literature, with performance influenced by factors like training data and document length.
We evaluate a rule-based (Lee et al., 2013) and neural (Lee et al., 2018) coreference system on Dutch datasets of two domains: literary novels and news/Wikipedia text. The results provide insight into the relative strengths of data-driven and knowledge-driven systems, as well as the influence of domain, document length, and annotation schemes. The neural system performs best on news/Wikipedia text, while the rule-based system performs best on literature. The neural system shows weaknesses with limited training data and long documents, while the rule-based system is affected by annotation differences. The code and models used in this paper are available at https://github.com/andreasvc/crac2020