Effective Measures of Domain Similarity for Parsing

8 years 11 months ago
Effective Measures of Domain Similarity for Parsing
It is well known that parsing accuracy suffers when a model is applied to out-of-domain data. It is also known that the most beneficial data to parse a given domain is data that matches the domain (Sekine, 1997; Gildea, 2001). Hence, an important task is to select appropriate domains. However, most previous work on domain adaptation relied on the implicit assumption that domains are somehow given. As more and more data is becoming available, automatic ways to select data that is beneficial for a new (unknown) target domain are becoming attractive. This paper evaluates various ways to automatically acquire related training data for a given test set. The results show that an unsupervised technique based on topic models is effective – it outperforms random data selection on both examined languages, English and Dutch. Moreover, the technique works better than manually assigned labels gathered from meta-data that is available for English.
Barbara Plank, Gertjan van Noord
Added 23 Aug 2011
Updated 23 Aug 2011
Type Journal
Year 2011
Where ACL
Authors Barbara Plank, Gertjan van Noord
Comments (0)