Detecting Multiword Verbs in the English Sublanguage of MEDLINE Abstracts

Detecting Multiword Verbs in the English Sublanguage of MEDLINE Abstracts Chun Xiao and Dietmar R¨osner Institut f¨ur Wissens- und Sprachverarbeitung Otto-von-Guericke Universit¨at Magdeburg Universit¨atsplatz 2, Magdeburg, Germany 39106 xiao|roesner@iws.cs.uni-magdeburg.de Abstract In this paper, we investigate the multiword verbs in the English sublanguage of MED- LINE abstracts. Based on the integration of the domain-specific named entity knowledge and syntactic as well as statistical information, this work mainly focuses on how to evaluate a proper multiword verb candidate. Our results present a sound balance between the low- and high-frequency multiword verb candidates in the sublanguage corpus. We get a F-measure of 0.753, when tested on a manual sample subset consisting of multiword candidates with both low- and high-frequencies. 1 Introduction During the construction of an information extraction (IE) system in the biomedical domain, we found that not only the task of named entity recognition (NER), but also the appropriate handling of verbs in this domain plays an important role. It is very helpful to determine the domain-specific verbs in a specific domain when extracting useful information, because the domain-specific verbs construct seman- tic relations between named entities (NEs). How- ever, three problems in the handling of verbs in a specific domain are still open: The first problem is how to determine domain- specific verbs. This problem has not received enough notice from most of the researchers yet. Ac- tually, domain-specific verbs have been mentioned quite often in biomedical text processing (Thomas et al., 2000; Ono et al., 2001; Xiao and R¨osner, 2004b), but often referred to a set of manually or experientially selected verbs. Spasi´c et al. (2003) briefly presented a method to find domain-specific verbs by filtering the verbs in a stoplist, at the same time, using the co-occurrence of a verb and specific terms in the text. In our experiment, the domain- specific verbs are determined through the compari- son between different corpora in different domains, or through genre analysis of the sublanguage domi- nated corpus. The second problem is how to determine multi- word verbs (MWVs). Here we do not make differences between the more detailed classifi- cation of multiword verbs, especially the verb- particle constructions and verb-preposition con- structions (Baldwin and Villavicencio, 2002). As a subcategory of multiword expressions (Sag et al., 2002), MWVs raise the complexity of our process- ing. Because some MWVs share the same verb head but lead to different semantic interpretations, like result in and result from, considering only verb heads in the processing is obviously not sufficient. A good IE system should deal with such MWVs au- tomatically and appropriately. The third problem is that there is a need to in- vestigate the inflectional and derivational forms of the verbs. An IE system may have to deal with a set of patterns, in which the inflectional and deriva- tional forms of the verbs should be taken into ac- count. For example, in biomedical texts, the verb interact defines a binary relation between two sub- stances, whereas its nominalization morpheme in a pattern such …

Content truncated to 500 words