- (Palmer, 2000) ⇒ Martha Palmer. (2000). "Consistent Criteria for Sense Distinctions." In: Computers and the Humanities, 34(1-2).
Subject Headings: Word Sense
- It analyzes Verb Polymemy.
- It analyzes the effectiveness of WordNet's Verb Sense divisions.
- For example WordNet separate senses of "cut" cutting grain, trees, hair, and for separating into pieces of a concrete object.
- This paper specifically addresses the question of polysemy with respect to verbs, and whether or not the sense distinctions that are made in on-line lexical resources such as WordNet are appropriate for computational lexicons. The use of sets of related syntactic frames and verb classes are examined as a means of simplifying the task of defining different senses, and the importance of concrete criteria such as different predicate argument structures, semantic class constraints and lexical co-occurrences is emphasized.
- The difficulty of achieving adequate hand-crafted semantic representations has limited the field of natural language processing to applications that can be contained within well-defined subdomains. The only escape from this limitation will be through the use of automated or semi-automated methods of lexical acquisition. However, the field has yet to develop a clear consensus on guidelines for a computational lexicon that could provide a springboard for such methods, in spite of all of the effort on different lexicon development approaches (Mel’cuk, 1988; Pustejovsky, 1991; Nirenburg et al., 1992; Copestake and Sanfilippo, 1993; Lowe et al., 1997; Dorr, 1997). One of the most controversial areas has to do with polysemy. What constitutes a clear separation into senses for any one verb or noun, and how can these senses be computationally characterized and distinguished? The answer to this question is the key to breaking the bottleneck of semantic representation that is currently the single greatest limitation on the general application of natural language processing techniques.
- In this paper we specifically address the question of polysemy with respect to verbs, and whether or not the sense distinctions that are made in on-line dictionary resources such as WordNet (Miller, 1990; Miller and Fellbaum, 1991), are appropriate for computational lexicons. We examine the use of sets of related syntactic frames and verb classes as a means of simplifying the task of defining different senses, and we focus on the mismatches between these types of distinctions and some of the distinctions that occur in WordNet.
- It has been suggested that WordNet sense distinctions are too fine-grained and coarser senses are needed to drive the word sense disambiguation task. For instance, in defining cut, WordNet distinguishes between WN1, separating into pieces of a concrete object, WN29, cutting grain, WN30, cutting trees, and WN33, cutting hair. For many purposes, the three more specialized senses, WN29, WN30 and WN33, which all involve separation into pieces of concrete objects could be collapsed into the more coarse-grained WN1. However, when searching for articles on recent changes in hair styles, the more fine-grained WN33 would still be useful. Computational lexicons actually lend themselves readily to moving back and forth between elements of an hierarchical representation based on concrete criteria, and this type of structuring should become more prevalent. The point is that they operate most effectively in the realm of concrete criteria for sense distinctions, such as changes in argument structure, changes in sets of syntactic frames and/or changes in semantic class constraints, and lexical co-occurrences. Distinctions that are based on world knowledge, no matter how diverse, are much more problematic. We must bear this in mind in order to design a word sense disambiguation task that will also encourage rational, incremental development of computational lexicons.,
|2000 ConsistentCriteriaForSenseDistinctions||Consistent Criteria for Sense Distinctions||Computers and the Humanities||http://www.coli.uni-saarland.de/~schulte/Teaching/ESSLLI-06/Referenzen/Senses/palmer-2000.pdf||2000|