Selectional constraints are limitations on the applicability of predicates to arguments. For example, the statement “The number two is blue” may be syntactically well formed, but at some level it is anomalous — BLUE is not a predicate that can be applied to numbers. According to the influential theory of (Katz and Fodor, 1964), a predicate associates a set of defining features with each argument, expressed within a restricted semantic vocabulary. Despite the persistence of this theory, however, there is widespread agreement about its empirical shortcomings (McCawley, 1968; Fodor, 1977). As an alternative, some critics of the Katz-Fodor theory (e.g. (Johnson-Laird, 1983)) have abandoned the treatment of selectional constraints as semantic, instead treating them as indistinguishable from inferences made on the basis of factual knowledge. This provides a better match for the empirical phenomena, but it opens up a different problem: if selectional constraints are the same as inferences in general, then accounting for them will require a much more complete understanding of knowledge representation and inference than we have at present. The problem, then, is this: how can a theory of selectional constraints be elaborated without first having either an empirically adequate theory of defining features or a comprehensive theory of inference? In this dissertation, I suggest that an answer to this question lies in the representation of conceptual

The automatic disambiguation of word senses has been an interest and concern since the earliest days of computer treatment of language in the 1950's. Sense disambiguation is an “intermediate task ” (Wilks and Stevenson, 1996) which is not an end in itself, but rather is necessary at one level or another to accomplish most natural language processing tasks. It is

Word sense disambiguation assumes word senses. Within the lexicography and linguistics literature, they are known to be very slippery entities. The paper looks at problems with existing accounts of `word sense' and describes the various kinds of ways in which a word's meaning can deviate from its core meaning. An analysis is presented in which word senses are abstractions from clusters of corpus citations, in accordance with current lexicographic practice. The corpus citations, not the word senses, are the basic objects in the ontology. The corpus citations will be clustered into senses according to the purposes of whoever or whatever does the clustering. In the absence of such purposes, word senses do not exist. Word sense disambiguation also needs a set of word senses to disambiguate between. In most recent work, the set has been taken from a general-purpose lexical resource, with the assumption that the lexical resource describes the word senses of English/French/. . . , between whi...

This article presents a method of word sense disambiguation that assigns a target word the sense that is most related to the senses of its neighboring words. We explore the use of measures of similarity and relatedness that are based on finding paths in a concept network, information content derived from a large corpus, and word sense glosses. We observe that measures of relatedness are useful sources of information for disambiguation, and in particular we find that two gloss based measures that we have developed are particularly flexible and e#ective measures for word sense disambiguation.

As large on-line corpora become more prewlent, a number of attempts have been made to automatically extract thesaurus-like relations directly from text using knowledge poor methods. In the absence of any specific application, comparing the results of these attempts is difficult. Here we propose an ewluation method using gold standards, i.e., pre-existing hand-compiled resources, as a means of comparing extraction techniques. Using this ewluation method, we compare two semantic extraction techniques which produce similar word lists, one using syntactic context of words , and the'other using windows of heuristiclly tagged words. The two techniques are very similar except that in one case selective natural language processing, a partial syntactic analysis, is performed. On a 4 megabyte corpus, syntactic contexts produce significantly better results against the gold standards for the most characteristic words in the corpus, while windows produce better results for rare words.

We develop a new computational model for representing the fine-grained meanings of near-synonyms and the differences between them. We also develop a sophisticated lexical-choice process that can decide which of several near-synonyms is most appropriate in a particular situation. This research has direct applications in machine translation and text generation. We first identify the problems of representing near-synonyms in a computational lexicon and show that no previous model adequately accounts for near-synonymy. We then propose a preliminary theory to account for near-synonymy, relying crucially on the notion of granularity of representation, in which the meaning of a word arises out of a context-dependent combination of a context-independent core meaning and a set of explicit differences to its near-synonyms. That is, near-synonyms cluster together. We then develop a clustered model of lexical knowledge, derived from the conventional ontological model. The model cuts off the ontology at a coarse grain, thus avoiding an awkward proliferation of language-dependent concepts in the ontology, and groups near-synonyms into subconceptual clusters that are linked to the ontology. A cluster differentiates near-synonyms in terms of fine-grained aspects of denotation, implication, expressed attitude, and style. The model is general enough to account for other types of variation, for instance, in collocational behaviour. An efficient, robust, and flexible fine-grained lexical-choice process is a consequence of a clustered model of lexical knowledge. To make it work, we formalize criteria for lexical choice as preferences to express certain concepts with varying indirectness, to express attitudes, and to establish certain styles. The lexical-choice process itself works on two tiers: between clusters and between near-synonyns of clusters. We describe our prototype implementation of the system, called I-Saurus.

We first reconsider the role of lexicographers in word-sense disambiguation as a computational task, as providers of both legacy material (dictionaries) and special test material for competitions like SENSEVAL. We suggest that the standard fine-grained division of senses and (larger) homographs by a lexicographer for use by a human reader may not be an appropriate goal for the computational WSD task. We argue that the level of sense-discrimination that NLP needs corresponds roughly to homographs, though we discuss psycholinguistic evidence that there are broad sense divisions with some etymological derivation (i.e. non-homographic) that are as distinct for humans as homographic ones and they may be part of the broad class of sensedivisions we seek to identify here. Fifteen years or more of WSD research has shown that it is this kind of discrimination that existing WSD programs are able to capture at the ~95% success level, whereas the full lexicographicallyderived division of senses seems to remain too hard for both programs and human discriminators. We link this discussion to the observation that major NLP tasks like MT and IR seem not to need independent WSD modules of the sort produced in the research field, even though they are undoubtedly doing WSD by other means. Our conclusion is that WSD should continue to focus on these broad discriminations, at which it can do very well, thereby possibly offering the close-to-100% success that IR needs (especially search-engine, rather than classic long-query) IR, and assume that this is what most NLP requires, with the possible exception of very fine questions of target word choice in MT. This proposal can be seen as reorienting WSD to what it can actually perform at the standard success levels, but we argue that this, rather...

The task of choosing between lexical near-equivalents in text generation requires the kind of knowledge of fine differences between words that is typified by the usage notes of dictionaries and books of synonym discrimination. These usage notes follow a fairly standard pattern, and a study of their form and content shows the kinds of differentiae adduced in the discrimination of nearsynonyms. For appropriate lexical choice in text generation and machine translation systems, it is necessary to develop the concept of formal `computational usage notes', which would be part of the lexical entries in a conceptual knowledge base. The construction of a set of `computational usage notes' adequate for text generation is a major lexicographic task of the future. 1 Lexicons for lexical choice 1.1 Lexical choice and plesionymy The problem of lexical choice in text generation is to determine the word that conveys most precisely the denotation and connotation that are to be expressed. The required...

Abstract. Prepositional Phrase (PP) attachment can be addressed by considering frequency counts of dependency triples seen in a non-annotated corpus. However, not all triples appear even in very big corpora. To solve this problem, several techniques have been used. We evaluate two different backoff methods, one based on WordNet and the other on a distributional (automatically created) thesaurus. We work on Spanish. The thesaurus is created using the dependency triples found in the same corpus used for counting the frequency of unambiguous triples. The training corpus used for both methods is an encyclopaedia. The method based on a distributional thesaurus has higher coverage but lower precision than the WordNet method. 1

The paper compares the tasks of part-of-speech (POS) tagging and word-sense-tagging or disambiguation (WSD), and argues that the tasks are not related by fineness of grain or anything like that, but are quite different kinds of task, particularly because there is nothing in POS corresponding to sense novelty. The paper also argues for the reintegration of sub-tasks that are being separated for evaluation.