This article presents a measure of semantic similarityinanis-a taxonomy based on the notion of shared information content. Experimental evaluation against a benchmark set of human similarity judgments demonstrates that the measure performs better than the traditional edge-counting approach. The article presents algorithms that take advantage of taxonomic similarity in resolving syntactic and semantic ambiguity, along with experimental results demonstrating their e#ectiveness. 1. Introduction Evaluating semantic relatedness using network representations is a problem with a long history in arti#cial intelligence and psychology, dating back to the spreading activation approach of Quillian #1968# and Collins and Loftus #1975#. Semantic similarity represents a special case of semantic relatedness: for example, cars and gasoline would seem to be more closely related than, say, cars and bicycles, but the latter pair are certainly more similar. Rada et al. #Rada, Mili, Bicknell, & Blett...

This article shows that the Web can be employed to obtain frequencies for bigrams that are unseen in a given corpus. We describe a method for retrieving counts for adjective-noun, noun-noun, and verb-object bigrams from the Web by querying a search engine. We evaluate this method by demonstrating: (a) a high correlation between Web frequencies and corpus frequencies; (b) a reliable correlation between Web frequencies and plausibility judgments; (c) a reliable correlation between Web frequencies and frequencies recreated using class-based smoothing; (d) a good performance of Web frequencies in a pseudodisambiguation task. 1.

This article concerns the estimation of a particular kind of probability, namely, the probability of a noun sense appearing as a particular argument of a predicate. In order to overcome the accompanying sparse-data problem, the proposal here is to define the probabilities in terms of senses from a semantic hierarchy and exploit the fact that the senses can be grouped into classes consisting of semantically similar senses. There is a particular focus on the problem of how to determine a suitable class for a given sense, or, alternatively, how to determine a suitable level of generalization in the hierarchy. A procedure is developed that uses a chi-square test to determine a suitable level of generalization. In order to test the performance of the estimation method, a pseudo-disambiguation task is used, together with two alternative estimation methods. Each method uses a different generalization procedure; the first alternative uses the minimum description length principle, and the second uses Resnik's measure of selectional preference. In addition, the performance of our method is investigated using both the standard Pearson chisquare statistic and the log-likelihood chi-square statistic

Manual development of large subcategorised lexicons has proved difficult because predicates change behaviour between sublanguages, domains and over time. Yet access to a comprehensive subcategorization lexicon is vital for successful parsing capable of recovering predicate-argument relations, and probabilistic parsers would greatly benefit from accurate information concerning the relative likelihood of different subcategorisation frames (scfs) of a given predicate. Acquisition of subcategorization lexicons from textual corpora has recently become increasingly popular. Although this work has met with some success, resulting lexicons indicate a need for greater accuracy. One significant source of error lies in the statistical filtering used for hypothesis selection, i.e. for removing noise from automatically acquired scfs. This thesis builds on earlier work in verbal subcategorization acquisition, taking as a starting point the problem with statistical filtering. Our investigation shows that statistical filters tend to work poorly because not only is the underlying distribution zipfian, but there is also very little correlation between conditional distribution of

We address the problem of clustering words (or con-structing a thesaurus) based on co-occurrence data, and using the acquired word classes to improve the accuracy of syntactic disambiguation. We view this problem as that of estimating a joint probability dis-tribution specifying the joint probabilities of word pairs, such as noun verb pairs. We propose an effi-cient algorithm based on the Minimum Description Length (MDL) principle for estimating such a prob-ability distribution. Our method is a natural ex-tension of those proposed in (Brown et al., 1992) and (Li and Abe, 1996), and overcomes their draw-backs while retaining their advantages. We then coinbined this clustering method with the disam-I)iguation method of (Li and Abe, 1995) to derive a disambiguation method that makes use of both auto-matically constructed thesauruses and a hand-made thesaurus. The overall disambiguation accuracy achieved by our method is 85.2%, which compares favorably against the accuracy (82.4%) obtained by the state-of-the-art disambiguation method of (Brill and Resnik, 1994). 1

The computation of selectional preferences, the admissible argument values for a relation, is a well-known NLP task with broad applicability. We present LDA-SP, which utilizes LinkLDA (Erosheva et al., 2004) to model selectional preferences. By simultaneously inferring latent topics and topic distributions over relations, LDA-SP combines the benefits of previous approaches: like traditional classbased approaches, it produces humaninterpretable classes describing each relation’s preferences, but it is competitive with non-class-based methods in predictive power. We compare LDA-SP to several state-ofthe-art methods achieving an 85 % increase in recall at 0.9 precision over mutual information (Erk, 2007). We also evaluate LDA-SP’s effectiveness at filtering improper applications of inference rules, where we show substantial improvement over Pantel et al.’s system (Pantel et al., 2007). 1

This paper describes an unsupervised algorithm for placing unknown words into a taxonomy and evaluates its accuracy on a large and varied sample of words. The algorithm works by first using a large corpus to find semantic neighbors of the unknown word, which we accomplish by combining latent semantic analysis with part-of-speech information. We then place the unknown word in the part of the taxonomy where these neighbors are most concentrated, using a class-labelling algorithm developed especially for this task. This method is used to reconstruct parts of the existing WordNet database, obtaining results for common nouns, proper nouns and verbs. We evaluate the contribution made by part-of-speech tagging and show that automatic filtering using the class-labelling algorithm gives a fourfold improvement in accuracy.

Techniques that exploit knowledge of distributional similarity between words have been proposed in many areas of Natural Language Processing. For example, in language modeling, the sparse data problem can be alleviated by estimating the probabilities of unseen co-occurrences of events from the probabilities of seen co-occurrences of similar events. In other applications, distributional similarity is taken to be an approximation to semantic similarity. However, due to the wide range of potential applications and the lack of a strict definition of the concept of distributional similarity, many methods of calculating distributional similarity have been proposed or adopted. In this work, a flexible, parameterized framework for calculating distributional similarity is proposed. Within this framework, the problem of finding distributionally similar words is cast as one of co-occurrence retrieval (CR) for which precision and recall can be measured by analogy with the way they are measured in document retrieval. As will be shown, a number of popular existing measures of distributional similarity are simulated with parameter settings within the CR framework. In this article, the CR framework is then used to systematically investigate three fundamental questions concerning distributional similarity. First, is the relationship of lexical similarity necessarily symmetric, or are there advantages to be gained from considering it as an asymmetric relationship? Second, are some co-occurrences inherently more salient than others in the calculation of distributional similarity? Third, is it necessary to consider the difference in the extent to which each word occurs in each co-occurrence type? Two application-based tasks are used for evaluation: automatic thesaurus generation and pseudo-disambiguation. It is possible to achieve significantly better results on both these tasks by varying the parameters within the CR framework rather than using other existing distributional similarity measures; it will also be shown that any single unparameterized measure is unlikely to be able to do better on both tasks. This is due to an inherent asymmetry in lexical substitutability and therefore also in lexical distributional similarity. 1.

Word sense disambiguation (WSD) is the ability to identify the meaning of words in context in a computational manner. WSD is considered an AI-complete problem, that is, a task whose solution is at least as hard as the most difficult problems in artificial intelligence. We introduce the reader to the motivations for solving the ambiguity of words and provide a description of the task. We overview supervised, unsupervised, and knowledge-based approaches. The assessment of WSD systems is discussed in the context of the Senseval/Semeval campaigns, aiming at the objective evaluation of systems participating in several different disambiguation tasks. Finally, applications, open problems, and future directions are discussed.

This article is aimed at quantifying the disambiguation performance of automatically acquired selectional preferences in regard to nouns, verbs, and adjectives with respect to a standard test corpus and evaluation setup (SENSEVAL-2) and to identify strengths and weaknesses. Although there is clearly a limit to coverage using preferences alone, because preferences are acquired only with respect to speci#c grammatical roles, we show that when dealing with running text, rather than isolated examples, coverage can be increased at little cost in accuracy by using the one-sense-per-discourse heuristic