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

As one attack on the "knowledge acquisition bottleneck", we are attempting to exploit a largely untapped source of general knowledge in texts, lying at a level beneath the explicit assertional content. This knowledge consists of relationships implied to be possible in the world, or, under certain conditions, implied to be normal or commonplace in the world. The goal of the work reported is to derive such general world knowledge (initially, from Penn Treebank corpora) in two stages: first, we derive general "possibilistic" propositions from noun phrases and clauses; then we try to derive stronger generalizations, based on the nature and statistical distribution of the possibilistic claims obtained in the first phase. Here we report preliminary results of the first phase, which indicate the feasibility of our project, and its likely limitations.

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

We have been developing techniques for extracting general world knowledge from miscellaneous texts by a process of approximate interpretation and abstraction, focusing initially on the Brown corpus. We apply interpretive rules to clausal patterns and patterns of modification, and concurrently abstract general "possibilistic" propositions from the resulting formulas. Two examples

Ontologies in current computer science parlance are computer based resources that represent agreed domain semantics. This paper first introduces ontologies in general and subsequently, in particular, shortly outlines the DOGMA ontology engineering approach that separates "atomic" conceptual relations from "predicative" domain rules. In the main part of the paper, we describe and experimentally evaluate work in progress on a potential method to automatically derive the atomic conceptual relations mentioned above from a corpus of English medical texts. Preliminary outcomes are presented based on the clustering of nouns and compound nouns according to co-occurrence frequencies in the subject-verbobject syntactic context.

. Two kinds of systems have been defined during the long history of WSD: principled systems that define which knowledge types are useful for WSD, and robust systems that use the information sources at hand, such as, dictionaries, light-weight ontologies or hand-tagged corpora. This paper tries to systematize the relation between desired knowledge types and actual information sources. We also compare the results for a wide range of algorithms that have been evaluated on a common test setting in our research group. We hope that this analysis will help change the shift from systems based on information sources to systems based on knowledge sources. This study might also shed some light on semi-automatic acquisition of desired knowledge types from existing resources. 1

Selectional preferences have a long history in both generative and computational linguistics. However, since the publication of Resnik's dissertation in 1993, a new approach has surfaced in the computational linguistics community. This new line of research combines knowledge represented in a pre-defined semantic class hierarchy with statistical tools including information theory, statistical modeling, and Bayesian inference. These tools are used to learn selectional preferences from examples in a corpus. Instead of simple sets of semantic classes, selectional preferences are viewed as probability distributions over various entities. We survey research that extends Resnik's initial work, discuss the strengths and weaknesses of each approach, and show how they together form a cohesive line of research.

The context of this paper is the application of unsupervised Machine Learning techniques to building ontology extraction tools for Natural Language Processing. Our method relies on exploiting large amounts of linguistically annotated text, and on linguistic concepts such as selectional restrictions and co-composition.

One of the main obstacles to highperformance Word Sense Disambiguation (WSD) is the knowledge acquisition bottleneck. In this paper, we present a methodology to automatically extend WordNet with large amounts of semantic relations from an encyclopedic resource, namely Wikipedia. We show that, when provided with a vast amount of high-quality semantic relations, simple knowledge-lean disambiguation algorithms compete with state-of-the-art supervised WSD systems in a coarse-grained all-words setting and outperform them on gold-standard domain-specific datasets. 1