and improving intercoder reliability in discourse tagging using statistical techniques. Biascorrected tags axe formulated and successfully used to guide a revision of the coding manual and develop an automatic classifier.

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.

We describe an algorithm for Word Sense Disambiguation (WSD) that relies on a lazy learner improved with automatic feature selection. The algorithm was implemented in a system that achieves excellent performance on the set of data released during the senseval-2 competition. We present the results obtained and discuss the performance of various features in the context of supervised learning algorithms for WSD.

This paper reports on experiments in classifying the semantic role annotations assigned to prepositional phrases in both the PENN TREEBANK (version II) and FRAMENET (version 0.75). This task can be viewed as word-sense disambiguation, treating the semantic roles of prepositional phrases as word senses for the associated preposition. Three sets of experiments are done: one evaluates crossfold validation over the TREEBANK role annotations; another does the same for the FRAMENET role annotations; the last evaluates the applicability of lexical associations across datasets. Each set of experiments compares the use of traditional lexical associations (i.e., collocations) versus class-based lexical associations using WordNet synsets. The latter generalize better to handle unknown datasets.

This paper reports on experiments in classifying the semantic role annotations assigned to prepositional phrases in both the Penn Treebank version II and FrameNet. This task can be viewed as word-sense disambiguation, treating the semantic roles of prepositional phrases as word senses for the associated preposition. Three sets of experiments are done: one set evaluates cross-fold validation over Treebank; another does the same for FrameNet; the final set evaluates the transfer of lexical associations from one dataset to the other. Each set of experiments compares the use of traditional lexical associations (i.e., collocations) versus classbased lexical associations using WordNet synsets. The latter type better facilitates the transfer of associations across datasets.

A context-based spelling error is a spelling or typing error that turns an intended word into another word of the language. For example, the intended word "sight" might become the word "site." A spell checker cannot identify such an error. In the English language---the case of interest here---a syntax checker may also fail to catch such an error since, among other reasons, the parts-of-speech of an erroneous word may permit an acceptable parsing. This chapter presents an effective method called Ltest for identifying the majority of context-based spelling errors. Ltest learns from

This paper reports on experiments in classifying the semantic role annotations assigned to prepositional phrases in both the PENN TREEBANK and FRAMENET.In both cases, experiments are done to see how the prepositions can be classified given the dataset's role inventory, using standard word-sense disambiguation features.

This paper presents a new approach for word prediction problem. Word prediction is a natural language processing problem that tries to predict the correct word in a given context. Word completion utilities, writing aids, and language translation are among the most common applications of word prediction. In this paper, we describe a new method to predict the correct word given its context. A data mining tool is used as a classification mean to predict the correct word in the given context. The method has been implemented; the testing results are promising. The approach requires a very small training text size compared with similar methods to produce an accuracy that approaches 93% correct predictions.

This article describes how semantic role resources can be exploited for preposition disambiguation. The main resources include the semantic role annotations provided by the Penn Treebank and FrameNet tagged corpora. The resources also include the assertions contained in the Factotum knowledge base, as well as information from Cyc and Conceptual Graphs. A common inventory is derived from these in support of definition analysis, which is the motivation for this work. The disambiguation concentrates on relations indicated by prepositional phrases, and is framed as word-sense disambiguation for the preposition in question. A new type of feature for word-sense disambiguation is introduced, using WordNet hypernyms as collocations rather than just words. Various experiments over the Penn Treebank and FrameNet data are presented, including prepositions classified separately versus together, and illustrating the effects of filtering. Similar experimentation is done over the Factotum data, including a method for inferring likely preposition usage from corpora, as knowledge bases do not generally indicate how relationships are expressed in English (in contrast to the explicit annotations on this in the Penn Treebank and FrameNet). Other experiments are included with the FrameNet data mapped into the common relation inventory developed for definition analysis, illustrating how preposition disambiguation might be applied in lexical acquisition. 1.

This thesis discusses the automatic acquisition of conceptual distinctions using empirical methods, with an emphasis on semantic relations. The goal is to improve semantic lexicons for computational linguistics, but the work can be applied to general-purpose knowledge bases as well.