We investigate one technique to produce a summary of an original text without requiring its full semantic interpretation, but instead relying on a model of the topic progression in the text derived from lexical chains. We present a new algorithm to compute lexical chains in a text, merging several robust knowledge sources: the WordNet thesaurus, a part-of-speech tagger and shallow parser for the ldentification of nominal groups, and a segmentation algorithm derived from (Hearst, 1994) Summarization proceeds in three steps: the original text m first segmented, lexical chains are constructed, strong chains are identified and significant sentences are extracted from the text. We present in this paper empirical results on the identification of strong chain and of significant sentences.

We introduce a memory-based approach to part of speech tagging. Memory-based learning is a form of supervised learning based on similarity-based reasoning. The part of speech tag of a word in a particular context is extrapolated from the most similar cases held in memory. Supervised learning approaches are useful when a tagged corpus is available as an example of the desired output of the tagger. Based on such a corpus, the tagger-generator automatically builds a tagger which is able to tag new text the same way, diminishing development time for the construction of a tagger considerably. Memory-based tagging shares this advantage with other statistical or machine learning approaches. Additional advantages specific to a memory-based approach include (i) the relatively small tagged corpus size sufficient for training, (ii) incremental learning, (iii) explanation capabilities, (iv) flexible integration of information in case representations, (v) its non-parametric nature, (vi) reasonably good results on unknown words without morphological analysis, and (vii) fast learning and tagging. In this paper we show that a large-scale application of the memory-based approach is feasible: we obtain a tagging accuracy that is on a par with that of known statistical approaches, ad with attractive space and time complexity properties when using IGTree, a tree-based formalism for indexing and searching huge case bases. The use of IGTree has as additional advantage that optimal context size for disambiguation is dynamically computed.

We propose methods for unsupervised learning of text profiles for music from unstructured text obtained from the web. The profiles can be used for classification, recommendation, and understanding, and may be used in conjunction with existing methods such as audio analysis and collaborative filtering to improve performance. A formal method for analyzing the quality of the learned profiles is given, and results indicate that they perform well when used to find similar artists. 1

Many ofthe processing steps in natural language engineering can be performed using nite state transducers. An optimal way tocreate such transducers is to compile them from regular expressions. This paper is an introduction to the regular expression calculus, extended with certain operators that have proved very useful in natural language applications ranging from tokenization to light parsing. The examples in the paper illustrate in concrete detail some of these applications. 1

We present a memory-based learning (MBL) approach to shallow parsing in which POS tagging, chunking, and identification of syntactic relations are formulated as nemory-based modules. The experiments reported in this paper show competitive results, the Fa= for the Wall Street Journal (WSJ) treebank is: 93.8% for NP chunking, 94.7% for VP chunking, 77.1% fox' subject detection and 79.0% for object detection.

In this paper we give a synoptic view of the growth text processing technology of information extraction (IE) whose function is to extract information about a pre-specified set of entities, relations or events from natural language textsand to record this information in structured representations called templates. Here we describe the nature of the IE task, review the history of the area from its origins in AI work in the 1960's and 70's till the present, discuss the techniques being used to carry out the task, describe application areas where IE systems are or are about to be at work, and conclude with a discussion of the challenges facing the area. What emerges is a picture of an exciting new text processing technology with a host of new applications, both on its own and in conjunction with other technologies, such as information retrieval, machine translation and data mining.

This thesis concerns the problem of unknown or out-of-vocabulary (00V) words in continuous speech recognition. Most of today's state-of-the-art speech recognition systems can recognize only words that belong to some predefined finite word vocabulary. When encountering an OOV word, a speech recognizer erroneously substitutes the OOV word with a similarly sounding word from its vocabulary. Furthermore, a recognition error due to an OOV word tends to spread errors into neighboring words; dramatically degrading overall recognition performance.

In this work we investigate the usefulness of n-grams for document indexing in text categorization (TCi We call-gram a set g k of n word stems, and we say that g k occurs in a document d j when a sequence of words appears in d j that, after stop word removal and stemming, consists exactly ofthe n stems in g k , in some order. Previous researches have investigated the use of n-grams (or some variant ofthem) in the context of specific learning algorithms, and thus have not obtained general answers on their usefulness for TC In this work we investigate the usefulness of n-grams inTC independently ofany specific learning algorithm. We do so by applying feature selection to the pool of all k-grams (k # n), and checking how many n-grams score high enough to be selected in the top #k-grams. We report the results of our experiments, using various feature selection measures and varying values of #, performed on theReuters-21 standardTC benchmark. We also report resul...

this paper, we combine different systems employing known representations. The observation that suggests this approach is that systems that are designed differently, either because they use a different formalism or because they contain different knowledge, will typically produce different errors. We hope to make use of this fact and reduce the number of errors with very little additional effort by exploiting the disagreement between different language models. Al- though the approach is applicable to any type of language model, we focus on the case of statistical disambiguators that are trained on annotated corpora. The examples of the task that are present in the corpus and its annotation are fed into a learning algorithm, which induces a model of the desired input-output mapping in the form of a classifier. * EO. Box 9103, 6500 HD Nijmegen, The Netherlands, hvh@let.ktm.nl t Universiteitsplein 1, 2610 Wilrijk, Belgium, {zavrel, daelem}@uia.ua.ac.be () 2000 Association for Computational Linguistics We use a number of different learning algorithms simultaneously on the same training corpus. Each type of learning method brings its own 'inductive bias' to the task and will produce a classifier with slightly different characteristics, so that different methods will tend to produce different errors

In the general framework of knowledge discovery, Data Mining techniques are usually dedicated to information extraction from structured databases. Text Mining techniques, on the other hand, are dedicated to information extraction from unstructured textual data and Natural Language Processing (NLP) can then be seen as an interesting tool for the enhancement of information extraction procedures. In this paper, we present two examples of Text Mining tasks, association extraction and prototypical document extraction, along with several related NLP techniques.