We study the problem of combining the outcomes of several different classifiers in a way that provides a coherent inference that satisfies some constraints. In particular, we develop two general approaches for an important subproblem - identifying phrase structure. The first is a Markovian approach that extends standard HMMs to allow the use of a rich observation structure and of general classifiers to model state-observation dependencies. The second is an extension of constraint satisfaction formalisms. We develop efficient combination algorithms under both models and study them experimentally in the context of shallow parsing. 1 Introduction In many situations it is necessary to make decisions that depend on the outcomes of several different classifiers in a way that provides a coherent inference that satisfies some constraints - the sequential nature of the data or other domain specific constraints. Consider, for example, the problem of chunking natural language sentences ...

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.

Dividing sentences in chunks of words is a useful preprocessing step for Parsing, information extraction and information retrieval. (Ramshaw and Marcus, 1995) have introduced a "convenient" data representation for chunking by converting it to a tagging task. In this paper we will examine seven different data representations for the problem of recognizing noun phrase chunks. We will show that the the data representation choice has a minor influence on chunking performance. However,

A SNoW based learning approach to shallow parsing tasks is presented and studied experimentally. The approach learns to identify syntactic patterns by combining simple predictors to produce a coherent inference. Two instantiations of this approach are studied and experimental results for Noun-Phrases (NP) and Subject-Verb (SV) phrases that compare favorably with the best published results are presented. In doing that, we compare two ways of modeling the problem of learning to recognize patterns and suggest that shallow parsing patterns are bet- ter learned using open/close predictors than using inside/outside predictors.

This paper presents a new approach to partial parsing of context-free structures. The approach is based

A parsing system returning analyses in the form of sets of grammatical relations can obtain high precision if it hypothesises a particular relation only when it is certain that the relation is correct. We operationalise this technique---in a statistical parser using a manually-developed wide-coverage grammar of English---by only returning relations that form part of all analyses licensed by the grammar. We observe an increase in precision from 75% to over 90% (at the cost of a reduction in recall) on a test corpus of naturally-occurring text.

Signi cant amount of work has been devoted recently to develop learning techniques that can be used to generate partial (shallow) analysis of natural language sentences rather than a full parse. In this work we set out to evaluate whether this direction is worthwhile by comparing a learned shallow parser to one of the best learned full parsers on tasks both can perform | identifying phrases in sentences. We conclude that directly learning to perform these tasks as shallow parsers do is advantageous over full parsers both in terms of performance and robustness to new and lower quality texts. 1

We present a unified technique to solve di#erent shallow parsing tasks as a tagging problem using a Hidden Markov Model-based approach (HMM). This technique consists of the incorporation of the relevant information for each task into the models. To do this, the training corpus is transformed to take into account this information. In this way, no change is necessary for either the training or tagging process, so it allows for the use of a standard HMM approach. Taking into account this information, we construct a Specialized HMM which gives more complete contextual models. We have tested our system on chunking and clause identification tasks using di#erent specialization criteria. The results obtained are in line with the results reported for most of the relevant state-of-the-art approaches.

Three data-driven publicly available part-of-speech taggers are applied to shallowparsin of Swedish texts. The phrase structure isrepresen ted byn00 types of phrasesin a hierarchical structurecon taintu labels for every con81E1wN t type the token belonR toin the parse tree. The enw din is basedon thecon0E91wN086 of the phrase tagson the path from lowest to highern des. Variouslinsw2---E2 features are usedin learn26--- the taggers aretrain2 on the basis of lexicalincalwB62R on , part-of-speechon0 ,an a combin1B1--- of both, to predict the phrase structure of the token with or without part-of-speech. Specialatten tion is directed to the taggers'senrs'w22 y todi#eren t types oflin0E0wN0 in0E0wN02 in0E0 in learnE0w as well as the taggers'senrs'w29 y to the sizean the various types oftrain6w data sets. The methodcan be easilytranw28R86 to otherlanw09EE2 Keywords: ChunBR1E Shallowparsin1 Part-of-speech taggers,Hidden Markov models, Maximumen tropy learn1wN TranERwN2E2E8wnwn learnRw 1. Introducti5 Machin learn0--- technwB62 in the last decade have permeated several areas ofnw1R21 lan1R21 processin (NLP). Thereason is that a vastn umber of machin learnR9 algorithms have proved to be able tolearn fromnomw29 lanw296 data given a relatively small correctly anrrectl corpus. Therefore, machin learn---2 algorithms make it possible towithin a short period of time developlanwB69 resources---dataansourc on variouslinsw1B8--- levels---that arenwB---19Ew forn umerousapplication in nplic lanlica processin8 On of the most popular NLP areas that machin learn------ algorithms have been successfully applied to is part-of-speech (PoS)taggin0 i.e. thean0R682wN of words with thecon textually appropriate PoS tags,often inen2E8 morphological features. The datadriven algorithms that have been su...

This paper proposes a new error-driven H/vIMbased text chunk tagger with context-dependent lexicon. Compared with standard HMM-based tagger, this tagger uses a new Hidden Markov Modelling approach which incorporates more contextual information into a lexical entry. Moreover, an error-driven learning approach is adopted to decrease the memory requirement by keeping only positive lexical entries and makes it possible to further incorporate more contextdependent lexical entries. Experiments show that this technique achieves overall precision and recall rates of 93.40% and 93.95% for all chunk types, 93.60% and 94.64% for noun phrases, and 94.64% and 94.75% for verb phrases when trained on PENN WSJ TreeBank section 00-19 and tested on section 20-24, while 25-fold validation experiments of PENN WSJ TreeBank show overall precision and recall rates of 96.40% and 96.47% for all chunk types, 96.49% and 96.99% for noun phrases, and 97.13% and 97.36% for verb phrases.