Lluís Padró

Named Entity Extraction using AdaBoost

This paper presents a Named Entity Extraction (NEE) system for the CoNLL 2002 competition. The two main sub-tasks of the problem, recognition (NER) and classification (NEC), are performed sequentially and independently with separate modules. Both modules are machine learning based systems, which make use of binary AdaBoost classifiers.

Mapping WordNets using structural information

We present a robust approach for linking already existing lexical/semantic hierarchies. We used a constraint satisfaction algorithm (relaxation labeling) to select - among a set of candidates- the node in a target taxonomy that bests matches each node in a source taxonomy. In particular, we use it to map the nominal part of WordNet 1.5 onto WordNet 1.6, with a very high precision and a very low remaining ambiguity.

A simple named entity extractor using AdaBoost

This paper presents a Named Entity Extraction (NEE) system for the CoNLL-2003 shared task competition. As in the past year edition (Carreras et al., 2002a), we have approached the task by treating the two main sub–tasks of the problem, recognition (NER) and classification (NEC), sequentially and independently with separate modules. Both modules are machine learning based systems, which make use of binary and multiclass AdaBoost classifiers. Named Entity recognition is performed as a greedy sequence tagging procedure under the well–known BIO labelling scheme. This tagging process makes use of three binary classifiers trained to be experts on the recognition of B, I, and O labels, respectively. Named Entity classification is viewed as a 4–class classification problem (with LOC, PER, ORG, and MISC class labels), which is straightforwardly addressed by the use of a multiclass learning algorithm. The system presented here consists of a replication, with some minor changes, of the system that obtained the best results in the CoNLL-2002 NEE task. Therefore, it can be considered as a benchmark of the state–of–the– art technology for the current edition, and will allow also to make comparisons about the training corpora of both editions.

A Flexible POS Tagger Using an Automatically Acquired Language Model

We present an algorithm that automatically learns context constraints using statistical decision trees. We then use the acquired constraints in a flexible POS tagger. The tagger is able to use information of any degree: n-grams, automatically learned context constraints, linguistically motivated manually written constraints, etc. The sources and kinds of constraints are unrestricted, and the language model can be easily extended, improving the results. The tagger has been tested and evaluated on the WSJ corpus.

A Machine Learning Approach to POS Tagging

We have applied the inductive learning of statistical decision trees and relaxation labeling to the Natural Language Processing (NLP) task of morphosyntactic disambiguation (Part Of Speech Tagging). The learning process is supervised and obtains a language model oriented to resolve POS ambiguities, consisting of a set of statistical decision trees expressing distribution of tags and words in some relevant contexts. The acquired decision trees have been directly used in a tagger that is both relatively simple and fast, and which has been tested and evaluated on the Wall Street Journal (WSJ) corpus with competitive accuracy. However, better results can be obtained by translating the trees into rules to feed a flexible relaxation labeling based tagger. In this direction we describe a tagger which is able to use information of any kind (n-grams, automatically acquired constraints, linguistically motivated manually written constraints, etc.), and in particular to incorporate the machine-learned decision trees. Simultaneously, we address the problem of tagging when only limited training material is available, which is crucial in any process of constructing, from scratch, an annotated corpus. We show that high levels of accuracy can be achieved with our system in this situation, and report some results obtained when using it to develop a 5.5 million words Spanish corpus from scratch.

Combining Supervised and Unsupervised Lexical Knowledge Methods for Word Sense Disambiguation

This work combines a set of available techniques – whichcould be further extended – to perform noun sense disambiguation. We use several unsupervised techniques (Rigau et al., 1997) that draw knowledge from a variety of sources. In addition, we also apply a supervised technique in order to show that supervised and unsupervised methods can be combined to obtain better results. This paper tries to prove that using an appropriate method to combine those heuristics we can disambiguate words in free running text with reasonable precision.

Learning a perceptron-based named entity chunker via online recognition feedback

We present a novel approach for the problem of Named Entity Recognition and Classification (NERC), in the context of the CoNLL-2003 Shared Task.

Developing a hybrid NP parser

We describe the use of energy function optimisation in very shallow syntactic parsing. The approach can use linguistic rules and corpus-based statistics, so the strengths of both linguistic and statistical approaches to NLP can be combined in a single framework. The rules are contextual constraints for resolving syntactic ambiguities expressed as alternative tags, and the statistical language model consists of corpus-based n-grams of syntactic tags. The success of the hybrid syntactic disambiguator is evaluated against a held-out benchmark corpus. Also the contributions of the linguistic and statistical language models to the hybrid model are estimated.

On the Evaluation and Comparison of Taggers: the Effect of Noise in Testing Corpora

This paper addresses the issue of POS tagger evaluation. Such evaluation is usually performed by comparing the tagger output with a reference test corpus, which is assumed to be error-free. Currently used corpora contain noise which causes the obtained performance to be a distortion of the real value. We analyze to what extent this distortion may invalidate the comparison between taggers or the measure of the improvement given by a new system. The main conclusion is that a more rigorous testing experimentation setting/designing is needed to reliably evaluate and compare tagger accuracies.

POS tagging using relaxation labelling

Relaxation labelling is an optimization technique used in many fields to solve contraint satisfcation problems. The algorithm finds a combination of values for a set of variables such that satisfies -to the maximum possible degree- a set of given constraints. This paper describes some experiments performed applying it to POS tagging, and the results obtained. It also ponders the possibility of applying it to Word Sense Disambiguation.