Entity relation detection is a form of information extraction that finds predefined relations between pairs of entities in text. This paper describes a relation detection approach that combines clues from different levels of syntactic processing using kernel methods. Information from three different levels of processing is considered: tokenization, sentence parsing and deep dependency analysis. Each source of information is represented by kernel functions. Then composite kernels are developed to integrate and extend individual kernels so that processing errors occurring at one level can be overcome by information from other levels. We present an evaluation of these methods on the 2004 ACE relation detection task, using Support Vector Machines, and show that each level of syntactic processing contributes useful information for this task. When evaluated on the official test data, our approach produced very competitive ACE value scores. We also compare the SVM with KNN on different kernels. 1

When complete, NomBank will provide annotation of noun arguments in Penn Treebank II (PTB). In PropBank, University of Pennsylvania annotators provide similar information for verbs. Given nominalization/verb mappings, the combination of NomBank and PropBank allows for generalization of arguments across parts of speech. This paper describes our annotation task including factors which make assigning role labels to noun arguments a challenging task. 1.

annotated the verbal argument structure for the Wall Street Journal Corpus of the Penn Treebank (PTB). This document outlines a new effort, called NomBank,

WordNet is a useful resource for lexical inference in applications. Inference over predicates, however, often requires a change in argument positions, which is not specified in WordNet. We propose a novel framework for augmenting WordNet-based inferences over predicates with corresponding argument mappings. We further present a concrete implementation of this framework, which yields substantial improvement to WordNet-based inference. 1

This paper addresses the automatic classification of semantic relations in noun phrases based on cross-linguistic evidence from a set of five Romance languages. A set of novel semantic and contextual English– Romance NP features is derived based on empirical observations on the distribution of the syntax and meaning of noun phrases on two corpora of different genre (Europarl and CLUVI). The features were employed in a Support Vector Machines algorithm which achieved an accuracy of 77.9 % (Europarl) and 74.31 % (CLUVI), an improvement compared with two state-of-the-art models reported in the literature. 1

Dedicated to my mother iii Acknowledgements I would like to thank my advisor Ralph Grishman for his guidance in academics. He is a great researcher with keen interests in science, constant efforts in doing things by hand and great personality. He is the example I followed and will continue to follow. Without him this thesis would not be possible.

Abstract not found

Semantic inference is often modeled as application of entailment rules, which specify generation of entailed sentences from a source sentence. Efficient generation and representation of entailed consequents is a fundamental problem common to such inference methods. We present a new data structure, termed compact forest, which allows efficient generation and representation of entailed consequents, each represented as a parse tree. Rule-based inference is complemented with a new approximate matching measure inspired by tree kernels, which is computed efficiently over compact forests. Our system also makes use of novel large-scale entailment rule bases, derived from Wikipedia as well as from information about predicates and their argument mapping, gathered from available lexicons and complemented by unsupervised learning. 1

We present GLARF, a framework for representing three linguistic levels and systems for generating this representation. We focus on a logical level, like LFG’s F-structure, but compatible with Penn Treebanks. While less finegrained than typical semantic role labeling approaches, our logical structure has several advantages: (1) it includes all words in all sentences, regardless of part of speech or semantic domain; and (2) it is easier to produce accurately. Our systems achieve 90 % for English/Japanese News and 74.5 % for Chinese News – these F-scores are nearly the same as those achieved for treebank-based parsing. 1

Resumen: En este artículo se describe un nuevo recurso: AnCora-Nom, un léxico de nominalizaciones deverbales del español. Actualmente, contiene 1.655 entradas léxicas y 3.094 sentidos, donde cada sentido tiene asociado el tipo denotativo y la estructura argumental con los papeles temáticos correspondientes. Este léxico se ha extraído automáticamente a partir de la información anotada en el corpus AnCora-Es. AnCora-Nom se derivó teniendo en cuenta no sólo la información estrictamente relacionada con las nominalizaciones deverbales sino también con información morfológica y sintáctico-semántica previamente anotada en el corpus. Palabras clave: nominalización deverbal, denotación, estructura argumental, recurso léxico, anotación de corpus. Abstract: This paper describes a new lexical resource: Ancora-Nom, a Spanish lexicon of deverbal nominalizations. At present, it contains 1,655 lexical entries and 3,094 senses. Each sense has a denotation type associated, and the mapping of nominal complements with arguments and the corresponding theta roles is also annotated. A particular interest of this lexicon is that it has been automatically extracted from the annotated AnCora-Es corpus. AnCora-Nom was derived taking into account the information directly related to nominalizations, but also the morphological and syntactic-semantic information annotated in the corpus.