Alfio Massimiliano Gliozzo

The role of domain information in Word Sense Disambiguation

This paper explores the role of domain information in word sense disambiguation. The underlying hypothesis is that domain labels, such as MEDICINE, ARCHITECTURE and SPORT, provide a useful way to establish semantic relations among word senses, which can be profitably used during the disambiguation process. Results obtained at the SENSEVAL-2 initiative confirm that for a significant subset of words domain information can be used to disambiguate with a very high level of precision.

Unsupervised and supervised exploitation of semantic domains in lexical disambiguation

Abstract Domains are common areas of human discussion, such as economics, politics, law, science, etc., which are at the basis of lexical coherence. This paper explores the dual role of domains in word sense disambiguation (WSD). On one hand, domain information provides generalized features at the paradigmatic level that are useful to discriminate among word senses. On the other hand, domain distinctions constitute a useful level of coarse grained sense distinctions, which lends itself to more accurate disambiguation with lower amounts of knowledge. In this paper we extend and ground the modeling of domains and the exploitation of WordNet Domains , an extension of WordNet in which each synset is labeled with domain information. We propose a novel unsupervised probabilistic method for the critical step of estimating domain relevance for contexts, and suggest utilizing it within unsupervised domain driven disambiguation for word senses, as well as within a traditional supervised approach. The paper presents empirical assessments of the potential utilization of domains in WSD at a wide range of comparative settings, supervised and unsupervised. Following the dual role of domains we report experiments that evaluate both the extent to which domain information provides effective features for WSD, as well as the accuracy obtained by WSD at domain-level sense granularity. Furthermore, we demonstrate the potential for either avoiding or minimizing manual annotation thanks to the generalized level of information provided by domains.

Exploiting Comparable Corpora and Bilingual Dictionaries for Cross-Language Text Categorization

Cross-language Text Categorization is the task of assigning semantic classes to documents written in a target language (e.g. English) while the system is trained using labeled documents in a source language (e.g. Italian).In this work we present many solutions according to the availability of bilingual resources, and we show that it is possible to deal with the problem even when no such resources are accessible. The core technique relies on the automatic acquisition of Multilingual Domain Models from comparable corpora.Experiments show the effectiveness of our approach, providing a low cost solution for the Cross Language Text Categorization task. In particular, when bilingual dictionaries are available the performance of the categorization gets close to that of monolingual text categorization.

Cross Language Text Categorization by Acquiring Multilingual Domain Models from Comparable Corpora

In a multilingual scenario, the classical monolingual text categorization problem can be reformulated as a cross language TC task, in which we have to cope with two or more languages (e.g. English and Italian). In this setting, the system is trained using labeled examples in a source language (e.g. English), and it classifies documents in a different target language (e.g. Italian). In this paper we propose a novel approach to solve the cross language text categorization problem based on acquiring Multilingual Domain Models from comparable corpora in a totally unsupervised way and without using any external knowledge source (e.g. bilingual dictionaries). These Multilingual Domain Models are exploited to define a generalized similarity function (i.e. a kernel function) among documents in different languages, which is used inside a Support Vector Machines classification framework. The results show that our approach is a feasible and cheap solution that largely outperforms a baseline.

Direct Word Sense Matching for Lexical Substitution

This paper investigates conceptually and empirically the novel sense matching task, which requires to recognize whether the senses of two synonymous words match in context. We suggest direct approaches to the problem, which avoid the intermediate step of explicit word sense disambiguation, and demonstrate their appealing advantages and stimulating potential for future research.

FBK-irst: Lexical Substitution Task Exploiting Domain and Syntagmatic Coherence

This paper summarizes FBK-irst participation at the lexical substitution task of the Semeval competition. We submitted two different systems, both exploiting synonym lists extracted from dictionaries. For each word to be substituted, the systems rank the associated synonym list according to a similarity metric based on Latent Semantic Analysis and to the occurrences in the Web 1T 5-gram corpus, respectively. In particular, the latter system achieves the state-of-the-art performance, largely surpassing the baseline proposed by the organizers.

Semantic Domains in Computational Linguistics

Semantic fields are lexically coherent the words they contain co-occur in texts. In this book the authors introduce and define semantic domains, a computational model for lexical semantics inspired by the theory of semantic fields. Semantic domains allow us to exploit domain features for texts, terms and concepts, and they can significantly boost the performance of natural-language processing systems. Semantic domains can be derived from existing lexical resources or can be acquired from corpora in an unsupervised manner. They also have the property of interlinguality, and they can be used to relate terms in different languages in multilingual application scenarios. The authors give a comprehensive explanation of the computational model, with detailed chapters on semantic domains, domain models, and applications of the technique in text categorization, word sense disambiguation, and cross-language text categorization. This book is suitable for researchers and graduate students in computational linguistics.

Kernel methods for minimally supervised wsd

We present a semi-supervised technique for word sense disambiguation that exploits external knowledge acquired in an unsupervised manner. In particular, we use a combination of basic kernel functions to independently estimate syntagmatic and domain similarity, building a set of word-expert classifiers that share a common domain model acquired from a large corpus of unlabeled data. The results show that the proposed approach achieves state-of-the-art performance on a wide range of lexical sample tasks and on the English all-words task of Senseval-3, although it uses a considerably smaller number of training examples than other methods.

Frame Detection over the Semantic Web

In the past, research in ontology learning from text has mainly focused on entity recognition, taxonomy induction and relation extraction. In this work we approach a challenging research issue: detecting semantic frames from texts and using them to encode web ontologies. We exploit a new generation Natural Language Processing technology for frame detection, and we enrich the frames acquired so far with argument restrictions provided by a super-sense tagger and domain specializations. The results are encoded according to a Linguistic MetaModel, which allows a complete translation of lexical resources and data acquired from text, enabling custom transformations of the enriched frames into modular ontology components.

Investigating Unsupervised Learning for Text Categorization Bootstrapping

We propose a generalized bootstrapping algorithm in which categories are described by relevant seed features. Our method introduces two unsupervised steps that improve the initial categorization step of the bootstrapping scheme: (i) using Latent Semantic space to obtain a generalized similarity measure between instances and features, and (ii) the Gaussian Mixture algorithm, to obtain uniform classification probabilities for unlabeled examples. The algorithm was evaluated on two Text Categorization tasks and obtained state-of-the-art performance using only the category names as initial seeds.