Extractive summarization using complex networks and syntactic dependency

The realization that statistical physics methods can be applied to analyze written texts represented as complex networks has led to several developments in natural language processing, including automatic summarization and evaluation of machine translation. Most importantly, so far only a few metrics of complex networks have been used and therefore there is ample opportunity to enhance the statistics-based methods as new measures of network topology and dynamics are created. In this paper, we employ for the first time the metrics betweenness, vulnerability and diversity to analyze written texts in Brazilian Portuguese. Using strategies based on diversity metrics, a better performance in automatic summarization is achieved in comparison to previous work employing complex networks. With an optimized method the Rouge score (an automatic evaluation method used in summarization) was 0.5089, which is the best value ever achieved for an extractive summarizer with statistical methods based on complex networks for Brazilian Portuguese. Furthermore, the diversity metric can detect keywords with high precision, which is why we believe it is suitable to produce good summaries. It is also shown that incorporating linguistic knowledge through a syntactic parser does enhance the performance of the automatic summarizers, as expected, but the increase in the Rouge score is only minor. These results reinforce the suitability of complex network methods for improving automatic summarizers in particular, and treating text in general. (C) 2011 Elsevier B.V. All rights reserved.

Unveiling the relationship between complex networks metrics and word senses

The automatic disambiguation of word senses (i.e., the identification of which of the meanings is used in a given context for a word that has multiple meanings) is essential for such applications as machine translation and information retrieval, and represents a key step for developing the so-called Semantic Web. Humans disambiguate words in a straightforward fashion, but this does not apply to computers. In this paper we address the problem of Word Sense Disambiguation (WSD) by treating texts as complex networks, and show that word senses can be distinguished upon characterizing the local structure around ambiguous words. Our goal was not to obtain the best possible disambiguation system, but we nevertheless found that in half of the cases our approach outperforms traditional shallow methods. We show that the hierarchical connectivity and clustering of words are usually the most relevant features for WSD. The results reported here shed light on the relationship between semantic and structural parameters of complex networks. They also indicate that when combined with traditional techniques the complex network approach may be useful to enhance the discrimination of senses in large texts. Copyright (C) EPLA, 2012

Vantaggi e limiti dell'integrazione tra sistemi di traduzione assistita e sistemi di traduzione automatica

Computer-assisted translation (or computer-aided translation or CAT) is a form of language translation in which a human translator uses computer software in order to facilitate the translation process. Machine translation (MT) is the automated process by which a computerized system produces a translated text or speech from one natural language to another. Both of them are leading and promising technologies in the translation industry; it therefore seems important that translation students and professional translators become familiar with this relatively new types of technology. Whether used together, not only might these two different types of systems reduce translation time, but also lead to a further improvement in the field of translation technologies. The dissertation consists of four chapters. The first one surveys the chronological development of MT and CAT tools, the emergence of pre-editing, post-editing and controlled language and the very last frontiers in this sector. The second one provide a general overview on the four main CAT tools that are used nowadays and tested hereto. The third chapter is dedicated to the experimentations that have been conducted in order to analyze and evaluate the performance of the four integrated systems that are the core subject of this dissertation. Finally, the fourth chapter deals with the issue of terminological equivalence in interlinguistic translation. The purpose of this dissertation is not to provide an objective and definitive solution to the complex issues that arise at any time in the field of translation technologies, this aim being well away from being achieved, but to supply information about the limits and potentiality that are typical of those instruments which are now essential to any professional translator.

Internazionalizzazione e traduzione: Un'esperienza presso l'azienda TR Turoni S.R.L.

This dissertation was conducted within the project Language Toolkit, which has the aim of integrating the worlds of work and university. In particular, it consists of the translation into English of documents commissioned by the Italian company TR Turoni and its primary purpose is to demonstrate that, in the field of translation for companies, the existing translation support tools and software can optimise and facilitate the translation process. The work consists of five chapters. The first introduces the Language Toolkit project, the TR Turoni company and its relationship with the CERMAC export consortium. After outlining the current state of company internationalisation, the importance of professional translators in enhancing the competitiveness of companies that enter new international markets is highlighted. Chapter two provides an overview of the texts to be translated, focusing on the textual function and typology and on the addressees. After that, manual translation and the main software developed specifically for translators are described, with a focus on computer-assisted translation (CAT) and machine translation (MT). The third chapter presents the target texts and the corresponding translations. Chapter four is dedicated to the analysis of the translation process. The first two texts were translated manually, with the support of a purpose-built specialized corpus. The following two documents were translated with the software SDL Trados Studio 2011 and its applications. The last texts were submitted to the Google Translate service and to a process of pre and post-editing. Finally, in chapter five conclusions are drawn about the main limits and potentialities of the different translations techniques. In addition to this, the importance of an integrated use of all available instruments is underlined.

Proposta di traduzione commentata di un articolo scientifico sulla sperimentazione animale: risorse, metodi e terminologia.

The aim of this dissertation is to provide a translation from English into Italian of a highly specialized scientific article published by the online journal ALTEX. In this text, the authors propose a roadmap for how to overcome the acknowledged scientific gaps for the full replacement of systemic toxicity testing using animals. The main reasons behind this particular choice are my personal interest in specialized translation of scientific texts and in the alternatives to animal testing. Moreover, this translation has been directly requested by the Italian molecular biologist and clinical biochemist Candida Nastrucci. It was not possible to translate the whole article in this project, for this reason, I decided to translate only the introduction, the chapter about skin sensitization, and the conclusion. I intend to use the resources that were created for this project to translate the rest of the article in the near future. In this study, I will show how a translator can translate such a specialized text with the help of a field expert using CAT Tools and a specialized corpus. I will also discuss whether machine translation can prove useful to translate this type of document. This work is divided into six chapters. The first one introduces the main topic of the article and explains my reasons for choosing this text; the second one contains an analysis of the text type, focusing on the differences and similarities between Italian and English conventions. The third chapter provides a description of the resources that were used to translate this text, i.e. the corpus and the CAT Tools. The fourth one contains the actual translation, side-by-side with the original text, while the fifth one provides a general comment on the translation difficulties, an analysis of my translation choices and strategies, and a comment about the relationship between the field expert and the translator. Finally, the last chapter shows whether machine translation and post-editing can be an advantageous strategy to translate this type of document. The project also contains two appendixes. The first one includes 54 complex terminological sheets, while the second one includes 188 simple terminological sheets.

Traduzione tecnica e internazionalizzazione d'impresa: Il caso di Technologica S.r.l.

This dissertation is part of the Language Toolkit project which is a collaboration between the School of Foreign Languages and Literature, Interpreting and Translation of the University of Bologna, Forlì campus, and the Chamber of Commerce of Forlì-Cesena. This project aims to create an exchange between translation students and companies who want to pursue a process of internationalization. The purpose of this dissertation is demonstrating the benefits that translation systems can bring to businesses. In particular, it consists of the translation into English of documents supplied by the Italian company Technologica S.r.l. and the creation of linguistic resources that can be integrated into computer-assisted translation (CAT) software, in order to optimize the translation process. The latter is claimed to be a priority with respect to the actual translation products (the target texts), since the analysis conducted on the source texts highlighted that the company could streamline and optimize its English language communication thanks to the use of open source CAT tools such as OmegaT. The work consists of five chapters. The first introduces the Language Toolkit project, the company (Technologica S.r.l ) and its products. The second chapter provides some considerations about technical translation, its features and some misconceptions about it. The difference between technical translation and scientific translation is then clarified and an overview is offered of translation aids such as those used for computer-assisted translation, machine translation, termbases and translation memories. The third chapter contains the analysis of the texts commissioned by Technologica S.r.l. and their categorization. The fourth chapter describes the translation process, with particular attention to terminology extraction and the creation of a bilingual glossary based on a specialized corpus. The glossary was integrated into the OmegaT software in order to facilitate the translation process both for the present task and for future applications. The memory deriving from the translation represents a sort of hybrid resource between a translation memory and a glossary. This was found to be the most appropriate format, given the specific nature of the texts to be translated. Finally, in chapter five conclusions are offered about the importance of language training within a company environment, the potentialities of translation aids and the benefits that they would bring to a company wishing to internationalize itself.

La traduzione automatica e i language service provider: il caso di CTI

La presente tesi nasce da un tirocinio avanzato svolto presso l’azienda CTI (Communication Trend Italia) di Milano. Gli obiettivi dello stage erano la verifica della possibilità di inserire gli strumenti automatici nel flusso di lavoro dell’azienda e l'individuazione delle tipologie testuali e delle combinazioni linguistiche a cui essi sono applicabili. Il presente elaborato si propone di partire da un’analisi teorica dei vari aspetti legati all’utilizzo della TA, per poi descriverne l’applicazione pratica nei procedimenti che hanno portato alla creazione dei sistemi custom. Il capitolo 1 offre una panoramica teorica sul mondo della machine translation, che porta a delineare la modalità di utilizzo della TA ad oggi più diffusa: quella in cui la traduzione fornita dal sistema viene modificata tramite post-editing oppure il testo di partenza viene ritoccato attraverso il pre-editing per eliminare gli elementi più ostici. Nel capitolo 2, partendo da una panoramica relativa ai principali software di traduzione automatica in uso, si arriva alla descrizione di Microsoft Translator Hub, lo strumento scelto per lo sviluppo dei sistemi custom di CTI. Nel successivo passaggio, l’attenzione si concentra sull’ottenimento di sistemi customizzati. Un ampio approfondimento è dedicato ai metodi per reperire ed utilizzare le risorse. In seguito viene descritto il percorso che ha portato alla creazione e allo sviluppo dei due sistemi Bilanci IT_EN e Atto Costitutivo IT_EN in Microsoft Translator Hub. Infine, nel quarto ed ultimo capitolo gli output che i due sistemi forniscono vengono rivisti per individuarne le caratteristiche e analizzati tramite alcuni tool di valutazione automatica. Grazie alle informazioni raccolte vengono poi formulate alcune previsioni sul futuro uso dei sistemi presso l’azienda CTI.

Book review: An Essay on Grammar-Parser Relations (J. van de Koot)

Review of this book, that is the author's Thesis Dissertation.

UPM system for WMT 2012

This paper describes the UPM system for the Spanish-English translation task at the NAACL 2012 workshop on statistical machine translation. This system is based on Moses. We have used all available free corpora, cleaning and deleting some repetitions. In this paper, we also propose a technique for selecting the sentences for tuning the system. This technique is based on the similarity with the sentences to translate. With our approach, we improve the BLEU score from 28.37% to 28.57%. And as a result of the WMT12 challenge we have obtained a 31.80% BLEU with the 2012 test set. Finally, we explain different experiments that we have carried out after the competition.

Multilingual number transcription for text-to-speech conversion

This paper describes the text normalization module of a text to speech fully-trainable conversion system and its application to number transcription. The main target is to generate a language independent text normalization module, based on data instead of on expert rules. This paper proposes a general architecture based on statistical machine translation techniques. This proposal is composed of three main modules: a tokenizer for splitting the text input into a token graph, a phrase-based translation module for token translation, and a post-processing module for removing some tokens. This architecture has been evaluated for number transcription in several languages: English, Spanish and Romanian. Number transcription is an important aspect in the text normalization problem.

Medida de distancia semántica en grafos UNL

El trabajo que se presenta a continuación desarrolla un modelo para calcular la distancia semántica entre dos oraciones representadas por grafos UNL. Este problema se plantea en el contexto de la traducción automática donde diferentes traductores pueden generar oraciones ligeramente diferentes partiendo del mismo original. La medida de distancia que se propone tiene como objetivo proporcionar una evaluación objetiva sobre la calidad del proceso de generación del texto. El autor realiza una exploración del estado del arte sobre esta materia, reuniendo en un único trabajo los modelos propuestos de distancia semántica entre conceptos, los modelos de comparación de grafos y las pocas propuestas realizadas para calcular distancias entre grafos conceptuales. También evalúa los pocos recursos disponibles para poder experimentar el modelo y plantea una metodología para generar los conjuntos de datos que permitirían aplicar la propuesta con el rigor científico necesario y desarrollar la experimentación. Utilizando las piezas anteriores se propone un modelo novedoso de comparación entre grafos conceptuales que permite utilizar diferentes algoritmos de distancia entre conceptos y establecer umbrales de tolerancia para permitir una comparación flexible entre las oraciones. Este modelo se programa utilizando C++, se alimenta con los recursos a los que se ha hecho referencia anteriormente, y se experimenta con un conjunto de oraciones creado por el autor ante la falta de otros recursos disponibles. Los resultados del modelo muestran que la metodología y la implementación pueden conducir a la obtención de una medida de distancia entre grafos UNL con aplicación en sistemas de traducción automática, sin embargo, la carencia de recursos y de datos etiquetados con los que validar el algoritmo requieren un esfuerzo previo importante antes de poder ofrecer resultados concluyentes.---ABSTRACT---The work presented here develops a model to calculate the semantic distance between two sentences represented by their UNL graphs. This problem arises in the context of machine translation where different translators can generate slightly different sentences from the same original. The distance measure that is proposed aims to provide an objective evaluation on the quality of the process involved in the generation of text. The author carries out an exploration of the state of the art on this subject, bringing together in a single work the proposed models of semantic distance between concepts, models for comparison of graphs and the few proposals made to calculate distances between conceptual graphs. It also assesses the few resources available to experience the model and presents a methodology to generate the datasets that would be needed to develop the proposal with the scientific rigor required and to carry out the experimentation. Using the previous parts a new model is proposed to compute differences between conceptual graphs; this model allows the use of different algorithms of distance between concepts and is parametrized in order to be able to perform a flexible comparison between the resulting sentences. This model is implemented in C++ programming language, it is powered with the resources referenced above and is experienced with a set of sentences created by the author due to the lack of other available resources. The results of the model show that the methodology and the implementation can lead to the achievement of a measure of distance between UNL graphs with application in machine translation systems, however, lack of resources and of labeled data to validate the algorithm requires an important effort to be done first in order to be able to provide conclusive results.

Método para la Construcción Automática de Ontologías Basado en Patrones Lingüísticos

La tesis que se presenta tiene como propósito la construcción automática de ontologías a partir de textos, enmarcándose en el área denominada Ontology Learning. Esta disciplina tiene como objetivo automatizar la elaboración de modelos de dominio a partir de fuentes información estructurada o no estructurada, y tuvo su origen con el comienzo del milenio, a raíz del crecimiento exponencial del volumen de información accesible en Internet. Debido a que la mayoría de información se presenta en la web en forma de texto, el aprendizaje automático de ontologías se ha centrado en el análisis de este tipo de fuente, nutriéndose a lo largo de los años de técnicas muy diversas provenientes de áreas como la Recuperación de Información, Extracción de Información, Sumarización y, en general, de áreas relacionadas con el procesamiento del lenguaje natural. La principal contribución de esta tesis consiste en que, a diferencia de la mayoría de las técnicas actuales, el método que se propone no analiza la estructura sintáctica superficial del lenguaje, sino que estudia su nivel semántico profundo. Su objetivo, por tanto, es tratar de deducir el modelo del dominio a partir de la forma con la que se articulan los significados de las oraciones en lenguaje natural. Debido a que el nivel semántico profundo es independiente de la lengua, el método permitirá operar en escenarios multilingües, en los que es necesario combinar información proveniente de textos en diferentes idiomas. Para acceder a este nivel del lenguaje, el método utiliza el modelo de las interlinguas. Estos formalismos, provenientes del área de la traducción automática, permiten representar el significado de las oraciones de forma independiente de la lengua. Se utilizará en concreto UNL (Universal Networking Language), considerado como la única interlingua de propósito general que está normalizada. La aproximación utilizada en esta tesis supone la continuación de trabajos previos realizados tanto por su autor como por el equipo de investigación del que forma parte, en los que se estudió cómo utilizar el modelo de las interlinguas en las áreas de extracción y recuperación de información multilingüe. Básicamente, el procedimiento definido en el método trata de identificar, en la representación UNL de los textos, ciertas regularidades que permiten deducir las piezas de la ontología del dominio. Debido a que UNL es un formalismo basado en redes semánticas, estas regularidades se presentan en forma de grafos, generalizándose en estructuras denominadas patrones lingüísticos. Por otra parte, UNL aún conserva ciertos mecanismos de cohesión del discurso procedentes de los lenguajes naturales, como el fenómeno de la anáfora. Con el fin de aumentar la efectividad en la comprensión de las expresiones, el método provee, como otra contribución relevante, la definición de un algoritmo para la resolución de la anáfora pronominal circunscrita al modelo de la interlingua, limitada al caso de pronombres personales de tercera persona cuando su antecedente es un nombre propio. El método propuesto se sustenta en la definición de un marco formal, que ha debido elaborarse adaptando ciertas definiciones provenientes de la teoría de grafos e incorporando otras nuevas, con el objetivo de ubicar las nociones de expresión UNL, patrón lingüístico y las operaciones de encaje de patrones, que son la base de los procesos del método. Tanto el marco formal como todos los procesos que define el método se han implementado con el fin de realizar la experimentación, aplicándose sobre un artículo de la colección EOLSS “Encyclopedia of Life Support Systems” de la UNESCO. ABSTRACT The purpose of this thesis is the automatic construction of ontologies from texts. This thesis is set within the area of Ontology Learning. This discipline aims to automatize domain models from structured or unstructured information sources, and had its origin with the beginning of the millennium, as a result of the exponential growth in the volume of information accessible on the Internet. Since most information is presented on the web in the form of text, the automatic ontology learning is focused on the analysis of this type of source, nourished over the years by very different techniques from areas such as Information Retrieval, Information Extraction, Summarization and, in general, by areas related to natural language processing. The main contribution of this thesis consists of, in contrast with the majority of current techniques, the fact that the method proposed does not analyze the syntactic surface structure of the language, but explores his deep semantic level. Its objective, therefore, is trying to infer the domain model from the way the meanings of the sentences are articulated in natural language. Since the deep semantic level does not depend on the language, the method will allow to operate in multilingual scenarios, where it is necessary to combine information from texts in different languages. To access to this level of the language, the method uses the interlingua model. These formalisms, coming from the area of machine translation, allow to represent the meaning of the sentences independently of the language. In this particular case, UNL (Universal Networking Language) will be used, which considered to be the only interlingua of general purpose that is standardized. The approach used in this thesis corresponds to the continuation of previous works carried out both by the author of this thesis and by the research group of which he is part, in which it is studied how to use the interlingua model in the areas of multilingual information extraction and retrieval. Basically, the procedure defined in the method tries to identify certain regularities at the UNL representation of texts that allow the deduction of the parts of the ontology of the domain. Since UNL is a formalism based on semantic networks, these regularities are presented in the form of graphs, generalizing in structures called linguistic patterns. On the other hand, UNL still preserves certain mechanisms of discourse cohesion from natural languages, such as the phenomenon of the anaphora. In order to increase the effectiveness in the understanding of expressions, the method provides, as another significant contribution, the definition of an algorithm for the resolution of pronominal anaphora limited to the model of the interlingua, in the case of third person personal pronouns when its antecedent is a proper noun. The proposed method is based on the definition of a formal framework, adapting some definitions from Graph Theory and incorporating new ones, in order to locate the notions of UNL expression and linguistic pattern, as well as the operations of pattern matching, which are the basis of the method processes. Both the formal framework and all the processes that define the method have been implemented in order to carry out the experimentation, applying on an article of the "Encyclopedia of Life Support Systems" of the UNESCO-EOLSS collection.

Grammar specification for the recognition of temporal expressions

This paper tells about the recognition of temporal expressions and the resolution of their temporal reference. A proposal of the units we have used to face up this tasks over a restricted domain is shown. We work with newspapers' articles in Spanish, that is why every reference we use is in Spanish. For the identification and recognition of temporal expressions we base on a temporal expression grammar and for the resolution on a dictionary, where we have the information necessary to do the date operation based on the recognized expressions. In the evaluation of our proposal we have obtained successful results for the examples studied.

A general proposal to multilingual information access based on syntactic-semantic patterns

In the last few years, there has been a wide development in the research on textual information systems. The goal is to improve these systems in order to allow an easy localization, treatment and access to the information stored in digital format (Digital Databases, Documental Databases, and so on). There are lots of applications focused on information access (for example, Web-search systems like Google or Altavista). However, these applications have problems when they must access to cross-language information, or when they need to show information in a language different from the one of the query. This paper explores the use of syntactic-sematic patterns as a method to access to multilingual information, and revise, in the case of Information Retrieval, where it is possible and useful to employ patterns when it comes to the multilingual and interactive aspects. On the one hand, the multilingual aspects that are going to be studied are the ones related to the access to documents in different languages from the one of the query, as well as the automatic translation of the document, i.e. a machine translation system based on patterns. On the other hand, this paper is going to go deep into the interactive aspects related to the reformulation of a query based on the syntactic-semantic pattern of the request.

AliQAn, Spanish QA system at multilingual QA@CLEF-2008

Comunicación presentada en Cross-Language Evaluation Forum (CLEF 2008), Aarhus, Denmark, September 17-19, 2008.