Jordi Turmo

Abstract: The goal of the project is to explore integrated environments allowing the cost-effective deployment of vertical information access portals for specific domains. The project started in January 2010, and will last three years. Keywords: Natural Language Processing, Syntactic Analysis, Semantic Interpretation, Knowledge Acquisition, Information Extraction, Information Retrieval

Similarity, which plays a key role in fields like cognitive science, psycholinguistics and natural language processing, is a broad and multifaceted concept. In this work we analyse how two approaches that belong to different perspectives, the corpus view and the psycholinguistic view, articulate similarity between verb senses in Spanish. Specifically, we compare the similarity between verb senses based on their argument structure, which is captured through semantic roles, with their similarity defined by word associations. We address the question of whether verb argument structure, which reflects the expression of the events, and word associations, which are related to the speakers’ organization of the mental lexicon, shape similarity between verbs in a congruent manner, a topic which has not been explored previously. While we find significant correlations between verb sense similarities obtained from these two approaches, our findings also highlight some discrepancies between them and the importance of the degree of abstraction of the corpus annotation and psycholinguistic representations.

This paper describes the experience of QAST 2008, the second time a pilot track of CLEF has been held aiming to evaluate the task of Question Answering in Speech Transcripts. Five sites submitted results for at least one of the five scenarios (lectures in English, meetings in English, broadcast news in French and European Parliament debates in English and Spanish).

En este trabajo nos centramos en la adquisición de clasificaciones verbales automáticas para el español. Para ello realizamos una serie de experimentos con 20 sentidos verbales del corpus Sensem. Empleamos diferentes tipos de atributos que abarcan información lingüística diversa y un método de clustering jerárquico aglomerativo para generar varias clasificaciones. Comparamos cada una de estas clasificaciones automáticas con un gold standard creado semi-automáticamente teniendo en cuenta construcciones lingüísticas propuestas desde la lingüística teórica. Esta comparación nos permite saber qué atributos son más adecuados para crear de forma automática una clasificación coherente con la teoría sobre construcciones y cuales son las similitudes y diferencias entre la clasificación verbal automática y la que se basa en la teoría sobre construcciones lingüísticas.

----
In this work we focus on the automatic acquisition of verbal classifications for Spanish. To do so, we perform a series of experiments with 20 verbal senses that belong to the Sensem corpus. We use di↵erent kinds of features that include diverse linguistic information and an agglomerative hierarchical clustering method to generate a number of classifications. We compare each of these automatic classifications with a semi-automatically created gold standard, which is built on the basis of linguistic constructions proposed by theoretical linguistics. This comparison allows us to investigate which features are adequate to build a verb classification coherent with linguistic constructions theory and which are the similarities and di↵erences between an automatic verbal classification and a verb classification based on the theory of linguistic constructions.

The biases of individual algorithms for non-parametric document clus- tering can lead to non-optimal solutions. Ensemble clustering methods may over- come this limitation, but have not been applied to document collections. This paper presents a comparison of strategies for non-parametric document ensemble cluster- ing.

In this article, we present a factoid question-answering system, Sibyl, specifically tailored for question answering (QA) on spoken-word documents. This work explores, for the first time, which techniques can be robustly adapted from the usual QA on written documents to the more difficult spoken document scenario. More specifically, we study new information retrieval (IR) techniques designed or speech, and utilize

ABSTRACT Cluster analysis lies at the core of most unsupervised learning tasks. However, the majority of clustering algorithms depend on the all-in assumption, in which all objects belong to some cluster, and perform poorly on minority clustering tasks, in which a small fraction of signal data stands against a majority of noise. The approaches proposed so far for minority clustering are supervised: they require the number and distribution of the foreground and background clusters. In supervised learning and all-in clustering, combination methods have been successfully applied to obtain distribution-free learners, even from the output of weak individual algorithms. In this work, we propose a novel ensemble minority clustering algorithm, Ewocs, suitable for weak clustering combination. Its properties have been theoretically proved under a loose set of constraints. We also propose a number of weak clustering algorithms, and an unsupervised procedure to determine the scaling parameters for Gaussian kernels used within the task. We have implemented a number of approaches built from the proposed components, and evaluated them on a collection of datasets. The results show how approaches based on Ewocs are competitive with respect to—and even outperform—other minority clustering approaches in the state of the art.

This paper describes the participation of the Technical University of Catalonia in the CLEF 2009 Question Answering on Speech Transcripts track. We have participated in the English and Spanish scenarios of QAST. For both manual and automatic transcripts we have used a robust factual Question Answering that uses minimal syntactic information. We have also developed a NERC designed to handle automatic transcripts. We perform a detailed analysis of our results and draw conclusions relating QA performance to word error rate and the difference between written and spoken questions.

This paper presents our experiments in question answering for speech corpora. These experiments focus on improving the answer extraction step of the QA process. We present two ap-proaches to answer extraction in question answering for speech corpora that apply machine ...

The biases of individual algorithms for non-parametric document clustering can lead to non-optimal solutions. Ensemble clustering methods may overcome this limitation, but have not been applied to document collections. This paper presents a comparison of strategies for non-parametric document ensemble clustering.

Abstract: The goal of the project is to explore integrated environments allowing the cost-effective deployment of vertical information access portals for specific domains. The project started in January 2010, and will last three years. Keywords: Natural Language Processing, Syntactic Analysis, Semantic Interpretation, Knowledge Acquisition, Information Extraction, Information Retrieval