De la Iltiŕta prerromana a la Ilerda tardorromana. Nuevos datos tras dos décadas de investigación continuada en Lérida

La antigua Ilerda (Hispania Citerior), bien conocida por las fuentes escritas y numismáticas republicanas e iniperiales, no contaba con un conocimiento arqueológico paralelo. Las excavaciones continuas de los últimos años han variado sustancialmente este panorama (sobre todo las más recientes) y hoy podemos aventurar su evolución histórica con unos datos más fiables, así como ,su extensión y principales características por etapas.

SenSem: sentidos verbales, semántica oracional y anotación de corpus

En este artículo presentamos el desarrollo del proyecto SenSem (BFF2003-06456), que tiene como objetivo describir y representar el comportamiento léxico, sintáctico y semántico de los verbos del español. En el desarrollo de este proyecto se están construyendo dos recursos: un corpus de oraciones asociadas a su interpretación sintáctico-semántica y un léxico donde cada sentido verbal se asocia a un conjunto de ejemplos anotados del corpus.

El llatí epigràfic de la Safor valenciana

In this paper, the author studies the main characteristics of the Latin epigraphy from a Valentian region, the Safor. His intention is to give us a whole exposition of the linguistic material. The words are classified according to their morphological condition, being included sigla, abbreviations, etc. Nominative and dative cases proliferate. This corpus is anchored in a rural region, so that political personalities are missing.

Biblio-MetReS: A bibliometric network reconstruction application and server

Background: Reconstruction of genes and/or protein networks from automated analysis of the literature is one of the current targets of text mining in biomedical research. Some user-friendly tools already perform this analysis on precompiled databases of abstracts of scientific papers. Other tools allow expert users to elaborate and analyze the full content of a corpus of scientific documents. However, to our knowledge, no user friendly tool that simultaneously analyzes the latest set of scientific documents available on line and reconstructs the set of genes referenced in those documents is available. Results: This article presents such a tool, Biblio-MetReS, and compares its functioning and results to those of other user-friendly applications (iHOP, STRING) that are widely used. Under similar conditions, Biblio-MetReS creates networks that are comparable to those of other user friendly tools. Furthermore, analysis of full text documents provides more complete reconstructions than those that result from using only the abstract of the document. Conclusions: Literature-based automated network reconstruction is still far from providing complete reconstructions of molecular networks. However, its value as an auxiliary tool is high and it will increase as standards for reporting biological entities and relationships become more widely accepted and enforced. Biblio- MetReS is an application that can be downloaded from http://metres.udl.cat/. It provides an easy to use environment for researchers to reconstruct their networks of interest from an always up to date set of scientific documents.

Criterios identificadores de las perífrasis verbales del español

Dels criteris proposats per la bibliogra a per identi car les perífrasis verbals (PV) de l'espanyol, només alguns ajuden a diferenciar amb e càcia les PV de les construccions no perifràstiques (CNP). L'objectiu d'aquest article és revisar aquests criteris i avaluar-los per tal de determinar quins són realment vàlids per identi car les PV. L'avaluació s'ha dut a terme amb un grup de 15 conjunts verbals. A més, s'ha realitzat un estudi experimental amb corpus per determinar la productivitat de les PV detectades.

Corpus Oral Dialectal (COD). 1. Presentation_English

Aquest document conté el text Presentation_English, una introducció en anglès al CD del Corpus Oral Dialectal (COD). El COD és un component del Corpus de Català Contemporani de la Universitat de Barcelona (CCCUB), un arxiu de corpus de llengua catalana oral contemporània que ha estat confegit pel grup de recerca Grup d'Estudi de la Variació (GEV) amb la finalitat de contribuir a l'estudi de la variació dialectal, social i funcional en la llengua catalana. Aquest i altres materials del CCCUB són accessibles directament al Dipòsit UB o a través del web del CCCUB (http://www.ub.edu/cccub).

Corpus Oral Dialectal (COD). 1. Presentation_English

Aquest document conté el text Presentation_English, una introducció en anglès al CD del Corpus Oral Dialectal (COD). El COD és un component del Corpus de Català Contemporani de la Universitat de Barcelona (CCCUB), un arxiu de corpus de llengua catalana oral contemporània que ha estat confegit pel grup de recerca Grup d'Estudi de la Variació (GEV) amb la finalitat de contribuir a l'estudi de la variació dialectal, social i funcional en la llengua catalana. Aquest i altres materials del CCCUB són accessibles directament al Dipòsit UB o a través del web del CCCUB (http://www.ub.edu/cccub).

Surface passivation of crystalline silicon by Cat-CVD amorphous and nanocrystalline thin silicon films

In this work, we study the electronic surface passivation of crystalline silicon with intrinsic thin silicon films deposited by Catalytic CVD. The contactless method used to determine the effective surface recombination velocity was the quasi-steady-state photoconductance technique. Hydrogenated amorphous and nanocrystalline silicon films were evaluated as passivating layers on n- and p-type float zone silicon wafers. The best results were obtained with amorphous silicon films, which allowed effective surface recombination velocities as low as 60 and 130 cms -1 on p- and n-type silicon, respectively. To our knowledge, these are the best results ever reported with intrinsic amorphous silicon films deposited by Catalytic CVD. The passivating properties of nanocrystalline silicon films strongly depended on the deposition conditions, especially on the filament temperature. Samples grown at lower filament temperatures (1600 °C) allowed effective surface recombination velocities of 450 and 600 cms -1 on n- and p-type silicon.

ClInt: A bilingual Spanish-Catalan spoken corpus of clinical interviews

In this paper we present ClInt (Clinical Interview), a bilingual Spanish-Catalan spoken corpus that contains 15 hours of clinical interviews. It consists of audio files aligned with multiple-level transcriptions comprising orthographic, phonetic and morphological information, as well as linguistic and extralinguistic encoding. This is a previously non-existent resource for these languages and it offers a wide-ranging exploitation potential in a broad variety of disciplines such as Linguistics, Natural Language Processing and related fields.