<html><body><a href="http://infoling.org" target="_blank"><img
src="http://infoling.org/img/infoling.png" alt="" width="255"
height="50" align="left" border="0" /></a>
<br /><br /><br />
<br /><font style="font-size:80%"><table border="0" bordercolor="#FFF"
width="520px"><tr><td width="130"><a style="text-decoration:none"
target="_blank" href="http://infoling.org/revista/" title="Ir a
Infoling Revista"><img style="vertical-align:-30%;width:130px"
alt="Infoling Revista"
src="http://delicious.uab.es/infoling2/img/infolingIR.png"></a></td><td
width="100"><a style="text-decoration:none"
href="http://www.facebook.com/infoling" target="_blank"><img
border="0" src="http://infoling.org/img/facebook-icon.png" alt=""
style="vertical-align:-30%;height:16px;width:16px" /> Infoling en
Facebook</a></td><td width="180"><a style="text-decoration:none"
href="http://www.twitter.com/infoling" target="_blank"><img border="0"
src="http://infoling.org/img/t_small-b.png" alt=""
style="vertical-align:-30%;height:16px;width:16px" /> Infoling en
Twitter</a></td></tr></table></font>
<br />Moderador/a: Carlos Subirats (U. Autónoma Barcelona), Mar Cruz
(U. Barcelona)
<br />Editoras: Paloma Garrido (U. Rey Juan Carlos), Laura Romero (UB)
<br />Programación, desarrollo: Marc Ortega (UAB)
<br />Directoras/es de reseñas: Alexandra Álvarez (U. Los Andes,
Venezuela), Yvette Bürki (U. Bern, Suiza), María Luisa Calero (U.
Córdoba, España), Luis Cortés (U. Almería)
<br />Asesoras/es: Maite Taboada (Simon Fraser U., Canadá), Isabel
Verdaguer (UB), Gerd Wotjak (U. Leipzig, Alemania)
<br />Asesor legal: Daniel Birba
<br />Colaboradoras/es: Julia Bernd (Int'l Computer Science Institute,
EE.UU), Miroslava Cruz (U. Autónoma del Estado de Morelos, México),
Mercedes García de Quesada (U. Granada), Matthias Raab (UB), Antonio
Ríos (UAB)
<br />
<br />Con el patrocinio de:
<br /><table border="0" bordercolor="#FFFFFF"
width="200px"><tr><!--<td><a href="http://www.fundacioncomillas.es/"
target="_blank"><img
src="http://www.infoling.org/img/logo-comillas.png" alt="Fundación
Comillas" width="85" height="49" align="left" border="0"
/></a></td>--><td><a
href="http://www.arcomuralla.com/Arco/Shop/default.asp"
target="_blank"><img style="border:0;margin-top:10px"
src="http://www.infoling.org/img/logoarco.jpg" alt="Arco Libros"
width="62" height="34" align="left" border="0"
/></a></td></tr></table><br /><hr /><font style="font-size:90%">
<br /><b>Infoling 9.17 (2014)</b><br />ISSN: 1576-3404 </font>
<br /><font style="font-size:90%">© Infoling 1996-2014. Reservados
todos los derechos</font>
<br />
<br /><hr /><b>Tesis doctoral: </b><br />Burgos, Diego A.. 2014.
Towards an image-term co-occurence model for multilingual terminology
alignment and cross-language image indexing. Universidad Pompeu Fabra
(España), Institut Universitari de Lingüística Aplicada.<br
/><b>Enlace externo</b>: <a
href="http://www.tdx.cat/bitstream/handle/10803/145644/tdabh.pdf?sequence=1"
target="_blank">http://www.tdx.cat/bitstream/handle/10803/145644/tdabh.pdf?sequence=1</a><br
/><b>Información de:</b> Diego A. Burgos <burgosda@wfu.edu><br
/>Compartir: <a
href="http://api.addthis.com/oexchange/0.8/forward/facebook/offer?url=http://www.infoling.org/informacion/T129.html&pubid=ra-4def7f4a7565a706"
target="_blank" rel="nofollow"><img border="0"
src="http://infoling.org/img/facebook-icon.png" alt="Send to Facebook"
title="Send to Facebook"
style="vertical-align:-30%;height:16px;width:16px"
/></a> <a
href="http://api.addthis.com/oexchange/0.8/forward/twitter/offer?url=http://www.infoling.org/informacion/T129.html&template=@infoling%20Burgos,%20D.%20Towards%20an%20image-term%20co-occurence%20model%20for%20multilingual%20terminology%20alignment%20and%20cross-language%20image%20indexing%20{{url}}&pubid=ra-4def7f4a7565a706&shortener=bitly&bitly.login=infoling&bitly.apiKey=R_60e1d6b1cb688030e7759b835f63d0c0"
target="_blank" rel="nofollow"><img border="0"
src="http://infoling.org/img/t_small-b.png" alt="Tweet this"
title="Tweet this" style="vertical-align:-30%;height:16px;width:16px"
/></a><hr /><p><a
href="http://infoling.org/english/search/tesis/ID/129"
target="_blank">View with English headings</a></p><hr /><br
/><b>Director/a de tesis: </b>Leo Wanner <br /><br
/><b>Descripción</b><br /><p>This thesis addresses the potential that
the relation between terms and images in multilingual specialized
documentation has for glossary compilation, terminology alignment, and
image indexing. It takes advantage of the recurrent use of these two
modes of communication (i.e., text and images) in digital documents to
build a bimodal co-occurrence model which aims at dynamically
compiling glossaries of a wider coverage. The model relies on the
developments of content-based image retrieval (CBIR) and text
processing techniques. CBIR is used to make two images from different
origin match, and text processing supports term recognition, artifact
noun classification, and image-term association. The model aligns one
image with its denominating term from collateral text, and then aligns
this image with another image of the same artifact from a different
document, which also enables the alignment of the two equivalent
denominating terms. The ultimate goal of the model is to tackle the
limitations and drawbacks of current static terminological
repositories by generating bimodal, bilingual glossaries that reflect
real usage, even when terms and images may originate from noisy
corpora.<br /><br />Más información: <a
href="http://www.tdx.cat/handle/10803/145644"
target="_blank">http://www.tdx.cat/handle/10803/145644</a></p><br
/><b>Área temática:</b> Lingüística computacional, Lingüística
de corpus, Otras especialidades, Terminología<br /><br
/><b>Índice</b><br /><p>1. Introduction <br />1.1. The problem <br
/>1.1.1. Corpus-based dictionary analysis <br />1.1.2. Usage-based
dictionary analysis <br />1.2. Thesis proposal: the BC model <br
/>1.2.1. The Bimodal Co-occurrence hypothesis <br />1.2.2. Theoretical
Background <br />1.2.3. Implementing the BC hypothesis <br />1.3.
Assumptions and hypothesis <br />1.3.1. Assumptions <br />1.3.2.
Hypothesis <br />1.4. Objectives and delimitation of scope <br
/>1.4.1. General Objective <br />1.4.2. Specific objectives <br
/>1.4.3. Delimitation of scope <br />1.5. Fields of application <br
/>1.6. The Structure of the Thesis <br /><br />2. State-of-the-Art<br
/>2.1. Introduction <br />2.2. Image- and text-based information
retrieval systems <br />2.2.1. Image-term alignment <br />2.2.2.
Content-based image retrieval <br />2.2.3. Cross-language image
retrieval <br />2.2.4. Domain-specific CLIR <br />2.3. MWT and
artifact noun recognition <br />2.3.1. Multi-word term recognition <br
/>2.3.2. Artifact noun recognition <br /><br />3. Search Space
Characterization<br />3.1. Outline <br />3.2. Data macrostructure <br
/>3.2.1. General figures <br />3.2.2. Category delimitation <br
/>3.2.3. Category filtering and distribution <br />3.3. Data
microstructure <br />3.3.1. URL analysis <br />3.3.2. Analysis of BC
components <br />3.4. Attained goals and drawbacks <br />3.4.1.
Attained goals <br />3.4.2. Search space drawbacks <br /><br />4.
Nominal, Artifact MWT Recognition <br />4.1. Introduction <br />4.2.
Nominal MWT recognition <br />4.2.1. Preliminary considerations <br
/>4.2.2. Rule-based MWT recognition <br />4.2.3. Seed-based bootstrap
MWT recognition<br />4.2.4. Predefined-categories-based MWT
recognition <br />4.2.5. Evaluation of MWT recognition <br />4.2.6.
Results <br />4.2.7. Discussion <br />4.3. Artifact MWT recognition
<br />4.3.1. Anchor-based selection <br />4.3.2. Noun classification
<br /><br />5. Implementation of the BC-Model <br />5.1. Introduction
<br />5.2. System overview <br />5.3. System Design <br />5.3.1.
Behavioral Model <br />5.3.2. Structural/Architectural Model: <br
/>5.4. User interface <br />5.5. System evaluation <br />5.5.1.
Evaluation of the CBIR component <br />5.5.2. Evaluation results <br
/>5.6. Remarks <br /><br />6. Conclusions <br />6.1. Hypothesis
validation <br />6.2. Objectives attainment <br />6.3. Contributions
<br />6.3.1. Characterization of the problem <br />6.3.2. The BC model
<br />6.3.3. The Search Space <br />6.3.4. MWT and artifact noun
recognition <br />6.3.5. Prototype <br />6.4. Limitations <br />6.4.1.
The BC model <br />6.4.2. MWT and artifact noun recognition <br
/>6.4.3. Prototype <br />6.5. Lines of future research <br /><br
/>References<br />Annexes</p><br /><b>Fecha de lectura o defensa:</b>
19 de mayo de 2014<br /><br /><b>Número de págs.:</b> 186<br /><br
/><b>Correo-e del autor/a:</b> <burgosda@wfu.edu><br /><br
/><b>Información en la web de Infoling:</b><br /> <a
href="http://www.infoling.org/informacion/T129.html" target="_blank">
http://www.infoling.org/informacion/T129.html</a></body></html>