AM07 START Conference Manager    

Study on the Influence of Vocabularies used for Image Indexing in a Multilingual Retrieval Environment

Elaine Ménard and Clément Arsenault

(Submission #1)


Summary

The Internet constitutes a vast universe of knowledge and human culture, allowing the dissemination of ideas and information without borders. The Web also became an important media for the diffusion of multilingual resources. However, its “worldwide” expansion is a rather recent phenomenon. Linguistic differences still form a major obstacle to scientific, cultural, and educational exchange. Besides this linguistic diversity, more and more databases and collections now contain documents in various formats (text, image, video...), which may also adversely affect the retrieval process. This research project aims to verify the existence of relations between each of these two indexing approaches: (1) traditional image indexing with a metadata schema recommending the use of controlled vocabularies or (2) free image indexing using natural language, and their respective performance for image retrieval, in a multilingual context. This research will verify if one or the other of these indexing approaches surpasses the other, in terms of effectiveness, efficiency, and satisfaction of the image searchers. In order to achieve this goal, three specific objectives were defined: (1) identify the characteristics of each of the two approaches used for image indexing, in a multilingual context; (2) explore the existence of relations between the characteristics of each of the two approaches used for image indexing and the retrieval performance, in a multilingual context; and finally, (3) determine the direction of the relations between the characteristics of each of the two approaches used for image indexing and the retrieval performance, in a multilingual context. This contributed research poster presents the context and the related problems of this research project. The experiment, which will be carried out to answer the research questions, is also described, as well as the data collection methods. This research poster will also present preliminary results of image retrieval, in a multilingual context.

 


  
START Conference Manager (V2.54.4)
Maintainer: asis@asis.org