Authors
Andreas Hense and Florian Quadt (Bonn-Rhein-Sieg University of Applied Sciences).
Abstract
From the very beginning, the Internet has been used increasingly by researchers of different mother tongues and research areas. This linguistic diversity is reflected in multilingual web content and scientific publications. While machine translation is the most desirable grade of multilingual collaboration, fully automated and technically mature translation systems still do not exist. Nevertheless, it is possible to use the knowledge of how to translate that has been gathered so far for an application that is not as complex as machine ranslation: cross-language information retrieval (CLIR).
CLIR relies on the fact that users of search engines can read and understand more than one single language. This is why search engines supporting CLIR present multilingual hits to the user. We will discuss the advantages of using a concept-based interlingua and resent a prototype of an interlingua-based CLIR system that has been tested with English, German, and Japanese documents. Additionally, we will focus on selected challenges in information retrieval as well as in machine translation.
About the speaker
Florian Quadt studied computer science at the Bonn-Rhein-Sieg University of Applied Sciences and received a master's degree in 2006.
During his studies he focused on information retrieval and implemented a cross-lingual search engine in his master thesis. He worked for an IT consultancy for two years, specializing in code analysis, architecture evaluation and verification. In April 2009 he returned to Bonn-Rhein-Sieg University as project manager of a research project that aims at the development of a publication system for meteorological experiment data