Constructing Chinese-English Concept Space

The information available in languages other than English on the World Wide Web is increasing significantly. According to a report from Computer Economics [1], 54% of Internet users are English speaking. However, it is predicted that there will be only 60% increase in Internet users among English speakers but there will be 150% growth among nonEnglish speakers for the next five years. By 2005, 57% of Internet users will be non-English speakers. A report by Techserver [2] shows that the number of Internet servers in China has increased from 505,000 to 1.175 million from January to June of 1999. All of these evidences reveal the importance of cross-lingual research to satisfy the needs in the near future. Cross-lingual information retrieval has been one of the focuses of digital library research recently. The NSF/DARPA/ANSA funded Digital Library Initiative 2 in U.S. is actively promoting activities and processes that cross the boundaries of language and politics. A new program in International Digital Libraries Collaborative Research is initiated [23]. The Internet 2 project will also be multilingual. In the Digital Library Initiative 1, researchers have done extensive work on structural interoperability and semantic interoperability. Searching and retrieving objects across variations in protocols, formats and disciplines are widely explored. However, research in crossing language boundaries, especially across European languages and Oriental languages, is still in the initial stage. In this proposal, we'll focus on cross-lingual semantic interoperability. In this paper, we present the construction of the Chinese-English cross-lingual concept space by Hopfield network based on a parallel corpus. Such concept space is important for building an accurate cross-lingual information retrieval system. Experiments are conducted to mesure the precision of the constructed concept space and the performance of the algorithm.