A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Advanced Technologies

Thesis (M.Eng.) - Asian Institute of Technology, 1998

With the recent explosive growth of the Internet resources, users are tending to seek information from a digital libra1y across national boundaries. The ability to query large multilingual collections using a single language could be beneficial. Multilingual Information Retrieval (MIR) is used to take queries in one language and retrieve relevant documents written in other languages. It has been found that the current MIR system leads to a drop in effectiveness of 40- 60% of monolingual retrieval. Hence, it is necessary to find a way to improve the performance of translation and searching in MIR services. In this research, a system model of MIR is proposed by using interactive query formulation under Resource Description Framework (RDF). The user query is structured in the same way as the resource description and then formulated according to the knowledge base and basic grammar rules assisted by user interaction. The system adopts dictionary and corpora for initial translations of the user query, and then interacts with user to reduce the ambiguities of such translations. This approach provides more effective results for query translations and also provides more information for retrieval as the Internet resources are described in the same structure. For the above theoretical system model, a simplified application system is implemented to achieve and simulate part of abilities. It is an application mainly used in Internet and implemented in Java, Java Servlet, and metadata database under RDF as simulant collections.