Applications using multilinguality: IR, summarization and generalization

MT Summit VII Sept. 1999 Applications Using Multilinguality: IR, Summarization and Generalization Vilas Wuwongse Computer Science and Information Management Program School of Advanced Technologies Asian Institute of Technology P.O. Box 4, Klong Luang, Pathumthani 12120, Thailand vw@cs.ait.ac.th Abstract Among the three main applications using multilinguality, i.e., information retrieval, summarization and text generation, the first one could be considered to be the core and the other two its supporting technologies. Information retrieval using multilinguality often appears in the form of allowing a query specified in a language to be answered by documents or information in one or more different languages. Summarization supports information retrieval by producing a database of intermediate representation of original documents, which contains only central and essential information. Text generation with multilingual capability helps create retrieved information in a desirable natural language. This brief paper identifies some issues regarding these three applications with emphasis on information retrieval. 1 Introduction With the rapid increase in the number and variety of information resources, there is a growing need for some techniques for efficient access to these resources. The authors of these resources want their documents to be translated and retrieved with greater accuracy, and thus to have a better chance to reach the right kind of readers. On the other hand, the readers wish to find the best- matched resources without wasting time in non-related documents. Moreover, appearances of these resources in more than one language is becoming common, i.e. (Peter and Picchi, 1997): ― in countries with more than one natural language ― in countries where both the national language and English are commonly used for scientific and technical documentation ― in Pan-European institutions such as research consortia ― in multinational companies Multilingual Information Retrieval (MIR) or Cross- Language Information Retrieval (CLIR) is defined as taking query in one language and retrieving relevant documents in one or more languages, not necessarily different from the query's language. The term MIR is less widely used than CLIR because MIR can also imply within-language retrieval in more than one language without anything to do with cross-language capability. Apart from enabling access to other language materials, MIR or CLIR also helps eliminate input limitations, search for non-text materials as well as increase efficiency by obviating multiple queries and allowing a user to use his most fluent language. 2 Approaches to CLIR Simple employment of bilingual dictionaries can easily yield 50% of monolingual information retrieval effectiveness. However, if higher effectiveness is desirable, more sophisticated techniques specially devised for CLIR must be utilized. Some basic approaches to CLIR are (Oard, 1997): • Controlled Vocabulary Retrieval: requires the indexing (manually or automatically) of multilingual documents using a predetermined vocabulary and that users express the query using terms drawn from the same vocabulary. • Free Text Retrieval which includes two basic approaches: ― Knowledge-based: can be further divided into dictionary-based and ontology-based, these two approaches are not mutually exclusive, however, the trend in cross- language free text retrieval research is to - 155- MT Summit VII __________________________________________________________________ …

Content truncated to 500 words