Term Builder: a lexical knowledge acquisition tool for the Logos machine translation system

MT Summit VII Sept. 1999 TermBuilder: A Lexical Knowledge Acquisition Tool for the Logos Machine Translation System Brigitte Orliac Kutz Arrieta Logos Corporation Logos Corporation 100 Enterprise Drive, Suite 501 601 South Washington Street, Apt. 301 Rockaway, NJ 07866 Seattle, WA 98104 Abstract Logos 8, the next generation of the Logos Machine Translation (MT) system, is a client server application, which realizes the latest advances in system design and architecture. A multi-user, networkable application, Logos 8 allows Internet or Intranet use of its applica- tions with client interfaces that communicate with dictionaries and translation servers through a common gateway. The new Lo- gos 8 technology is based on a relational da- tabase for storage and organization of the lexical data. In this paper, we present Term- Builder, the Lexical Knowledge Acquisition tool developed for Logos 8. The new auto- matic coding functionality within Term- Builder is significantly improving the process of acquiring new lexicons for MT and other applications. 1 Introduction Machine translation may be succinctly defined as the mapping of one language into another by elec- tronic means. Mapping between two languages can be “direct” or it can be achieved by means of an interme- diate representation, in a transfer-based or interlin- gua-based approach. In the transfer-based approach, the source language is mapped into an abstract repre- sentation that retains language-specific information (the result of analysis). Bilingual modules convert the source language representations into equivalent target language representations. These are in turn input to a generation module. In the interlingua-based approach, the source lan- guage is mapped into one or more language-neutral representations (such as an ontology or a knowledge base) from which the target language is generated. In its current incarnation, the Logos MT system is a hybrid system which combines features of both rule-based and interlingua-based systems. In the next section, we present the basic characteristics of the new Logos MT system, hereafter referred to as Logos 8. In section 3, we discuss the major issues surrounding the design, maintenance and development of the lexical module in linguistic applications. In section 4, we introduce TermBuilder, the Lexical Knowledge Ac- quisition tool developed for Logos 8. We conclude the presentation of TermBuilder by listing remaining is- sues and future development items. 2 Logos 8: the next generation 2.1 Logos 8 System Architecture Logos 8 represents the next generation of Logos language products. Logos 8 is a client server applica- tion, which realizes the latest advances in system de- sign and architecture. A multi-user, networkable ap- plication, Logos 8 allows Internet or Intranet use of its applications with client interfaces that communicate with dictionaries and translation servers through a common gateway. The Logos client interfaces and gateway are written in Java to run on all platforms while the translation server is written in C++. The three client interfaces are: • Translation Client for the submission and re- trieval of terminology extraction and translation jobs • TermBuilder for terminology management • Administration Client for user registration and permissions Logos 8 …

Content truncated to 500 words