The foundation for GSI >Erli’s many applications is its linguistic engineering toolbox ALETH, not to be confused with ALEP, the Commission,funded development environment, in which the company is also involved (see sidebar page five). ALETH, which is written in C and C++ and operates under the Unix operating system, has been under continual development at GSIfor a decade. ALETHDic is GSI’s multilingual dictionary for French (70,000 entries) and English (50,000 entries) and coded with morphological, syntactic, and semantic information in GENELEX,compliant format. ALETHGram is GSI’s collection of grammar rules which are can be used for a variety of applications, such as morphological analysis and gen, eration, recognition of noun phrases, indexing, natural language-to-SQL translation, and others. GSI currently has grammars for French and Engl, ish, with grammars for German, Italian, and Spanish under development. ALETHTrad is a col, lection of tools for analyzing translation source texts for repetitions, consistency, and unknown terms. In addition, ALETHCorp is GSI-Erli’s statistical-based corpus toolbox, ALETHGT is its thesaurus and termbase manager, ALETHGD its dictionary manager, and ALETHGen is a text generation module. lastly, ALETHInd is GSI-Erli’s automatic text indexing and natural language query tool. At the moment, the GSI-Erli is working with the US company Verity to integrate Erli’s linguistic modules into Verity’s Topic information retrieval system.

(See  article  that this sidebar accompanied)