MorphoLogic, the only Hungarian private institution active in research and development solely in the field of natural language processing, was established in 1991 by several researchers of various academic laboratories. Their mission has been to develop
The vocabulary of morphologically complex languages such as Hungarian, Polish or Czech cannot be represented on a computer as simple word lists because they would exceed the resources of the entire computer system. So MorphoLogic had worked out other models of morphologically describing these languages. This resulted in a morphological analyzer called HUMOR (High-speed Unification MORphology) capable of handling billions of word forms even with very limited computer resources.
This model turned out to be a great advantage for languages with less complex morphologies such as English and even German, and now MorphoLogic can say it has a language independent technology as the basis of proofing tools and other language-related software modules. Currently, MorphoLogic has morphological descriptions for the English, German, Hungarian, Polish, French, Spanish, Czech, Romanian, Slovak and some minor languages.
Based on the HUMOR morphological analyzer, MorphoLogic has created a wide range of products. These include proofing tools - spelling checkers, hyphenators, grammar checkers, and thesauri-, and there are many linguistic modules - such as linguistic stemming tools, morphological and syntactic analyzers - that can be integrated with other online/offline applications and services.
From the beginning, MorphoLogic has been committed itself to create translation support and machine translation tools. The 5 year-long research since 2001 resulted in the MetaMorpho machine translation engine, which is the basis of the recently released English-Hungarian-English translation application range.
- a flexible language engineering platform-for morphology and syntax first-in order to describe a wide range of languages in a uniform way, and, at
the same time, make marketable language technology applications;
- implementations of all types of proofing tools for many languages;
- an intelligent, easy to use dictionary system for various languages;
- implementations of linguistic modules supporting intelligent text processing, free text search and database indexing;
- methods for information extraction from documents and develop a representation formalism that is suitable for further automatic processing;
- machine translation systems for the common European linguistic infrastructure.