MetaMorpho: A Pattern-Based Machine Translation System
Gábor Prószéky, László Tihanyi · Repository of the Academy's Library (Library of the Hungarian Academy of Sciences) · 2002
This paper describes an efficient real-time comprehension assistance and machine translation method. Combining the advantages of example-based (EBMT) and rule-based machine translation (RBMT), a new paradigm, pattern-based translation is presented. A system based on these principles that features an innovative user-friendly interface has been built. Called MetaMorpho, the system has been tested for English-Hungarian translation, and showed very promising results both in translation quality and speed.