| Persian Linguistics | System Architecture | Dictionary Structure | Demos | Publications | Persian Resources |
| Persian Linguistics | System Architecture | Dictionary Structure | Demos | Publications | Persian Resources |
|
The Shiraz machine translation system is a MT prototype that
translates Persian text into English. The project began in October 1997
and the final version was delivered in August 1999. The system uses typed feature
structures and an underlying unification-based formalism to describe
Persian linguistic phenomena. It is able to run on Unix as well as
Windows machines. The Shiraz system uses an electronic bilingual
Persian to English dictionary consisting of approximately 50,000
terms, a complete morphological analyzer and a syntactic parser. The
system components were tested on a bilingual tagged
corpus developed from a large Persian corpus of on-line material
(approximately 10MB). The machine translation system is mainly
targeted at translating news material.
The dictionary was built by a team of Persian lexicographers and includes single words, compounds and phrasal expressions. The dictionary contains information about the orthography, morphosyntactic category and syntactic properties of lexical items as well as the English word-sense equivalents. The current system performs tokenization and full morphological analysis. Compounds and light verbs are also recognized. The syntactic parser can analyze noun phrases (including relative clauses), preposition phrases and basic sentential constructions. |
|
Rémi Zajac, Project Manager
Karine Megerdoomian, Computational Linguist Jan Amtrup, Computer Specialist Hamid M. Rad , Computational Linguist, Technical Writer Mohammad Reza Aidinejad, Lexicographer Jane Freider, Computer Specialist Mike Freider, Computer Specialist |
|