Industrial Projects


Aim: Upgrading/proposing new spell and grammar checking algorithms for Persian language and implementing the related software.

For: Software and Information Technology Group, Information Technology Research Center, Iran Telecommunication Research Center.

Abstract: With advancements in industry and information technology, large volumes of electronic text are produced daily. In spite of having considerable benefits such as easy organization and data management, electronic documents are too time taking and expensive to maintain. Therefore, existence of automatic systems such as spell checker/corrector can help in reducing costs and increasing the electronic texts.

This project aims at improving existing algorithms for automatic correction of Persian texts to be used to develop an automatic spell checking system. Written mistakes are usually categorized as spelling, grammatical and real word errors. If a word is not observed in the word list or can not be indirectly produced using morphological rules, this word is considered to have spelling error. Grammatical errors which happen in sentence level or word relation involve inaccordance error (eg. object and verb), order of words in a sentence, and in general anything which does not agree with grammatical rules of that language. Real word errors correspond to the wrong usage of a word in a sentence. For instance if the word “match” and “catch” are both correct words which are available in word list, but if the latter one appears in a sentence where the first one was considered to be, there exists a real word error.