Aleph or Aleph-Maddah, That is the Question! Spelling Correction for Search Engine Autocomplete Service

In this paper, we proposed a combinational framework, called Aleph checker (similar to spell checker), to select the correct form of Aleph in Perso-Arabic words. This system tries to check spelling of Persian words by using user queries of a web search engine. For each word, a Soundex algorithm extracts a list of candidates, which are suspicious to be misspelled of the word. An Ngram based filtering, prunes the list of equivalences and the remained items are counted to generate the frequency of different spellings of the word. At the last step, correct spelling is determined by a linear binary classifier. In conducted experiment, the system shows the accuracy of 0.92 in determining correct form of Aleph.