Shima Foolad - Electrical and Computer Engineering Department, Semnan University, Semnan, Iran
Farzin Yaghmaee - Electrical and Computer Engineering Department, Semnan University, Semnan, Iran

in applications such as search engines the ability of word prediction along as user typing is critical especially for non-native persons who use a foreign language. Some works have done in this topic for English language and very few works in Persian language. In this paper we propose a method for Persian word based on her/his previous search phrases and N-gram language model. Our experimental results show that the proposed method decreases the number of keystrokes of a Persian user about 55%. This means that user can type words with twice speed.