Question and answering service is one of the popular services in the World Wide Web. The main goal of these services is to finding the best answer for user's input question as quick as possible. In order to achieve this aim, most of these use new techniques foe question matching. . We have a lot of question and answering services in Persian web, so it seems that developing a question matching model might be useful. This paper introduces a new question matching model for Persian. This model is based on statistical language model and employed generalized bigram and trigram model. We also describe some results regarding the employment natural language processing in question matching model. Most of the Q&A services have large number of questions and answers; hence we considered an optimized implementation for the model. We evaluated our model with Rasekhoon question and answering archive which contains about 18000 pairs of questions and their answers. The results show the improvement of precision and recall measures by using this model.