Main purpose is to extract key words and collocations, which is deviant to Zipf's distribution law for given language. These collocats then combined to text using weighted MDP, which is not suitable for such task, it is better to use neural networks for such task.