Research Of Mongolian Information Retrieval Model Based On Markov Random Field

With the development of computer science and internet, the web is becoming a universal repository of human knowledge where user now can get abundant information rapidly. How to get exact information from the repository is becoming a focus of user. In this context, the information retrieval has been developing vigorously.At present, information retrieval has obtained good results in the study of the Chinese English,Japanese etc. But because of the difference of each language, it is still short of the research on minority languages. And it is severity encumbrance to the spread of minority languages. Mongolian minority languages in China plays an important role, but also very influential in the world. Mongolian information increasingly rich, has promoted the development of Mongolian information retrieval. As a result, more and more people are interest in Mongolian information retrieval system.In this paper we use some theories and methods of Markov random fields in the study of the M-ongolian Information Retrieval.A Markov random fields is undirected graphical network that is ca-pable of efficiently representing relevance in knowledge with strong learning and inferring capabili-ty. According to the Mongolian language characteristics, we propose Mongolian Information model based on Markov random fields that put MRF model and Mongolian language structure combinatio-n. Through a lot of experiment, Analysis and verify the performance of Mongolian Information mo-del based on Markov random fields:Full independence model,Phrase dependence model and Mod-ification dependence model. The experiments show that our models make significant improvements in retrieval performance.