Some Problems On Natural Language Processing For Mathematical Academic Literature

Since the beginning of reform and opening-up,the scientific research in China has developed rapidly.During these years,Mathematics of our country made great progress under the spotlight all over the world.Especially in the recent decades,articles written by Chinese Mathematicians published in major journals are impressive in any respect.And with no doubt,all of these articles are published in English,which is one of the most universal languages.Chinese word segmentation is a pretty important study direction in translation,the speed and accuracy of which influence directly whether the work of translation could be finished successfully.Therefore,good method of Chinese word segmentation should have both high accuracy rate and the advantage of segmenting quickly.In the developing modern society,the coming of new words is a common thing.Then how could we deal with such situation,how to make use of the computer to distinguish and split the words is a problem.If the computer could not distinguish the new words well,it could be tricky and may affect the process of the research.Although the traditional method of segmentation is relatively effective,it is not useful for new words as we mentioned before or context with ambiguity for segmentation.Dictionary is the central element of mechanical word segmentation,and then the renew of dictionary caused by the coming new words is also a difficulty.As far as I am concerned,word the smallest unit of context with specific meaning.So the priority is to segment the words we need with the help of Chinese word segment technology.However,the successive Chinese words in the context could not be segment with blank space like English,traditional method cannot determine which Chinese words appear in the sentence.So a better method is needed to deal with this situation.