Chinese-mining Preprocessing Technology Based on Text Trait Optimizing

How to get the target text quickly becomes a technical limitation with the using of massive data. While obtaining the Chinese target information, the segmentation of the sentence is supposed to be the key according to research. To mine the segmentation of English text is relatively simple for the space is used as a interval, meanwhile the Chinese segmentation is much more difficult. So in this paper the reciprocal crossing segmentation algorithm and the trait-optimizing vector model are designed to improve the mining efficiency of Chinese information. Based on dictionary, an improved segmentation algorithm is adopted in text pretreatment processing, which is based on vector space module, to do experiments on the segmentation algorithm and to analyze the segmentation results. And that segmentation algorithm is already proved to be very effective in the text mining of text trait vector module.