Research of Automatic Indexing Based on Semantic and Statistic Feature

Automatic indexing is the foundation and core technology of automatic documents processing. Currently most of the documents don´t have Keywords, and manual indexing consumes too much time and laborious, it is also highly subjective. This paper discusses the automatic indexing method on the calculation of a statistical and semantic analysis. In the basis of statistical methods, the use of semantic information to improve the accuracy of indexing. At the same time, based on the characteristics of the corpus, the right to the feature selection and optimization in terms of aspects of the program. Experiments show this method not only improves the accuracy of indexing, and improve efficiency.