Research on template automatic extraction based on systematic cluster

With the development of the storage technology and the increase of the storage capacity, many large-scale corpuses have been built up. But the corpuses include much redundant information. It leads to the low translation efficiency and speed. In order to get a highly accurate result and high translation speed, it¿s necessary to reduce the redundant information by extracting template. This paper proposes a method based on system cluster to extract templates from the corpuses and the detail algorithm about how to extract template. The experiment results show that the method has a good performance in extracting template.