Abstract：A method based on statistics of automatically generating information of Chinese character and it'sneighbours is described in this paper.it's also applied in Chinese automatic abstracts. It can also automat-ically learn during the building and using of the system.
姜贤塔;陈根才. 利用字频统计及机器学习的自动文摘方法[J]. 浙江大学学报（理学版）, 1997, 24(3): 220-225.
Jiang Xianta Chen Gencai. Automatic Chinese Abstracts Based on the Statistics ofChinese Character Frequency and Machine Learning. Journal of ZheJIang University(Science Edition), 1997, 24(3): 220-225.