计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (4): 226-229.DOI: 10.3778/j.issn.1002-8331.2010.04.071

• 工程与应用 • 上一篇    下一篇

现当代文学作品的作者身份识别研究

年洪东,陈小荷,王东波   

  1. 南京师范大学 文学院,南京 210097
  • 收稿日期:2008-09-18 修回日期:2008-12-05 出版日期:2010-02-01 发布日期:2010-02-01
  • 通讯作者: 年洪东

Research on authorship attribution of contemporary literature

NIAN Hong-dong,CHEN Xiao-he,WANG Dong-bo   

  1. School of Chinese Language and Literature,Nanjing Normal University,Nanjing 210097,China
  • Received:2008-09-18 Revised:2008-12-05 Online:2010-02-01 Published:2010-02-01
  • Contact: NIAN Hong-dong

摘要: 主要利用了SVM统计机器学习模型对中国现当代文学八位代表人物的作品进行了作者身份识别研究,在识别过程中选取了以词汇为基础的多种统计量作为识别特征,并且采取了基于低密度多特征的训练方法,在跨文体的作品的作者身份识别中取得了非常优异的识别性能。

关键词: 作者身份识别, 机器学习, 计算风格学, 现当代文学

Abstract: This paper uses the statistical model(SVM) for the identification of the author of contemporary Chinese literature works to eight representatives.In the identification process to select a vocabulary based on a variety of statistics as identifying features,and to take training methods based on the low-density and more features,having achieved better result in cross-style works of the author identification.

Key words: authorship attribution, machine learning, computational stylistics, contemporary literature

中图分类号: