何琳,刘竟,侯汉清.基于《中图法》的多层自动分类影响因素分析[J].中国图书馆学报,2009,35(6):
An Analysis of the Impact Factors in the Multi Layer Automatic Classification Based on CLC
基于《中图法》的多层自动分类影响因素分析
  
DOI：
Key words:CLC,Multi layer hierarchy,Automatic classification,Impact factors
中文关键词:  中图法，多层分类，文本分类，影响因素
基金项目:
Author NameAffiliation
He Lin 南京农业大学信息管理系 南京 210095 
Liu Jing 江苏大学 苏州 212013 
Hou Hanqing 南京农业大学信息管理系 南京 210095 
Hits: 4324
Download times: 3711
Abstract:
In former automatic text classification research，most of the prevalent classification technologies divided texts into several classes at one layer．However，with the increased quantities of information retrieval，this flat organizational classification is more and more unsuitable to the information retrieval task. This paper tries to analyze the impact factors in the multi layer classification, including training data, classification algorithms, classification hierarchy systems and evaluation method. It also discusses the main difficulties and potential solutions in the multi layer text classification. 4 figs. 2 tabs. 9 refs.
中文摘要:
      系统总结基于《中图法》知识库的多层自动分类项目的研究经验，分析训练数据、特征词选择、分类算法、类目体系和评估方法等因素对多层自动分类的影响。围绕《中图法》，对自动分类的适应性、稀有类别的处理、知识库更新、明显正确或错误数据的标注、标准数据集的制定等进行探讨。图4。表2。参考文献9。
View Full Text   View/Add Comment  Download reader