文献类型： 会议论文

第一作者： Yuexiang Yang

作者： Yuexiang Yang ¹ ; Shouhui Pan ² ; Bo Xu ³ ; Yiyang Wang ³ ; Chao Lei ³ ;

作者机构： 1.China National Institute of Standardization

2.Beijing Research Center for Information Technology in Agriculture, Beijing Academy of Agriculture and Forestry Sciences

3.School of Economics and Management, Beihang University

关键词： Quality and safety;Web mining;Text classification;Consumer product

会议名称： International Conference on Advanced Computer Control

主办单位：

页码： 47-54

摘要： At present, the research of quality and safety Web information mining is still in its infancy, and the quality and safety information on Web is not effectively explored and utilized due to the lack of relevant theoretical and technical support. A multi-level Web text classification method is proposed and a three-tier Web text classifier is constructed for classifying the quality and safety Web text. The first- and second-tier classifiers are both rule-based classifier, which classify the Web text by the custom-made rule template, whereas the third-tier classifier is a KNN classifier based on machine learning. The effect of HTML tags is taken into account sufficiently for classification, and an improved term weight method based on vector space model is proposed. The experimental results show that our proposed method can effectively classify the quality and safety Web text.

分类号： TP3-53