摘要
提出了一种基于文本集密度的特征词选择与权值计算的方法AMTW (ApproachofModifyingTermWeighting) .该方法可以找出不损失文本有效信息的最小特征词语集 ,设计出更为合理权值计算方案 .
A method of feature selection and weighting scheme based on text set density is proposed, by which the set containing least elements and representing all variable information of a text can be found. A more reasonable weighting scheme is presented also, with its validity proved by an evaluating criterion, meta-scoring.
出处
《山东大学学报(工学版)》
CAS
2004年第3期92-95,共4页
Journal of Shandong University(Engineering Science)
关键词
信息检索
文本集密度
权值计算
元打分法
information retrieval
text set density
weighting
meta-scoring