| 基于关联规则的科技文献分类研究 |
| |
| 引用本文： | 周丽红,刘勘.基于关联规则的科技文献分类研究[J].图书情报工作,2012,56(4):12-119. |
| |
| 作者姓名： | 周丽红 刘勘 |
| |
| 作者单位： | 中南财经政法大学信息与安全工程学院 |
| |
| 基金项目： | 教育部人文什会科学项目 |
| |
| 摘 要： | 针对面向结构特性的科技文献分类问题，通过关联规则的分类方法将科技文献划分为不同的类型：综述型、理论型和应用型。首先对科技文献数据分词等进行预处理；然后通过PredictiveApriori关联算法挖掘关于类别特征项的频繁项集，构造科技文献分类的分类器；接着对分类科技文献进行分类规则匹配，判定所属类别；最后通过实验对分类性能进行评估，并通过对比证明了本方法的有效性。
|
| 关 键 词： | 科技文献 关联文本分类 PredictiveApriori算法 |
| 收稿时间： | 2011-10-25 |
| 修稿时间： | 2011-09-06 |
Scientific Literature Classification Based on Associated Rules |
| |
| Zhou Lihong Liu Kan.Scientific Literature Classification Based on Associated Rules[J].Library and Information Service,2012,56(4):12-119. |
| |
| Authors: | Zhou Lihong Liu Kan |
| |
| Institution: | Information and Safety Engineering School, Zhongnan University of Economics and Law, |
| |
| Abstract: | For the problem of scientific literature classification by structural properties,an associative classification method based on frequent item sets has been proposed.There are three types of scientific literature:Review,theory-based and applied.Firstly,pretreat the scientific literature data.Then,use the Predictive Apriori algorithm to mine frequent item sets among categories and construct the rule base.After that,match the classification rules for the literatures which waiting for classification.Finally,assess the performance of classification.Experimental results show that,the scientific literature associative classification method by structural properties has higher classification performance. |
| |
| Keywords: | scientific literature associative text classification PredictiveApriori algorithm |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《图书情报工作》浏览原始摘要信息 |
| 点击此处可从《图书情报工作》下载免费的PDF全文 |
