基于粗糙集的多标签文本分类算法  

Multi-Label Text Classification Based on Rough Set

作　　者：吕小勇[1] 石洪波[1] 

机构地区：[1]山西财经大学信息管理学院,山西太原030006

出　　处：《广西师范大学学报（自然科学版）》2009年第3期150-153,共4页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金资助项目(60873100);山西省自然科学基金资助项目(2009011017-4)

摘　　要：将粗糙集优越的约简理论应用于多标签文本分类,提出了基于粗糙集理论的多标签文本分类算法,该算法利用训练阶段得到的各个类别的分类规则与测试实例逐一匹配,得出实例的类标签集合,扩展了粗糙集理论在文本分类中的应用,实验证明算法有效可行。Range of applicat of Text classification has been widely concerned,but the researchers mainly research single-label classified,and have fewer studies on multi-label relatively. Rough Set Theory is a hot spot in the academic community currently, and researchers have recently applied it to single-label text categorization. The excellent reduct feature of rough set theory will be applied to multi-label text classification,and a multi-label text classification algorithm based on rough set is proposed,which uti- lizes classification rules of each category acquired in the training phase to match test instance one by one and obtains the label set of the instance. The algorithm expands the application of rough set theory in the text classification. The experiment result shows that the algorithm is effective and feasible.

关 键 词：多标签 粗糙集 约简 分类规则 匹配 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP181[自动化与计算机技术—计算机科学与技术]