Title :
A Text Association Rules Mining Method Based on Concept Algebra
Author :
Feiyue Ye ; Jiannan Xiong ; Lingyu Xu
Author_Institution :
Sch. of Comput. Eng. & Sci., Shanghai Univ., Shanghai, China
Abstract :
The Concept Algebra (CA) based text representation method which can be auto-constructed and used in ordinary texts´ includes more semantic information compared to Keyword methods. This method provides a new way of thinking of more accurately mining meaningful association rules. In this paper concepts association rules mining is based on CA represented texts, and use the relations among concepts to optimize the mining results with definition and calculation of association weight and rule strength. Compared with traditional method regarding keyword as independent to each other lacks semantic information while domain-oriented method would be applicable to specific areas with specialists construction. Method in this paper use CA´s superiority on utilizing Ri and Ro of concepts to remedy the two deficiencies effectively which is proved by the theoretical and experiment.
Keywords :
algebra; data mining; text analysis; CA; concept algebra; domain-oriented method; keyword methods; semantic information; text association rules mining; text representation; Algebra; Association rules; Context; Educational institutions; Knowledge representation; Semantics; Concept Algebra; association rules; auto-constructed; concept relation; rule strength; semantic;
Conference_Titel :
Green Computing and Communications (GreenCom), 2013 IEEE and Internet of Things (iThings/CPSCom), IEEE International Conference on and IEEE Cyber, Physical and Social Computing
Conference_Location :
Beijing
DOI :
10.1109/GreenCom-iThings-CPSCom.2013.406