Title :
A feature selection technique based on equivalent relation
Author :
Wang, Meng ; Sun, Shudong ; Niu, Ganggang ; Tu, Yuanzhi ; Guo, Shihui
Author_Institution :
Inst. of Syst. Integration & Eng. Manage., Northwestern Polytech. Univ., Xi´´an, China
Abstract :
Traditional feature selection for classification can only treat with decision tables. The paper proposes a feature selection technique based on equivalent relation, which works well for both decision tables and information systems. The approach can compute the significance attribute from a database. The measure also takes into account the attribute support degree in decision tables. The experiment results confirm that the performance of the approach is efficient.
Keywords :
decision tables; feature extraction; pattern classification; decision tables; equivalent relation; feature classification; feature selection technique; information systems; Accuracy; Classification algorithms; Information systems; Machine learning; Pain; Pattern recognition; Search methods; attribute significance; equivalent relation; feature selection; information system;
Conference_Titel :
Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC), 2011 2nd International Conference on
Conference_Location :
Deng Leng
Print_ISBN :
978-1-4577-0535-9
DOI :
10.1109/AIMSEC.2011.6010707
