作者: Zhonghua Tang , Qin Liao
DOI:
关键词:
摘要: applying the association rule into classification can improve accuracy and obtain some valuable rules information that cannot be captured by other approaches. However, generation procedure is very time-consuming when encountering large data set. Besides, traditional classifier building organized in several separate phases which may also degrade efficiency of these In this paper, a new class based associative approach (CACA) proposed. The label taken good advantage mining step so as to cut down searching space. proposed algorithm synchronize phases, shrinking space help speed up generation. Experimental result suggested CACA making better performances Associative