Title :
An Improved DM Algorithm Based on Rough Set Theory
Author :
Yang Zu-Qiao ; Xiao Xiao-Hong ; Gao Han-ping
Author_Institution :
Sch. of Comput. Sci. & Technol., HuangGang Normal Univ., Huanggang
Abstract :
The basic idea of an improved algorithm is outlined in this paper firstly, includes main steps and the advantages and disadvantages of CART algorithm; and then the definitions of attribute importance and the system comprehensive information entropy and the algorithm of maintaining maximum system comprehensive information entropy are presented. Compares to some disadvantages of CART algorithm such as low computation efficiency of extraction rules and long rule length, the improved algorithm has effect in the experimental result.
Keywords :
data mining; entropy; rough set theory; CART algorithm; attribute importance; data mining; improved DM algorithm; maximum system comprehensive information entropy; rough set theory; Computer science; Data mining; Decision trees; Delta modulation; Erbium; Information entropy; Information systems; Neural networks; Rail to rail inputs; Set theory;
Conference_Titel :
Wireless Communications, Networking and Mobile Computing, 2007. WiCom 2007. International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-1311-9
DOI :
10.1109/WICOM.2007.769
