DocumentCode :
480594
Title :
New Computation for Rules Extraction Based on Block Matrix
Author :
Cheng Yusheng ; Jiang Xiaoyao ; Zhang Xiaoliang
Author_Institution :
Sch. of Comput. & Inf., Anqing Teachers´ Coll., Anqing
Volume :
1
fYear :
2008
fDate :
20-22 Dec. 2008
Firstpage :
75
Lastpage :
79
Abstract :
As a new mathematical tool to deal with uncertainty, rough sets theory has been successfully applied in many field such as pattern recognition, data reduction and so on. Because the limitations of discernible matrix and equivalence matrix for rules extraction, a new definition of block equivalence matrix and its strategy of divide and conquer for rules extraction in information system have been put forward based on rules extraction of equivalence matrix; According to equal division principle, the process of transforming rules extraction in information system to parallel rules extraction in agents corresponding to the multiple subsystems after division effectively solves the problem of rules extraction in information system. Complexity analysis shows that the algorithm is more efficient than those existing algorithms. An example is used to illustrate the efficiency of the new algorithm based on matrix block.
Keywords :
computational complexity; divide and conquer methods; information systems; knowledge acquisition; matrix algebra; rough set theory; uncertain systems; block matrix; divide and conquer method; equal division principle; equivalence matrix; information system rule extraction; mathematical tool; parallel rule extraction; rough sets theory; uncertain system; Acceleration; Application software; Data mining; Information systems; Information technology; Intelligent vehicles; Knowledge acquisition; Pattern recognition; Rough sets; Uncertainty; discernible matrix; matrix block computation; rough sets theory; rules extraction;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Information Technology Application, 2008. IITA '08. Second International Symposium on
Conference_Location :
Shanghai
Print_ISBN :
978-0-7695-3497-8
Type :
conf
DOI :
10.1109/IITA.2008.107
Filename :
4739538
Link To Document :
بازگشت