Open AccessRule Extraction Algorithm of Ordered Decision Information System
Author(s) -
Jing-Lan Mo
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1651/1/012053
Subject(s) - data mining , computer science , attribution , algorithm , property (philosophy) , function (biology) , rough set , value (mathematics) , dominance (genetics) , decision rule , set (abstract data type) , information extraction , mathematics , artificial intelligence , machine learning , biology , genetics , gene , psychology , social psychology , philosophy , epistemology , programming language
In ordered decision information system, an improved LEM2 algorithm (DRI-LEM2) is proposed based on the generalized decision function, aiming at the low efficiency and low quality of LEM2 algorithm in extracting rules. The algorithm computes the candidate attribution-value pair set according to the generalized decision function, and removes redundant attribution-value pair from the candidate attribution-value pair set to gradually reduce the size of the property-value pair set. Through UCI experiments, it is proved that the improved LEM2 algorithm (DRI-LEM2) based on dominance relations can effectively improve the efficiency and quality of rules extraction.