2011 | OriginalPaper | Chapter
A Novel Feature Selection Method for the Conditional Information Entropy Model
Authors : Jing Ruan, Changsheng Zhang
Published in: Emerging Research in Artificial Intelligence and Computational Intelligence
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
In this paper, a novel feature selection method of discernibility object pair set is provided. At first, the feature selection definition of new method is presented. What’s more, it is proved that the above feature selection definition is equal to the feature selection definition based on conditional information entropy. In order to compute discernibility object pair set, a quick algorithm for simplified decision system is introduced, whose time complexity is
O
(|
C
∥
U
|). On this condition, an efficient and novel algorithm based on discernibility object pair set for feature selection in conditional information entropy model is designed, whose time and space complexity are
O
(|
C
∥
U
|) +
o
(|
c
∥
u
|
c
|
2
) and
O
(|
U
|
C
|
2
) +
O
(|
U
|) respectively. At last, an example is employed to illustrate the efficiency of the new algorithm.