One approach among a number of strategies for multiple classifier combination is to calculate the beliefs of confidence based on confusion matrices from individual classifiers. To achieve precise belief computation, based on previous researchers' work, presented is a better algorithm with a more generic capability, showing improved performance.
References
-
-
1)
-
J.R. Parker
.
Rank and response combination from confusion matrix data.
Inf. Fusion
,
113 -
120
-
2)
-
H.L. Tang ,
R. Hanka ,
H.S.H. Ip
.
Histological image retrieval based on semantic content analysis.
IEEE Trans. Inf. Technol. Biomed.
,
1 ,
26 -
36
-
3)
-
D. Tax
.
Combining multiple classifiers by averaging or by multiplying?.
Pattern Recognit.
,
1475 -
1485
-
4)
-
L. Xu ,
A. Krzyzak ,
C.Y. Suen
.
Methods of combining classifiers and their applications to handwriting recognition.
IEEE Trans. Syst. Man Cybern.
,
3 ,
418 -
435
-
5)
-
Tang, H.L.: `Semantic analysis of image content for intelligent retrieval and automatic annotation of medical images', 2000, PhD, University of Cambridge.
-
6)
-
J. Kittler
.
On combining classifiers.
IEEE Trans. Pattern Anal. Mach. Intell.
,
3 ,
226 -
239
http://iet.metastore.ingenta.com/content/journals/10.1049/el_20040176
Related content
content/journals/10.1049/el_20040176
pub_keyword,iet_inspecKeyword,pub_concept
6
6