Sage Journals: Discover world-class research

Abstract

Currently, Multiple Classifier System (MCS) attracts more and more attentions and has become one of the research hotspots in the pattern recognition field. Classifier selection is a commonly used strategy for MCS to achieve the final decision. A classifier selection method based on clustering and weighted mean is proposed in this paper. In the method, multiple clusters are selected according to the distances between cluster centers and the input sample. Then, the average performance of each classifier on selected clusters is calculated. The best classifier on the nearest cluster and the classifier with the best average performance are picked out. According to the reliability of their outputs which are estimated by confusion matrix, one of them is selected to make the final decision of the system. A number of benchmark data sets from KDD’99, UCI and ELENA database were used to evaluate the proposed method. It can be seen form the experimental results that the proposed method performs well.

Keywords

Multiple classifier system classifier selection clustering weighted mean

Get full access to this article

View all access options for this article.

References

Wozniak

, Graña

and Corchado

, A survey of multiple classifier systems as hybrid systems, Information Fusion16(1) (2014), 3–17.

Chowdhury

, Verma

and Stockwell

, A novel texture feature based multiple classifier technique for roadside vegetation classification, Expert Systems with Applications42(12) (2015), 5047–5055.

Chacko

A.M.M.

and Dhanya

P.M.

, Multiple classifier system for offline malayalam character recognition, Procedia Computer Science46 (2015), 86–92.

De-la-Torre

, Granger

, Radtke

P.V.W.

, Sabourin

and Gorodnichy

D.O.

, Partially-supervised learning from facial trajectories for face recognition in video surveillance, Information Fusion24 (2015), 31–53.

Dainotti

, Pescapé

, Sansone

and Quintavalle

, Using a behaviour knowledge space approach for detecting unknown IP traffic flow, Lecture Notes in Computer Science6713 (2011), 360–369.

Zahid

M.A.

and Swart

, The borda majority count, Information Sciences295 (2015), 429–440.

Bigdeli

, Samadzadegan

and Reinartz

, Fusion of hyperspectral and LIDAR data using decision template-based fuzzy multiple classifier system, International Journal of Applied Earth Observation and Geoinformation38 (2015), 309–320.

Yang

, Han

D.Q.

and Han

C.Z.

, A novel diversity measure of multiple classifier systems based on distance of evidence, Acta Aeronautica et Astronautica Sinica33(6) (2012), 1093–1099.

Abdallah

A.C.B.

, Hichem

and Gader

, Adaptive local fusion with neural networks, Lecture Notes in Computer Science6353 (2010), 486–491.

10.

Zhang

W.B.

, Ji

H.B.

and Wang

, Adaptive weighted feature fusion classification algorithm, Systems Engineering and Electronics35(6) (2013), 1133–1137.

11.

Kuncheva

L.I.

, Switching between selection and fusion in combining classifiers: An experiment, IEEE Transactions on Systems, Man and Cybernetics32(2) (2002), 146–156.

12.

Hao

H.W.

, Wang

Z.B.

, Yin

X.C.

and Chen

Z.Q.

, Dynamic selection and circulating combination for multiple classifier systems, Acta Automatica Sinica37(11) (2011), 1290–1295.

13.

Cavalin

P.R.

, Sabourin

and Suen

C.Y.

, Dynamic selection approaches for multiple classifier systems, Neural Computing and Applications22(3-4) (2013), 673–688.

14.

Britto

A.S.

, Sabourin

and Oliveira

L.E.

, Dynamic selection of classifiers-A comprehensive review, Pattern Recognition47(11) (2014), 3665–3680.

15.

A.Z.

and Huo

Z.Q.

, Experimental comparison of six fixed classifier fusion rules, Procedia Engineering23 (2011), 429–433.

Classifier selection method based on clustering and weighted mean

Abstract

Keywords

Get full access to this article

References