An Analytical Survey on Classification for Method Incomplete Pattern
Keywords:
Prototype Based classification, Belief function, credal classification, evidential reasoning, incomplete pattern, missing data, k -means clustering.Abstract
The classification of incomplete patterns is an astoundingly troublesome task in light of the way that the dissent (incomplete case) with different possible estimations of missing qualities may yield specific classification happens. The shakiness (ambiguity) of classification is generally realized by the nonappearance of data of the missing data. Another model based credal classification (PCC) system is proposed to oversee incomplete patterns in light of the conviction work structure used generally as a piece of evidential thinking approach. The class models obtained by means of getting ready tests are separately used to check the missing qualities. Consistently, in a c-class issue, one needs to oversee c models, which yield c estimations of the missing qualities. The various changed patterns, in light of all conceivable possible estimation have been gathered by a standard classifier and we can get at most c unmistakable classification comes to fruition for an incomplete case. Since all these unmistakable classification comes about are possibly satisfactory, we propose to combine every one of them to get the last classification of the incomplete case. Another credal mix procedure is introduced for taking consideration of the classification issue, and it can depict the unavoidable insecurity in view of the possible conflicting outcomes passed on by different estimations of the missing qualities. The incomplete patterns that are uncommonly difficult to assemble in a specific class will be sensibly and normally committed to some genuine meta-classes by PCC procedure with a particular ultimate objective to diminish mistakes. The sufficiency of PCC system has been attempted through four examinations with fake and honest to goodness data sets. In this paper, we talk about various incomplete illustration classification and evidential thinking systems used as a piece of the area of data mining.
References
- Zhun-Ga Liu, Quan Pan, Grgoire Mercier, and Jean Dezert, “A New Incomplete Pattern Classication Method Based on Evidential Reasoning”, North western Polytechnical Uni-versity, Xian 710072, China,4, APRIL 2015
- J. Luengo, J. A. Saez, and F. Herrera, “Missing data imputation for fuzzy rule-based classification systems,” Soft Comput., vol. 16, no. 5, pp. 863-881, 2012.
- T. Denoeux, “Maximum likelihood estimation from uncertain data in the belief function framework,” IEEE Trans. Knowl. Data Eng., vol. 25, no. 1, pp. 119-130, Jan. 2013.
- P. Garcia-Laencina, J. Sancho-Gomez, and A. Figueiras-Vidal, “Pattern classification with missing data: A review,” Neural Comput. Appl. vol. 19, no. 2, pp. 263–282, 2010.
- P. Smets, “Analyzing the combination of conflicting belief functions,” Inform. Fusion, vol. 8, no. 4, pp. 387-412, 2007.
- K. Pelckmans, J. D. Brabanter, J. A. K. Suykens, and B. D. Moor, “Handling missing values in support vector machine classifiers,” Neural Netw., vol. 18, nos. 5-6, pp. 684-692, 2005.
- O. Troyanskaya et al., “Missing value estimation methods for DNA microarrays,”Bioinformatics, vol. 17, no. 6, pp. 520-525, 2001.
- M.-H. Masson and T. Denoeux, “ECM: An evidential version of the fuzzy C-means algorithm,” Pattern Recognit., vol. 41, no. 4, pp. 1384-1397, 2008.
- G. Batista and M. C. Monard, “A study of K-nearest neighbour as an imputation method,” in Proc. 2nd Int. Conf. Hybrid Intell. Syst., 2002, pp. 251-260.
- Z. Ghahramani and M. I. Jordan, “Supervised learning from incomplete data via an EM approach,” in Advances in Neural Information Processing Systems, vol. 6, J. D. Cowan et al., Eds. San Mateo, CA, USA: Morgan Kaufmann, 1994, pp. 120-127.
- D. J. Mundfrom and A. Whitcomb, “Imputing missing values: The effect on the accuracy of classification,” MLRV, vol. 25, no. 1, pp. 13–19, 1998.
- D. Li, J. Deogun, W. Spaulding, and B. Shuart, “Towards missing data imputation: A study of fuzzy k-means clustering method,” in Proc. 4th Int. Conf. Rough Sets Current Trends Comput. (RSCTC04), Uppsala, Sweden, Jun. 2004, pp. 573–579.
- P. Smets, “The combination of evidence in the transferable belief model,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 12, no. 5, pp. 447 458, May 1990.
- T. Denoeux and P. Smets, “Classification using belief functions: Relationship between case-based and model-based approaches,” IEEE Trans. Syst., Man, Cybern. B, Cybern., vol. 36, no. 6, pp. 1395–1406, Dec. 2006.
- T. Denoeux, “A neural network classifier based on Dempster–Shafer theory,” IEEE Trans. Syst., Man, Cybern. A, Syst. Humans, vol. 30, no. 2, pp. 131–150, Mar. 2000.
Downloads
Published
Issue
Section
License
Copyright (c) IJSRCSEIT
This work is licensed under a Creative Commons Attribution 4.0 International License.