Dimensionality Reduction by Feature Co-Occurrence based Rough Set

doi:10.23940/ijpe.19.01.p31.307316

Int J Performability Eng ›› 2019, Vol. 15 ›› Issue (1): 307-316.doi: 10.23940/ijpe.19.01.p31.307316

Previous Articles Next Articles

Dimensionality Reduction by Feature Co-Occurrence based Rough Set

Lei La^a*(), Qimin Cao^b, and Ning Xu^b

a School of Information Technology & Management, University of International Business and Economics, Beijing, 100029,China
b Library,China University of Political Science and Law, Beijing, 100088, China

Revised on ; Accepted on
Contact: La Lei E-mail:lalei1984@aliyun.com
About author:Lei La is an associate professor of School of Information Technology & Management, University of International Business and Economics. His research interests include semi-supervised machine learning and network data analysis.|Qimin Cao is an associate research fellow of Library, China University of Political Science and Law. Her research interests include system engineering and information analysis.|Ning Xu is an undergraduate student of School of Information Technology & Management, University of International Business and Economics. His research interests include machine learning algorithm development and coding.

Abstract

Abstract:

Feature selection is the key issue of unstructured data mining related fields. This paper presents a dimensionality reduction method which uses a rough set as the feature selection tool. Different from previous rough set based classification algorithm, it takes feature co-occurrence into account when make attribution reduction to get a more accurate feature subset. The novel method called Feature Co-occurrence Quick Reduction algorithm is in this article. Experimental results show it has a high efficiency in dimensionality reduction—time consumption by approximately 23% less than traditional rough set based dimensionality reduction methods. Moreover, classification based on the feature set selected by Feature Co-occurrence Quick Reduction algorithm is more precise. The proposed algorithm is helpful to us for refining knowledge from massive unstructured data.

Key words: dimensionality reduction, feature co-occurrence, rough set, time consumption, unstructured data

Lei La, Qimin Cao, and Ning Xu. Dimensionality Reduction by Feature Co-Occurrence based Rough Set [J]. Int J Performability Eng, 2019, 15(1): 307-316.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

Figures/Tables 10

Figure 1

Table 1

Figure 2

Figure 3

Figure 4

Table 2

Figure 5

Table 3

Figure 6

Figure 7

References 18

[1]	A. Yindalon, D. F. Lawrence, Z. Li, R.P. Eric, E. Efstathiadis , “A Comprehensive Empirical Comparison of Modern Supervised Classification and Feature Selection Methods for Text Categorization,” Journal of the Association for Information Science and Technology, Vol.42, No.10, pp. 1964-1987, October 2014
[2]	P. T. Fernando, J. Cardiff, P. Rosso, P. David , “Weblog and Short Text Feature Extraction and Impact on Categorization,” Journal of Intelligent & Fuzzy Systems, Vol.27, No. 52014, pp. 2529-2544, May 2016
[3]	R. S. Bhalerao, J. Y. Ollitrault, P. Subrata , “Principal Component Analysis of Event-by-Event Fluctuations,”Physical Review Letters, Vol.114, pp. 1-6, December 2015
[4]	M. Z. Ming , “Sparse Principal Component Analysis and Iterative Thresholding,” Annals of Statistics, Vol.41, No. 2, pp. 772-801, February 2013
[5]	J. Guia, Z. Suna, W. Jia , “Discriminant Sparse Neighborhood Preserving Embedding for Face Recognition,”Pattern Recognition, Vol.45, No. 8, pp. 2884-2893, August 2012
[6]	R. Karbauskaitė, O. Kurasova, G. Dzemyda , “Selection of the Number of Neighbors of Each Data Point for the Locally Linear Embedding Algorithm,” Information Technology and Control, Vol.36, No.4, pp. 359-364, April 2015
[7]	F. Y. Cao, J.Y. Liang, D.Y. Li , “A Dissimilarity Measure for the k-Modes Clustering Algorithm,” KNOWLEDGE-BASED SYSTEMS, Vol.26, No. 2, pp. 120-127, May 2014
[8]	K. Devarajan, G. L. Wang, N. Ebrahimi , “A Unified Statistical Approach to Non-Negative Matrix Factorization and Probabilistic Latent Semantic Indexing,”MACHINE LEARNING,Vol. 99, pp. 137-163, 2015
[9]	D. Q. Miao, H. Y. Duan, N. J. Zhang , “Rough Set based Hybrid Algorithm for Text Classification,” Expert Systems with Applications, Vol.36, No. 5, pp. 9168-9174, 2009
[10]	Y.S. Lee, L. Rocky, C. Y. Chen, P. C. Lin, J. C. Wang , “News Topics Categorization using Latent Dirichlet Allocation and Sparse Representation Classifier,” in Proceedings of IEEE International Conference on Consumer Electronics, pp. 126-13, Taipei, Taiwan, June 2015
[11]	Polkowski, Lech, Shusaku Tsumoto, and Tsau Y. Lin,Rough Set Methods and Applications: New Developments in Knowledge Discovery in Information Systems,Vol. 56, Physica, 2012
[12]	J. Dai and H. Tian, “Fuzzy Rough Set Model for Set-Valued Data,”Fuzzy Set and Systems,Vol. 229, No. 7, pp. 54-68, July 2013
[13]	S. B. Kanti, S. S. Sankar, C. Kripasindhu , “A Genetic Algorithm-based Rule Extraction System,” Applied Soft Computing,Vol.12, No. 1, pp. 238-254 January 2013
[14]	C. C. Yeh, D. J. Chi, T. Y. Lin, S. H. Chiu , “A Hybrid Detecting Fraudulent Financial Statements Model using Rough Set Theory and Support Vector Machines,” Cybernetics and Systems, Vol.47, No. 4, pp. 261-276, April 2016
[15]	H. M. Chen, T. R. Li, C. Luo, S. J. Horng, G. Y. Wang , “A Decision-Theoretic Rough Set Approach for Dynamic Data Mining,” IEEE Transactions on Fuzzy Systems, Vol.23, No. 6, pp. 1958-1970, June 2015
[16]	X. Y . Chen and S. Y. Wu, “The Optimization Assignment Model of Multi-Sensor Resource Management based on Rough Entropy,” International Journal of Grid and Utility Computing, Vol.8, No. 5, pp. 233-244, August 2015
[17]	Y. H. Han, Y. Yang, Z. G. Ma, Y. Yan, N. Sebe, X.F. Zhou , “Semi-Supervised Feature Selection via Spline Regression for Video Semantic Recognition,” IEEE Transactions on Neural Networks and Learning Systems, Vol.26, No. 2, pp. 252-264, February 2015
[18]	W. Z. Wu, Y. H. Xu, M. W. Shao, G. Y. Wang , “Axiomatic Characterizations of (S, T)-Fuzzy Rough Approximation Operators,” Information Sciences, Vol. 334, pp. 17-43, September 2016

D	T				C
D	t₁	t₂	…	t_m	C
d₁	1	1	…	0	c₁
d₂	0	1	…	0	c₂
…	…	…	…	…	…
d_n_-1	0	1		0	c_l_-1
d_n	1	0	…	1	c_l

Number of documents	Original feature number	Reductionfeature number	Number ratio
100	2153	34	63.32
300	3362	47	63.83
1000	4101	67	61.21
3000	4887	74	66.04
10000	5239	83	63.12

AlgorithmClass	SVM	Na?ve Bayes	Neural Networks
Politics	0.895	0.813	0.864
Economics	0.897	0.799	0.853
Education	0.912	0.830	0.862
Sports	0.909	0.824	0.841
Culture & Art	0.911	0.822	0.867
Science	0.898	0.815	0.855

Dimensionality Reduction by Feature Co-Occurrence based Rough Set

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 18

Related Articles 4

Recommended 0

[1]	Rushali A. Deshmukh. Naive Bayes and Neural Network Techniques for Marathi Poem Classification into Nine Rasa using Feature Selection [J]. Int J Performability Eng, 2022, 18(9): 626-636.
[2]	Risu Na, Weiguo Zhang, Kaifa Jia, and Quan Zhang. Key Factors of Seal Ring Reliability based on QFD [J]. Int J Performability Eng, 2022, 18(11): 759-769.
[3]	Wanjuan Zhang, Xiaodan Wang, Diego Cabrera, and Yun Bai. Product Quality Reliability Analysis based on Rough Bayesian Network [J]. Int J Performability Eng, 2020, 16(1): 37-47.
[4]	Zhifeng Zhang and Junxia Ma. Feature Selection Combined Feature Resolution with Attribute Reduction based on Correlation Matrix of Equivalence Classes [J]. Int J Performability Eng, 2019, 15(4): 1131-1140.