New Feature Subset Selection Algorithm Using Class Association Rules Mining

Abstract:

A new feature subset selection algorithm using class association rules mining is proposed in this paper. Firstly, the algorithm mines rules with features as antecedences and class attributes as consequences. Then, it selects the strongest rules one by one, and all the rules’ antecedences make up of the selected feature subset. Experimental results on 10 real data sets show that the proposed algorithm produces a remarkable advantage in enormously reducing the number of the features while keeping quite high classification accuracy compared with other existing seven algorithms. This algorithm can offer an available preprocess technique for pattern recognition, machine learning and data mining.

• Periodical: Key Engineering Materials (Volumes 474-476)
• Pages: 622-625
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.474-476.622
• Online since: April 2011
• Authors: Shou Juan Zhang, Quan Zhou
• Keywords: Apriori Algorithm, C4.5 Decision Tree Classification, Class Association Rules Mining, Feature Selection

References

[1] JiaweiHan, MichelineKamber, DataMining: Concepts and Techniques, 2nd ed., Elsevier Inc. 2006, pp.230-231.

[2] Quinlan, J.R. C4. 5: Programs for Machine Learning. Morgan Kaufmann Publishers, (1993).

[3] R. Agrawal,R. Srikant, Fast algorithms for mining association rules, Proceedings of the 20th Very Large DataBases Conference(VLDB'94), Santiago de Chile, Chile, 1994, pp.487-499.

[4] Jianwen Xie, Ijanhua Wu, Qingquan Qian, Feature Selection Algorithm Based on Association Rules Mining Method, IEEE2009 Eight IEEE/ACIS International Conference on Computer and Information Science, pp.357-362.

DOI: 10.1109/icis.2009.103

[5] Asuncion, A. & Newman, D.J. (2007). UCI Machine Learning Repository [http: /www. ics. uci. edu/~mle arn/MLRepository. html]. Irvine, CA: University of California, School of Information and Computer Science.

[6] Liu Wei, Zhang Feng-li, Cheng Hong-rong, Wan Ming-cheng, Improved ReliefF algorithm applied in detecting image spam, Application Research of Computers2009, Vol. 26, No. 9, pp.3256-32588.

[7] DaiLei; YunXiaochun; XiaoJun, Optimizing traffic classification using hybrid feature selection, IEEE2009 Ninth International Conference on Web-Age Information Management (WAIM), pp.520-525.

DOI: 10.1109/waim.2008.30

[8] Ohara, Kouzou; Hara, Masahiro; Takabayashi, Kiyoto; Motoda, Hiroshi; Washio, Takashi, Pruning strategies based on the upper bound of information gain for discriminative subgraph mining, Lecture Notes in Computer Science2009 (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 5465, LNAI, pp.50-60.

DOI: 10.1007/978-3-642-01715-5_5

[9] Jingnian Chen; Shujun Fu; Taorong Qiu; Gain-ratio-based selective classifiers for incomplete data, 2009 IEEE International Conference on Granular Computing (GrC 2009), pp.57-60.

DOI: 10.1109/grc.2009.5255162

[10] Senthamarai Kannan, Ramaraj, A novel hybrid feature selection via symmetrical uncertainty ranking based local memetic search algorithm, Knowledge-Based Systems2010, Vol. 23, No. 6, pp.580-585.

DOI: 10.1016/j.knosys.2010.03.016

[11] Kabir, M.M.; Shahjahan, M.; Murase, K.; Involving New Local Search in Hybrid Genetic Algorithm for Feature Selection, ICONIP 2009, Neural Information Processing. Proceedings 16th International Conference, pp.150-158, Springer Verlag, Berlin.

DOI: 10.1007/978-3-642-10684-2_17