Research on Multi-granularity Attribute Reduction Method for Continuous Parameters

doi:10.3778/j.issn.1673-9418.2006061

Abstract

Abstract:

To measure the degree of information granulation, granularity has attracted many researchers?? extensive attention in the field of granular computing. One of the important and widely accepted patterns is parameterized granularity. Based on such parameterized granularity, when solving the problem of attribute reduction, it is often necessary to calculate the reducts related to each parameter until all of the reducts have been obtained. Obviously, this method will result in high time consumption. To fill such a gap, a multi-granularity attribute reduction approach based on continuous parameters is proposed. Firstly, a new constraint related to attribute reduction is constructed by using the interval of continuous parameters and the monotonicity of uncertainty measure in rough set. Secondly, a forward greedy searching algorithm is designed to derive the continuous parameters based reducts. Finally, 8 UCI data sets are selected for experimental comparisons and analyses. The results show that compared with single granularity based reducts in terms of multiple parameters, attribute reduction related to continuous parameters can greatly reduce the elapsed time of obtaining reduct without causing significant changes in the classification performance. This study provides a new solution for multi-granularity based modeling and feature selection from a continuous perspective.

Key words: attribute reduction, continuous parameter, multi-granularity, neighborhood rough set

摘要：

作为度量粒化程度的方式，在粒计算研究领域中，粒度受到了众多学者的广泛关注，其中一种重要且广为接受的模式是参数化粒度。利用这种参数化的粒度表现形式，在面向属性约简的求解问题时，往往需要计算每一个参数所对应的粒度下约简，直至找出所有参数下的约简结果。显然，这种方式会带来巨大的时间消耗。为解决这一问题，提出了一种连续参数意义下的多粒度属性约简策略：首先利用连续参数的区间及粗糙集中不确定性度量的单调性，构造了连续参数下属性约简的约束条件；其次设计了连续参数意义下约简求解的前向贪心搜索算法；最后选取了8组UCI数据集进行实验对比分析，结果表明，相较于多个离散参数下的单粒度属性约简，连续参数意义下的属性约简可以在使得约简中属性的分类性能不发生显著变化的情况下，极大地提升约简求解的时间性能。这一研究为从连续视角进行多粒度建模及相关属性选择工作提供了新的解决方案。

关键词: 属性约简, 连续参数, 多粒度, 邻域粗糙集

WU Jiang, SONG Jingjing, CHENG Fuhao, WANG Pingxin, YANG Xibei. Research on Multi-granularity Attribute Reduction Method for Continuous Parameters[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(8): 1555-1562.

吴将, 宋晶晶, 程富豪, 王平心, 杨习贝. 面向连续参数的多粒度属性约简方法研究[J]. 计算机科学与探索, 2021, 15(8): 1555-1562.

References

[1] PAWLAK Z. Rough sets: theoretical aspects of reasoning about data[M]. Hingham: Kluwer Academic Publishers, 1992.
[2] DOWLATSHAHI M B, DERHAMI V, NEZAMABADI P H. Ensemble of filter-based rankers to guide an epsilon-greedy swarm optimizer for high-dimensional feature subset selection [J]. Information, 2017, 8: 152.
[3] CHEN H M, LI T R, LUO C, et al. A decision-theoretic rough set approach for dynamic data mining[J]. IEEE Transactions on Fuzzy Systems, 2015, 23: 1958-1970.
[4] MANORANJAN D, HUAN L. Consistency-based search in feature selection[J]. Artificial Intelligence, 2003, 151: 155-176.
[5] HU Q H, YU D R, XIE Z X. Neighborhood classifiers[J]. Expert Systems with Applications, 2008, 34: 866-876.
[6] YAO Y Y, ZHANG X Y. Class-specific attribute reducts in rough set theory[J]. Information Sciences, 2017, 418/419: 601-618.
[7] CHANDRIMA S, SARAH C, JAIDEEP S. Robust feature selection technique using rank aggregation[J]. Applied Arti-ficial Intelligence, 2014, 28: 243-257.
[8] GAO Y, CHEN X J, WANG P X, et al. Attribute reduction over consistent samples[J]. CAAI Transactions on Intelligent Systems, 2019, 14(6): 1170-1178.
高媛, 陈向坚, 王平心, 等. 面向一致性样本的属性约简[J]. 智能系统学报, 2019, 14(6): 1170-1178.
[9] YANG X B, YAO Y Y. Ensemble selector for attribute redu-ction[J]. Applied Soft Computing, 2018, 70: 1-11.
[10] WANG C Z, HU Q H, WANG X Z, et al. Feature selection based on neighborhood discrimination index[J]. IEEE Tran-sactions on Neural Networks and Learning Systems, 2018, 29(7): 2986-2999.
[11] LI Y W, LIN Y J, LIU J H, et al. Feature selection for multi-label learning based on kernelized fuzzy rough set[J]. Neur-ocomputing, 2018, 318: 271-286.
[12] JIANG Z H, YANG X B, YU H L, et al. Accelerator for multi-granularity attribute reduction[J]. Knowledge-Based Systems, 2019, 177: 145-158.
[13] LIU K Y, YANG X B, YU H L, et al. An efficient selector for multi-granularity attribute reduction[J]. Information Sci-ences, 2019, 505: 457-472.
[14] GAO Y, CHEN X J, YANG X B, et al. Neighborhood attri-bute reduction: a multicriterion strategy based on sample selection[J]. Information, 2018, 9(11): 282-302.
[15] WITOLD P, GIANCARLO S, ALBERTO S, et al. Data des-cription: a general framework of information granules[J]. Knowledge-Based Systems, 2015, 80: 98-108.
[16] JIANG Z H, WANG Y B, XU G, et al. Multi-scale based accelerator for attribute reduction[J]. Computer Science, 2019, 46(12): 250-256.
姜泽华, 王怡博, 徐刚, 等. 面向多尺度的属性约简加速器[J]. 计算机科学, 2019, 46(12): 250-256.
[17] XU W, YU J. A novel approach to information fusion in multi-source data sets: a granular computing viewpoint[J]. Information Sciences, 2017, 378: 410-423.
[18] QIAN Y H, CHENG H H, WANG J T, et al. Grouping gran-ular structures in human granulation intelligence[J]. Inform-ation Sciences, 2017, 382/383: 150-169.
[19] LIU K Y, YANG X B, YU H L, et al. Rough set based semi-supervised feature selection via ensemble selector[J]. Knowledge-Based Systems, 2019, 165: 282-296.
[20] LI Z Y, YANG X B, XU S P, et al. Attribute reduction app-roach to neighborhood decision agreement[J]. Journal of Henan Normal University (Natural Science Edition), 2017, 45(5): 68-73.
李智远, 杨习贝, 徐苏平, 等. 邻域决策一致性的属性约简方法研究[J]. 河南师范大学学报(自然科学版), 2017, 45(5): 68-73.
[21] ZHANG X, MEI C L, CHEN D G, et al. Feature selection in mixed data: a method using a novel fuzzy rough set-based information entropy[J]. Pattern Recognition, 2016, 56: 1-15.