Attribute reduction based on optimized granular ball neighborhood rough sets

Yufei Wang; Nan Zhang; Yufei Wang; Nan Zhang

doi:10.3934/era.2026319

Electronic Research Archive

2026, Volume 34, Issue 10: 7392-7419. doi: 10.3934/era.2026319

Previous Article Next Article

Research article

Attribute reduction based on optimized granular ball neighborhood rough sets

Yufei Wang ^,,
Nan Zhang

School of Computer and Control Engineering, Yantai University, Yantai 264005, China

Received: 17 June 2026 Revised: 04 August 2026 Accepted: 12 August 2026 Published: 27 August 2026

Neighborhood rough sets (NRS) are commonly applied to attribute reduction because they can process numerical data directly without discretization. Granular ball neighborhood rough sets extend NRS by adaptively determining neighborhood radii according to the data distribution. However, existing granular ball generation and reduction algorithms can be further enhanced in terms of granule representation, radius computation, and attribute evaluation. To address these issues, an optimized granular ball neighborhood rough set model (OGNRS) and a corresponding attribute reduction algorithm are proposed. In this model, a scoring function integrating density and centroid-based distance is employed to select centers, and a median absolute deviation-based strategy is introduced to determine the radii of granular balls. In addition, the minimal-redundancy-maximal-relevance criterion is incorporated into the reduction process to provide search guidance and reduce redundancy among selected attributes. Experiments on 14 UCI datasets demonstrate that OGNRS achieves competitive overall performance in terms of classification accuracy, reduct size, and computational efficiency.
- rough set,
- granular computing,
- granular ball,
- attribute reduction
Citation: Yufei Wang, Nan Zhang. Attribute reduction based on optimized granular ball neighborhood rough sets[J]. Electronic Research Archive, 2026, 34(10): 7392-7419. doi: 10.3934/era.2026319

Related Papers:

Abstract

Neighborhood rough sets (NRS) are commonly applied to attribute reduction because they can process numerical data directly without discretization. Granular ball neighborhood rough sets extend NRS by adaptively determining neighborhood radii according to the data distribution. However, existing granular ball generation and reduction algorithms can be further enhanced in terms of granule representation, radius computation, and attribute evaluation. To address these issues, an optimized granular ball neighborhood rough set model (OGNRS) and a corresponding attribute reduction algorithm are proposed. In this model, a scoring function integrating density and centroid-based distance is employed to select centers, and a median absolute deviation-based strategy is introduced to determine the radii of granular balls. In addition, the minimal-redundancy-maximal-relevance criterion is incorporated into the reduction process to provide search guidance and reduce redundancy among selected attributes. Experiments on 14 UCI datasets demonstrate that OGNRS achieves competitive overall performance in terms of classification accuracy, reduct size, and computational efficiency.

References

[1]	Z. Pawlak, Rough sets, Int. J. Comput. Inf. Sci., 11 (1982), 341–356. https://doi.org/10.1007/BF01001956
[2]	M. Modrzejewski, Feature selection using rough sets theory, in Machine Learning: ECML-93, 667 (1993), 213–226. https://doi.org/10.1007/3-540-56602-3_138
[3]	R. W. Swiniarski, A. Skowron, Rough set methods in feature selection and recognition, Pattern Recognit. Lett., 24 (2003), 833–849. https://doi.org/10.1016/S0167-8655(02)00196-4 doi: 10.1016/S0167-8655(02)00196-4
[4]	N. Parthaláin, Q. Shen, R. Jensen, A distance measure approach to exploring the rough set boundary region for attribute reduction, IEEE Trans. Knowl. Data Eng., 22 (2010), 305–317. https://doi.org/10.1109/TKDE.2009.119 doi: 10.1109/TKDE.2009.119
[5]	V. K. H. Turaga, S. Chebrolu, Rapid and optimized parallel attribute reduction based on neighborhood rough sets and MapReduce, Expert Syst. Appl., 260 (2025), 125323. https://doi.org/10.1016/j.eswa.2024.125323 doi: 10.1016/j.eswa.2024.125323
[6]	S. Wu, L. Wang, S. Ge, Z. Hao, Y. Liu, Neighborhood rough set with neighborhood equivalence relation for feature selection, Knowl. Inf. Syst., 66 (2024), 1833–1859. https://doi.org/10.1007/s10115-023-01999-z doi: 10.1007/s10115-023-01999-z
[7]	J. Dai, Q. Hu, H. Hu, D. Huang, Neighbor inconsistent pair selection for attribute reduction by rough set approach, IEEE Trans. Fuzzy Syst., 26 (2018), 937–950. https://doi.org/10.1109/TFUZZ.2017.2698420 doi: 10.1109/TFUZZ.2017.2698420
[8]	W. Ziarko, Variable precision rough set model, J. Comput. Syst. Sci., 46 (1993), 39–59. https://doi.org/10.1016/0022-0000(93)90048-2 doi: 10.1016/0022-0000(93)90048-2
[9]	W. Ziarko, Probabilistic approach to rough sets, Int. J. Approximate Reasoning, 49 (2008), 272–284. https://doi.org/10.1016/j.ijar.2007.06.014 doi: 10.1016/j.ijar.2007.06.014
[10]	D. Slezak, W. Ziarko, The investigation of the Bayesian rough set model, Int. J. Approximate Reasoning, 40 (2005), 81–91. https://doi.org/10.1016/j.ijar.2004.11.004 doi: 10.1016/j.ijar.2004.11.004
[11]	Q. Hu, D. Yu, J. Liu, C. Wu, Neighborhood rough set based heterogeneous feature subset selection, Inf. Sci., 178 (2008), 3577–3594. https://doi.org/10.1016/j.ins.2008.05.024 doi: 10.1016/j.ins.2008.05.024
[12]	L. Shen, F. E. H. Tay, A discretization method for rough sets theory, Intell. Data Anal., 5 (2001), 431–438. https://doi.org/10.3233/IDA-2001-5506 doi: 10.3233/IDA-2001-5506
[13]	R. Ali, M. H. Siddiqi, S. Lee, Rough set-based approaches for discretization: A compact review, Artif. Intell. Rev., 44 (2015), 235–263. https://doi.org/10.1007/s10462-014-9426-2 doi: 10.1007/s10462-014-9426-2
[14]	Q. Hu, H. Zhao, D. Yu, Efficient symbolic and numerical attribute reduction with neighborhood rough sets, Pattern Recognit. Artif. Intell., 21 (2008), 732–738.
[15]	W. Xu, Z. Yuan, Z. Liu, Feature selection for unbalanced distribution hybrid data based on $k$-nearest neighborhood rough set, IEEE Trans. Artif. Intell., 5 (2024), 229–243. https://doi.org/10.1109/TAI.2023.3237203 doi: 10.1109/TAI.2023.3237203
[16]	J. Dai, W. Chen, L. Xia, Feature selection based on neighborhood complementary entropy for heterogeneous data, Inf. Sci., 682 (2024), 121261. https://doi.org/10.1016/j.ins.2024.121261 doi: 10.1016/j.ins.2024.121261
[17]	C. Liu, B. Lin, D. Miao, A novel adaptive neighborhood rough sets based on sparrow search algorithm and feature selection, Inf. Sci., 679 (2024), 121099. https://doi.org/10.1016/j.ins.2024.121099 doi: 10.1016/j.ins.2024.121099
[18]	S. Xia, Y. Liu, X. Ding, G. Wang, H. Yu, Y. Luo, Granular ball computing classifiers for efficient, scalable and robust learning, Inf. Sci., 483 (2019), 136–152. https://doi.org/10.1016/j.ins.2019.01.010 doi: 10.1016/j.ins.2019.01.010
[19]	S. Xia, H. Zhang, W. Li, G. Wang, E. Giem, Z. Chen, GBNRS: A novel rough set algorithm for fast adaptive attribute reduction in classification, IEEE Trans. Knowl. Data Eng., 34 (2022), 1231–1242. https://doi.org/10.1109/TKDE.2020.2997039 doi: 10.1109/TKDE.2020.2997039
[20]	S. Zeng, D. Liu, IEWGBNRS: A novel attribute reduction approach based on information entropy-weighted granular-ball neighborhood rough set, J. Northwest Univ. (Nat. Sci. Ed.), 56 (2026), 128–140. https://doi.org/10.16152/j.cnki.xdxbzr.2026-01-012 doi: 10.16152/j.cnki.xdxbzr.2026-01-012
[21]	S. Xia, C. Wang, G. Wang, X. Gao, W. Ding, J. Yu, et al., GBRS: A unified granular ball learning model of Pawlak rough set and neighborhood rough set, IEEE Trans. Neural Networks Learn. Syst., 36 (2025), 1719–1733. https://doi.org/10.1109/TNNLS.2023.3325199 doi: 10.1109/TNNLS.2023.3325199
[22]	Z. Zhang, J. Song, H. Li, E. C. C. Tsang, A weighted granular-ball rough set: model and attribute reduction, in 2024 International Conference on Machine Learning and Cybernetics (ICMLC), (2024), 433–438. https://doi.org/10.1109/ICMLC63072.2024.10935042
[23]	Z. Jia, Z. Zhang, W. Pedrycz, Generation of granular balls for clustering based on the principle of justifiable granularity, IEEE Trans. Cybern., 55 (2025), 1687–1700. https://doi.org/10.1109/TCYB.2025.3534195 doi: 10.1109/TCYB.2025.3534195
[24]	J. Zhang, K. Sun, B. Huang, T. Wang, X. Wang, Attribute reduction based on a rapid variable granular ball generation model, Expert Syst. Appl., 265 (2025), 126030. https://doi.org/10.1016/j.eswa.2024.126030 doi: 10.1016/j.eswa.2024.126030
[25]	J. Pan, G. Lang, Q. Xiao, T. Yang, A framework of granular ball generation for classification via granularity tuning, Appl. Intell., 55 (2025), 63. https://doi.org/10.1007/s10489-024-05904-1 doi: 10.1007/s10489-024-05904-1
[26]	S. Xia, X. Dai, G. Wang, X. Gao, E. Giem, An efficient and adaptive granular ball generation method in classification problem, IEEE Trans. Neural Networks Learn. Syst., 35 (2024), 5319–5331. https://doi.org/10.1109/TNNLS.2022.3203381 doi: 10.1109/TNNLS.2022.3203381
[27]	Q. Xie, Q. Zhang, S. Xia, F. Zhao, C. Wu, G. Wang, et al., GBG++: A fast and stable granular ball generation method for classification, IEEE Trans. Emerging Top. Comput. Intell., 8 (2024), 2022–2036. https://doi.org/10.1109/TETCI.2024.3359091 doi: 10.1109/TETCI.2024.3359091
[28]	H. Peng, F. Long, C. Ding, Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy, IEEE Trans. Pattern Anal. Mach. Intell., 27 (2005), 1226–1238. https://doi.org/10.1109/TPAMI.2005.159 doi: 10.1109/TPAMI.2005.159
[29]	A. Rodriguez, A. Laio, Clustering by fast search and find of density peaks, Science, 344 (2014), 1492–1496. https://doi.org/10.1126/science.1242072 doi: 10.1126/science.1242072
[30]	F. Liu, Q. Zhang, S. Xia, Q. Xie, W. Liao, S. Zhang, A granular ball generation method based on local density for classification, Inf. Sci., 717 (2025), 122295. https://doi.org/10.1016/j.ins.2025.122295 doi: 10.1016/j.ins.2025.122295

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)