基于三支决策的密度聚类算法
Density Based Spatial Clustering of Application with Noise Based on Three-Way Decision
摘要: 三支聚类使用核心域，边界域和琐碎域三个集合来表示类簇，将确定的元素放入核心域中，不确定的元素放入边界域中延迟决策，降低了决策风险。本文将含有噪声的基于密度的聚类算法(Density Based Spatial Clustering of Application with Noise, DBSCAN)与三支聚类进行结合，利用数学形态学中的腐蚀和膨胀思想，用自然最近邻算法定义了一个结构算子，对二支聚类的结果通过收缩和膨胀得到核心域和边界域。在UCI数据集和Shape数据集上的实验结果显示，该方法可以有效降低DBI的值，同时提高ACC和AS的值。
Abstract: Three-way clustering uses three sets of core region, boundary region and trivial region to represent the clusters. The determined elements are put into the core region, while the uncertain elements are put into the boundary region to delay the decision, thus reducing the decision risk. In this paper, DBSCAN (Density Based Spatial Clustering of Application with Noise) is combined with the three-way clustering, and a structural operator is defined by using the natural nearest neighbor algorithm based on the corrosion and expansion ideas in mathematical morphology. The core region and boundary region are obtained by shrinking and expanding the results of the two-way clustering. Experimental results on UCI datasets and shape datasets show that this method can effectively reduce the value of DBI and improve the value of ACC and AS.
文章引用：姜凡. 基于三支决策的密度聚类算法[J]. 应用数学进展, 2022, 11(2): 858-865. https://doi.org/10.12677/AAM.2022.112092

参考文献

[1] Elalami, M.E. (2011) Supporting Image Retrieval Framework with Rule Base System. Knowledge-Based Systems, 24, 331-340. [Google Scholar] [CrossRef] 
[2] Chang, M.S., Chen, L.H., Hung, L.J., et al. (2014) Exact Algorithms for Problems Related to the Densest k-Set Problem. Information Processing Letters, 114, 510-513. [Google Scholar] [CrossRef] 
[3] Xu, D. and Tian, Y. (2015) A Comprehensive Survey of Clustering Algorithms. Annals of Data Science, 2, 165-193. [Google Scholar] [CrossRef] 
[4] Ester, M., Kriegel, H.P., Sander, J., et al. (1996) A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining, Portland, 2-4 August 1996, 226-231.
[5] Rangaprakash, D., Odemuyiwa, T., Narayana, D.N., et al. (2020) Density-Based Clustering of Static and Dynamic Functional MRI Connectivity Features Obtained from Subjects with Cognitive Impairment. Brain Informatics, 7, Article No. 19. [Google Scholar] [CrossRef] [PubMed]
[6] 岳晓新, 贾君霞, 陈喜东, 李广安. 改进YOLO V3的道路小目标检测[J]. 计算机工程与应用, 2020, 56(21): 218-223.
[7] Lingras, P. and West, C. (2004) Interval Set Clustering of Web Users with Rough k-Means. Journal of Intelligent Information Systems, 23, 5-16. [Google Scholar] [CrossRef] 
[8] Yao, Y.Y., Lingras, P., Wang, R.Z., et al. (2009) Interval Set Cluster Analysis: A Re-Formulation. International Workshop on Rough Sets, Fuzzy Sets, Data Mining, and Granular Soft Computing, Delhi, 15-18 December 2009, 398-405. [Google Scholar] [CrossRef] 
[9] Yu, H., Wang, X., Wang, G., et al. (2018) An Active Three-Way Clustering Method via Low-Rank Matrices for Multi-View Data. Information Sciences, 507, 823-839. [Google Scholar] [CrossRef] 
[10] Yu, H., Zhang, C. and Wang, G.Y. (2016) A Tree-Based Incremental Overlapping Clustering Method Using the Three-Way Decision Theory. Knowledge-Based Systems, 91, 189-203. [Google Scholar] [CrossRef] 
[11] Yu, H. (2017) A Framework of Three-Way Cluster Analysis. International Joint Conference on Rough Sets, Olsztyn, 3-7 July 2017, 300-312. [Google Scholar] [CrossRef] 
[12] Wang, P.X. and Yao, Y.Y. (2018) CE3: A Three-Way Clustering Method Based on Mathematical Morphology. Knowledge-Based Systems, 155, 54-65. [Google Scholar] [CrossRef] 
[13] Wang, P.X., Shi, H., Yang, X.B. and Mi, J. (2019) Three-Way k-Means: Integrating k-Means and Three-Way Decision. International Journal of Machine Learning & Cybernetics, 10, 2767-2777. [Google Scholar] [CrossRef] 
[14] Yang, B. and Li, J.H. (2020) Complex Network Analysis of Three-Way Decision Researches. International Journal of Machine Learning and Cybernetics, 11, 973-987. [Google Scholar] [CrossRef] 
[15] Yu, H., Chen, L.Y., Yao, J.T., et al. (2019) A Three-Way Clustering Method Based on an Improved DBSCAN Algorithm. Physica A: Statistical Mechanics and its Applications, 535, Article ID: 122289. [Google Scholar] [CrossRef] 
[16] Yao, Y.Y. (2011) The Superiority of Three-Way Decisions in Probabilistic Rough Set Models. Information Sciences, 181, 1080-1096. [Google Scholar] [CrossRef] 
[17] Yao, Y.Y. (2012) An Outline of a Theory of Three-Way Decisions. International Conference on Rough Sets and Current Trends in Computing, Chengdu, 17-20 August, 1-17. [Google Scholar] [CrossRef] 
[18] 刘强, 施虹, 王平心, 杨习贝. 基于ε邻域的三支决策聚类分析[J]. 计算机工程与应用, 2019, 55(6): 140-144.
[19] Yu, H., Jiao, P., Yao, Y.Y., et al. (2016) Detecting and Refining Overlapping Regions in Complex Networks with Three-Way Decisions. Information Sciences, 373, 21-41. [Google Scholar] [CrossRef] 
[20] Zhang, Q., Lyu, G., Chen, Y., et al. (2018) A Dynamic Three-Way Decision Model Based on the Updating of Attribute Values. Knowledge-Based Systems, 142, 71-84. [Google Scholar] [CrossRef] 
[21] 周红芳, 王鹏. DBSCAN算法中参数自适应确定方法的研究[J]. 西安理工大学学报, 2012, 28(3): 291-293.
[22] Huang, J., Zhu, Q., Yang, L., et al. (2016) A Non-Parameter Outlier Detection Algorithm Based on Natural Neighbor. Knowledge-Based Systems, 92, 71-77. [Google Scholar] [CrossRef] 
[23] Serra, J. (1986) Introduction to Mathematical Morphology. Computer Vision Graphics & Image Processing, 35, 283-305. [Google Scholar] [CrossRef] 
[24] Banerji, A. (2000) An Introduction to Image Analysis Using Mathematical Morphology. IEEE Engineering in Medicine and Biology Magazine, 19, 13-14. [Google Scholar] [CrossRef] 
[25] Bloch, I. (2000) On Links between Mathematical Morphology and Rough Sets. Pattern Recognition, 33, 1487-1496.
[26] Pasi, F. and Sami, S. (2018) K-Means Properties on Six Clustering Benchmark Datasets. Applied Intelligence, 48, 4743-4759. [Google Scholar] [CrossRef] 

为你推荐