Visual Clustering Method of Quasi-Circular Mapping Based on Dimension Extension and Rearrangement
HUANG Shan1, LI Ming1,2, CHEN Hao1,2, LI Junhua1,2, ZHANG Congxuan2
1.School of Information Engineering, Nanchang Hangkong University, Nanchang 330063 2.Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition, Nanchang Hangkong University, Nanchang 330063
Abstract：The non-linear structure of high-dimensional data cannot be captured by the existing radial layout visualization method. Therefore, visual clustering method of quasi-circular mapping based on dimension extension and rearrangement is proposed. The dimension of high-dimensional data is expanded by affinity propagation clustering algorithm and multi-objective clustering visualization evaluation index. Then, the dimension correlation rearrangement of the extended high-dimensional data is carried out. Finally, the high-dimensional data is reduced to two-dimensional visualization space by quasi-circular mapping mechanism to realize effective visual clustering. Experiments show that the proposed dimension extension and rearrangement strategy can effectively improve the visual clustering effect of quasi-circular mapping visualization. The dimension extension strategy can also significantly improve the clustering effect of other radial layout visualization methods with better generalization performance. Moreover, the proposed method has obvious advantages in visual clustering accuracy, topology
[1] TANG J, LIU J Z, ZHANG M, et al. Visualizing Large-Scale and High-Dimensional Data // Proc of the 25th International Conference on World Wide Web. New York, USA: ACM, 2016: 287-297. [2] LIU S S, MALJOVEE D, WANG B, et al. Visualizing High-Dimensional Data: Advances in the Past Decade. IEEE Transactions on Visualization and Computer Graphics, 2017, 23(3): 1249-1268. [3] DANIELS K, GRINSTEIN G, RUSSELL A, et al. Properties of Normalized Radial Visualizations. Information Visualization, 2012, 11(4): 273-300. [4] VAN LONG T. Another Look at Radial Visualization for Class-Preserving Multivariate Data Visualization. Informatica-Journal of Computing and Informatics, 2017, 41(2): 159-168. [5] RUBIO-SÁNCHEZ M, RAYA L, DIAZ F, et al. A Comparative Study between Radviz and Star Coordinates. IEEE Transactions on Visualization and Computer Graphics, 2015, 22(1): 619-628. [6] NOVÁKOVÁ L, STEPÁNKOVÁ O. Visualization of Trends Using RadViz. Journal of Intelligent Information Systems, 2011, 37(3): 355-369. [7] LEHMANN D J, THEISEL H. Orthographic Star Coordinates. IEEE Transactions on Visualization and Computer Graphics, 2013, 19(12): 2615-2624. [8] 张志豪,张军平,陈德铭,等.基于可交互相关性矩阵的维度重排径向坐标可视化方法.模式识别与人工智能, 2017, 30(7): 637-645. (ZHANG Z H, ZHANG J P, CHEN D M, et al. Interactive Dimension Reordering in RadViz with Correlation Matrix. Pattern Recognition and Artificial Intelligence, 2017, 30(7): 637-645.) [9] VAN LONG T, LINSEN L. Visualizing High Density Clusters in Multidimensional Data Using Optimized Star Coordinates. Computational Statistics, 2011, 26(4): 655-678. [10] VAN LONG T. iRadviz: An Inversion RadViz for Class Visualization of Multivariate Data Visualization // Proc of the 7th Sympo-sium on Information and Communication Technology. New York, USA: ACM, 2016: 131-138. [11] ALBUQUERQUE G, EISEMANN M, LEHMANN D J, et al. Improving the Visual Analysis of High-Dimensional Datasets Using Quality Measures // Proc of the IEEE Symposium on Visual Analytics Science and Technology. Washington, USA: IEEE, 2010: 19-26. [12] ZHOU F F, HUANG W, LI J C, et al. Extending Dimensions in RadViz Based on Mean Shift // Proc of the IEEE Pacific Visualization Symposium. Washington, USA: IEEE, 2015, I: 111-115. [13] ZANABRIA G G, NONATO L G, GOMEZ-NIETO E. iStar (i*): An Interactive Star Coordinates Approach for High-Dimensional Data Exploration. Computers and Graphics, 2016, 60: 107-118. [14] SHARKO J, GRINSTEIN G, MARX K A. Vectorized Radviz and Its Application to Multiple Cluster Datasets. IEEE Transactions on Visualization and Computer Graphics, 2008, 14(6): 1444-1451. [15] 周芳芳,李俊材,黄 伟,等.基于维度扩展的Radviz可视化聚类分析方法.软件学报, 2016, 27(5): 1127-1139. (ZHOU F F, LI J C, HUANG W, et al. Extending Dimensions in Radviz for Visual Clustering Analysis. Journal of Software, 2016, 27(5): 1127-1139.) [16] 朱胜利,朱善安.基于卡尔曼滤波器组的Mean Shift模板更新算法.中国图象图形学报, 2007, 12(3): 460-465. (ZHU S L, ZHU S A. An Algorithm of Mean Shift Template Update Based a Group of Kalman Filters. Journal of Image and Graphics, 2007, 12(3): 460-465.) [17] WANG L M, ZHENG K Y, TAO X, et al. Affinity Propagation Clustering Algorithm Based on Large-Scale Dataset. International Journal of Computers and Applications, 2018, 40(3): 1-6. [18] KONIG A. Interactive Visualization and Analysis of Hierarchical Neural Projections for Data Mining. IEEE Transactions on Neural Networks, 2000, 11(3): 615-624. [19] KOU G, PENG Y, WANG G X. Evaluation of Clustering Algorithms for Financial Risk Analysis Using MCDM Methods. Information Sciences, 2014, 275: 1-12. [20] LAURENS V D M. Accelerating t-SNE Using Tree-Based Algorithms. Journal of Machine Learning Research, 2014, 15: 3221-3245. [21] AHMED A H, ASHOUR W. An Initialization Method for the K-means Algorithm Using RNN and Coupling Degree. International Jour- nal of Computer Applications, 2011, 25(1): 1-6.
2019, Vol. 32 
