基于SOM聚类的垃圾处理方式可视化分析方法
|
摘要
针对全国各省份垃圾处理方式的数据,提出一种混合可视分析方法。为了从多角度分析数据,混合U矩阵、平行坐标以及Small-Multiple 3种可视化技术,设计并实现了3种可视化视图的交互联动。首先,对数据进行聚类处理,将各省份近年的垃圾处理方式划分类别,采用SOM神经网络聚类算法实现聚类。然后,针对SOM聚类结果,采用U矩阵的方式进行可视化,并采用平行坐标描述每个聚类结果的各个属性。为了分析数据的地理属性及时序属性,采用Small-Multiple可视化技术。最后,实现多视图联动、刷新技术等交互方式,帮助用户自行探索数据,实现多视图的交互展示与分析。实验表明,这种混合可视方式可达到较好的多属性交互可视化效果,能够帮助用户了解并分析我国垃圾处理方式的分布及趋势。
A hybrid visual analysis method was proposed for the data of garbage disposal modes in various provinces of China.In order to analyze the data from multiple views,this paper combined three visualization techniques including unified distance matrix,parallel coordinate and Small-Multiple,designed and realized the interactions among the three views.Firstly,the data are processed by clustering so that the garbage disposal modes of different provinces in recent years can be divided into different categories.The clustering used in this paper is based on self-organizing maps(SOM)neural network clustering algorithm.Secondly,according to the results of SOM clustering,the unified distance matrix is used for visualization,and the parallel coordinate is used to describe the attributes of each clustering result.In order to analyze the geographical and sequential attributes of data,the Small-Multiple visualization technology is used.Thirdly,interactions such as multi-view linkage and refreshing are implemented to help users explore the data on their own and achieve interactive display and analysis of multiple views.The experimental results show that the hybrid visualization method can achieve a good visualization effect of multi-attributes interaction and can help users to understand the distribution and trend analysis of garbage disposal modes in China.
A hybrid visual analysis method was proposed for the data of garbage disposal modes in various provinces of China.In order to analyze the data from multiple views,this paper combined three visualization techniques including unified distance matrix,parallel coordinate and Small-Multiple,designed and realized the interactions among the three views.Firstly,the data are processed by clustering so that the garbage disposal modes of different provinces in recent years can be divided into different categories.The clustering used in this paper is based on self-organizing maps(SOM)neural network clustering algorithm.Secondly,according to the results of SOM clustering,the unified distance matrix is used for visualization,and the parallel coordinate is used to describe the attributes of each clustering result.In order to analyze the geographical and sequential attributes of data,the Small-Multiple visualization technology is used.Thirdly,interactions such as multi-view linkage and refreshing are implemented to help users explore the data on their own and achieve interactive display and analysis of multiple views.The experimental results show that the hybrid visualization method can achieve a good visualization effect of multi-attributes interaction and can help users to understand the distribution and trend analysis of garbage disposal modes in China.
引文
[1] TUFTE E R.Beautiful Evidence[M].Cheshire:Graphics Press,2006.
[2] REIJNER H.The development of the horizon graph[C]∥Proceedings of IEEE Visualization Workshop From Theory to Practice:Design,Vison and Visualization.2008.
[3] TUFTE E R.Envisioning information[J].Optometry&Vision Science,1991,68(4):322-324.
[4] WEBER M,ALEXA M,MLLER W.Visualizing Time-Series on Spirals[C]∥Proceedings of the IEEE Symposium on Information Visualization 2001(InfoVis01).2001:7-14.
[5] VAN WIJK J J,VAN SELOW E R.Cluster and calendar based visualization of time series data[C]∥Proceedings of the IEEE Symposium on Information Visualization 1999(InfoVis’99).1999:4-9.
[6] GOODWIN S,DYKES J,SLINGSBY A,et al.Visualizing multiple variables across scale and geography[J].IEEE Transactions on Visualization and Computer Graphics,2016,22(1):599-608.
[7] HEILMANN R,KEIM D A,PANSE C,et al.Recmap:Rectangular map approximations[C]∥Proceedings of the IEEE Symposium on Information Visualization 2004(InfoVis’04).2004:33-40.
[8] KEIM D A,PANSE C,NORTH S C.Medial-axis-based cartograms[J].IEEE Computer Graphics and Applications,2005,25(3):60-68.
[9] COLLINS C,PENN G,CARPENDALE S.Bubble sets:Revealing set relations with isocontours over existing visualizations[J].IEEE Transactions on Visualization and Computer Graphics,2009,15(6):1009-1016.
[10]QU H,CHAN W Y,XU A,et al.Visual analysis of the air pollution problem in Hong Kong[J].IEEE Transactions on Visualization and Computer Graphics,2007,13(6):1408-1415.
[11]IM J F,MCGUFFIN M J,LEUNG R.GPLOM:the generalized plot matrix for visualizing multidimensional multivariate data[J].IEEE Transactions on Visualization and Computer Graphics,2013,19(12):2606-2614.
[12]GUO D,CHEN J,MACEACHREN A M,et al.A visualization system for space-time and multivariate patterns(vis-stamp)[J].IEEE Transactions on Visualization and Computer Graphics,2006,12(6):1461-1474.
[13]PENG W,WARD M O,RUNDENSTEINER E A.Clutter Reduction in Multi-Dimensional Data Visualization Using Dimension Reordering[C]∥Proceedings of the IEEE Symposium on Information Visualization 2004(InfoVis’04).2004:89-96.
[14]WARD M,GRINSTEIN G,KEIM D.Interactive Data Visualization:Foundations,Techniques,and Applications[J].Seed&Plant Production Journal,2010,51(4):188-195.
[15]WARD M O.Xmdvtool:Integrating multiple methods for visualizing multivariate data[C]∥Proceedings of the Conference on Visualization’94.IEEE Computer Society Press,1994:326-333.
[16]国家统计局.中国统计年鉴[M].北京:中国统计出版社,2016.
[2] REIJNER H.The development of the horizon graph[C]∥Proceedings of IEEE Visualization Workshop From Theory to Practice:Design,Vison and Visualization.2008.
[3] TUFTE E R.Envisioning information[J].Optometry&Vision Science,1991,68(4):322-324.
[4] WEBER M,ALEXA M,MLLER W.Visualizing Time-Series on Spirals[C]∥Proceedings of the IEEE Symposium on Information Visualization 2001(InfoVis01).2001:7-14.
[5] VAN WIJK J J,VAN SELOW E R.Cluster and calendar based visualization of time series data[C]∥Proceedings of the IEEE Symposium on Information Visualization 1999(InfoVis’99).1999:4-9.
[6] GOODWIN S,DYKES J,SLINGSBY A,et al.Visualizing multiple variables across scale and geography[J].IEEE Transactions on Visualization and Computer Graphics,2016,22(1):599-608.
[7] HEILMANN R,KEIM D A,PANSE C,et al.Recmap:Rectangular map approximations[C]∥Proceedings of the IEEE Symposium on Information Visualization 2004(InfoVis’04).2004:33-40.
[8] KEIM D A,PANSE C,NORTH S C.Medial-axis-based cartograms[J].IEEE Computer Graphics and Applications,2005,25(3):60-68.
[9] COLLINS C,PENN G,CARPENDALE S.Bubble sets:Revealing set relations with isocontours over existing visualizations[J].IEEE Transactions on Visualization and Computer Graphics,2009,15(6):1009-1016.
[10]QU H,CHAN W Y,XU A,et al.Visual analysis of the air pollution problem in Hong Kong[J].IEEE Transactions on Visualization and Computer Graphics,2007,13(6):1408-1415.
[11]IM J F,MCGUFFIN M J,LEUNG R.GPLOM:the generalized plot matrix for visualizing multidimensional multivariate data[J].IEEE Transactions on Visualization and Computer Graphics,2013,19(12):2606-2614.
[12]GUO D,CHEN J,MACEACHREN A M,et al.A visualization system for space-time and multivariate patterns(vis-stamp)[J].IEEE Transactions on Visualization and Computer Graphics,2006,12(6):1461-1474.
[13]PENG W,WARD M O,RUNDENSTEINER E A.Clutter Reduction in Multi-Dimensional Data Visualization Using Dimension Reordering[C]∥Proceedings of the IEEE Symposium on Information Visualization 2004(InfoVis’04).2004:89-96.
[14]WARD M,GRINSTEIN G,KEIM D.Interactive Data Visualization:Foundations,Techniques,and Applications[J].Seed&Plant Production Journal,2010,51(4):188-195.
[15]WARD M O.Xmdvtool:Integrating multiple methods for visualizing multivariate data[C]∥Proceedings of the Conference on Visualization’94.IEEE Computer Society Press,1994:326-333.
[16]国家统计局.中国统计年鉴[M].北京:中国统计出版社,2016.