Title :
Integration of unsupervised clustering, interaction and parallel coordinates for the exploration of large multivariate data
Author :
Johansson, Jimmy ; Treloar, Robert ; Jern, Mikael
Author_Institution :
Linkoping Univ., Sweden
Abstract :
Parallel coordinates are widely used in many applications for visualization of multivariate data. Because of the nature of parallel coordinates, the visualization technique is often used for data overview. However, when the number of tuples to be visualized becomes very large, this technique makes it difficult to distinguish the overall structure. In This work we present a novel technique which uses a classification approach, the self-organizing map (an unsupervised learning algorithm), to solve this problem by creating an initial clustering of the data. By initially only visualizing the resulting representational clusters, the inherited global structure can be shown. Using linked views and allowing the user to perform drill-down and filtering on these representations reveals the single data items without loss of context.
Keywords :
data visualisation; pattern clustering; self-organising feature maps; unsupervised learning; data clustering; data overview; interaction coordinates; interactive visualization; large multivariate data exploration; linked views; multivariate data visualization; parallel coordinates; representational cluster visualization; self-organizing map; tuple visualization; unsupervised clustering; unsupervised learning algorithm; Aggregates; Clustering algorithms; Data analysis; Data visualization; Displays; Filtering; Multidimensional systems; Navigation; Unsupervised learning;
Conference_Titel :
Information Visualisation, 2004. IV 2004. Proceedings. Eighth International Conference on
Print_ISBN :
0-7695-2177-0
DOI :
10.1109/IV.2004.1320124