Enhancing the Visualization Process with Principal Component Analysis to Support the Exploration of Trends

Mueller, W., Nocke, T. and Schumann, H.

This paper describes the integration of the Principal Component Analysis into the Visualization Process. Although, the combination of Principal Component Analysis (PCA) and visual methods is a common approach to the analysis of high-dimensional datasets, it is mostly limited to a pure preprocessing step for dimension reduction. In this paper we will discuss, how PCA results can be used to control all steps of the visualization pipeline to generate more effective visual representations, and thus, a higher degree of understanding of the PCA values as well as of original data.

Cite as: Mueller, W., Nocke, T. and Schumann, H. (2006). Enhancing the Visualization Process with Principal Component Analysis to Support the Exploration of Trends. In Proc. Asia Pacific Symposium on Information Visualisation (APVIS2006), Tokyo, Japan. CRPIT, 60. Misue, K., Sugiyama, K. and Tanaka, J., Eds. ACS. 121-130.

(from crpit.com) (local if available)