T-Map: A Topological Approach to Time Series Volume Data Mining

摘要

This article presented T-map as a topologically-based visual data mining environment for large-scale time-series volume datasets. The T-map relies on a mathematically rigorous theory, and provides the user with a set of robust algorithms to describe the topological skeleton of 4D scalar field. The present scheme can be extended to handling a higher dimensional array of datasets for the purpose of parameter study. For example, the IS and EIS can serve as a 2D-scrollable strip chart for visual data mining spreadsheet called Hyper-Cells [DeCoste 2001]. The feasibility of the T-map scheme has been proved empirically with an application to large-scale time-varying volume datasets from atomic collision research. See [Otsuka et at. 2002] for more details.