Issue No. 10 - October (2003 vol. 25)
<p><b>Abstract</b>—We propose a novel visualization algorithm for high-dimensional time-series data. In contrast to most visualization techniques, we do <it>not</it> assume consecutive data points to be independent. The basic model is a linear dynamical system which can be seen as a dynamic extension of a probabilistic principal component model. A further extension to a particular <it>switching</it> linear dynamical system allows a representation of complex data onto multiple and even a hierarchy of plots. Using sensible approximations based on expectation propagation, the projections can be performed in essentially the same order of complexity as their static counterpart. We apply our method on a real-world data set with sensor readings from a paper machine.</p>
Data visualization, time-series, latent variables, principal component analysis, switching linear dynamical systems, approximate inference.
O. Zoeter and T. Heskes, "Hierarchical Visualization of Time-Series Data Using Switching Linear Dynamical Systems," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 25, no. , pp. 1202-1214, 2003.