Stimulation Spectrum Based High-dimensional Data Visualization
Kan Liu, Ping Liu, Dawei Jin · 2006
One of the main goals of high-dimensional data visualization is to reduce dimensions, so that the data can be projected onto 2-D or 3-D space. This paper proposes a stimulation spectrum based visualization approach, in which each high-dimension data is regarded as a stimulation spectrum and the change of each value of data attribute corresponds to the change of wavelength of visible spectrum. According to the relationship between stimulation spectrum and color space, we can project high-dimensional data onto 3- dimensional space, and easily observe the distribution of the data. We also demonstrate this novel approach with synthetic data and real data, and the result is very promising.