Time Series Clustering Methods for Analysis of Astronomical Data

摘要

Cluster analysis, often referred to as segmentation in business contexts, is used to identify and describe subgroups of individuals with common characteristics that distinguish them from the rest of the population. While segments are often identified using static characteristics, evolving systems may be better described by how things change over time. A medical patient may be classified by the amount of time since an important event such as a diagnosis, economic activity may be segmented by stages in an economic cycle, and neighborhoods grouped by stages in generational evolution. In astrostatistics, this technique is used to classify a supernova by how the amount of light it produces changes over time.