Review of Clustering Methods for Functional Data

Functional data clustering is to identify heterogeneous morphological patterns in the continuous functions underlying discrete measurements/observations. Application of functional has appeared many publications across various fields sciences, including but not limited biology, (bio)chemistry, engineering, environmental science, medical psychology, social and so on. The phenomenal growth application indicates urgent need for a systematic approach develop efficient methods scalable algorithmic implementations. On other hand, there abundant literature on cluster analysis time series, trajectory data, spatio-temporal on, which are all related data. Therefore, an overarching structure existing will enable cross-pollination ideas research fields. We here conduct comprehensive review original propose taxonomy that explores connections differences among relates them conventional multivariate methods. built three main attributes method therefore more reliable than categorizations. aims bridge gap between community generate new principles clustering.