Spatio-Temporal Data Clustering using Deep Learning: A Review

Spatial and temporal information are recorded with each measurement in so-called spatiotemporal (ST) data. The space-time information added to each measurement makes the data complex enough to dissent with classical statistical data mining methods. Modeling the evolution of a process over time and space is a fundamental challenge in various disciplines. Spatiotemporal clustering defines the process of grouping data points that are similar in both space and time. While classical clustering algorithms solve the problem by considering time as an additional dimension, deep learning models with powerful data learning capability open up new possibilities. This paper examines recent significant works on clustering spatiotemporal data using deep learning in the context of advancements in deep clustering.