A New Way to Describe Spatiotemporal Data
Mingli Song · Proceedings of the 2nd International Conference on Computer Science and Electronics Engineering (ICCSEE 2013) · 2013
Data analysis especially dealing with spatiotemporal data in a human-centric way becomes essential in modern engineering and science fields.In this study, we present a granular way of data analysis where the data are described by information granules defined in spatial and temporal domain.We show that the data can be effectively described through a collection of Cartesian products of information granules and their convolution.The quality of the codebooks (vocabulary) is evaluated by data granulation-degranulation. Finally a synthetic example is given to illustrate our algorithm.