Refinement and Trust Modeling of Spatio-Temporal Big Data

Abstract

The conventional studies of spatio-temporal data models and their big data applications cannot reliably reflect the large volume, heterogeneity and dynamics of spatio-temporal big data. In this paper, the structure and function expression of spatio-temporal metadata is analyzed. With fused and normalized spatio-temporal reference and data structure, the constraint rules of spatio-temporal big data refinement are proposed. Using the domain specific modeling (DSM) and the data granulation algorithms, an object-oriented modeling language, the thrust modeling of spatio-temporal big data, and the aggregated status correlation of unified model data are established. This work utilizes the trust modeling theory and the spatio-temporal data processing methods and defines a case study that converts spatio-temporal data into dynamic complex big data. This research paves the way for the trust modeling and validation of spatio-temporal big data.