Spatiotemporal Modeling

Spatiotemporal dataset has three components in general, attributes, space, and time. Modeling approaches of spatiotemporal data have covered a broad spectrum of applications in many fields, including environmental applications, crime hotspot analysis, healthcare informatics, transportation modeling, social media, and many others. Clustering, predictive learning, frequent pattern mining, anomaly detection, change detection, and relationship mining are the few broad categories of the modeling approaches irrespective of the applications (Atluri et al. 2018). This chapter discusses some modeling approaches used for environmental applications in general. Further, one spatiotemporal modeling approach of outlier detection is chosen and presented here. Outlier detection within the application data is an essential preprocessing step for most of the spatiotemporal applications. Some important literature on spatiotemporal outlier detection methods is also discussed. Though the applications and methods presented here are not exhaustive, this chapter gives the initial pointers for further exploration of the spatiotemporal models.