Spatial-temporal data mining, or knowledge discovery in spatial/temporal databases, is the extraction of implicit knowledge, and discovery of interesting characteristics and patterns that are not explicitly represented in the spatial-temporal data. These techniques can play an important role in understanding the data and in capturing intrinsic relationships among data. The amount of spatial data obtained from satellite, medical imagery and other sources has been growing tremendously in recent years. Terabytes of data are generated everyday. As a consequence, interesting patterns may change over time. It is preferable to make the system monitor certain patterns specified by users and take proper actions upon occurrence. STING+ extends current spatial data mining techniques to support user-defined triggers, i.e., active spatial data mining.

In general, for any large database with time varying numerical attributes, interesting patterns are often numerous and complicated. This is both a challenging problem and one with significant practical application in business, science, and medicine. Many patterns can be represented in the form of association rules. We proposed a parameterizable model for temporal sequences of numerical attributes and devised efficient ways to search for parameter values that will result in a good fit to (at least a significant portion of) the data. Metrics for how well instances of the model fit portion of the data includethe familiar measures of support and strength used in association rule mining and a new metric called density. A user specifies thresholds for these metrics and, based on structural properties of the class of models we are attempting to fit to the data, the search space can be drastically pruned by using these thresholds.