A generalized framework for mining spatio-temporal patterns in scientific data

Abstract

In this paper, we present a general framework to discover spatial associations and spatio-temporal episodes for scientific datasets. In contrast to previous work in this area, features are modeled as geometric objects rather than points. We define multiple distance metrics that take into account objects' extent and thus are more robust in capturing the influence of an object on other objects in spatial neighborhood. We have developed algorithms to discover four different types of spatial object interaction (association) patterns. We also extend our approach to accommodate temporal information and propose a simple algorithm to derive spatio-temporal episodes. We show that such episodes can be used to reason about critical events. We evaluate our framework on real datasets to demonstrate its efficacy. The datasets originate from two different areas: Computational Molecular Dynamics and Computational Fluid Flow. We present results highlighting the importance of the identified patterns and episodes by using knowledge from the underlying domains. We also show that the proposed algorithms scale linearly with respect to the dataset size.

Analyzing change in spatial data by utilizing polygon models

“…In summary, the work in [17] and [7] is similar to our work in that they focus on spatial data. Unlike [17] which uses association analysis, our work utilizes cluster analysis.…”

“…In work by Asur et al [1], a technique for mining evolutionary behavior of interaction graphs is proposed. Yang et al [17] proposed a technique to discover the evolution of spatial objects. Association analysis is used to detect changes.…”

“…Unlike [17] which uses association analysis, our work utilizes cluster analysis. Similar to our technique, the techniques in [1] [17], [7], [13], and [6] analyze changes in data but require that the identity of objects must be known or restrict analysis to objects that are characterized by nominal attributes.…”

Spatio-Temporal Modelling of Frequent Human Mobility Pattern to Analyse the Dynamics of Epidemic Disease

“…The change of pattern [6] [7] in data with respect to space and time is considered to be interesting part of spatio-temporal data mining.W.O. Kermack (1927) laid the foundation [8] for modeling the spread of epidemics.…”

Graph Mining for Object Tracking in Videos

“…Note that most existing algorithms which mine dynamic graphs (e.g., dynamic networks) consider graphs with only edges insertions or deletions i.e., the time series of graphs share the same set of nodes over time (see, e.g., [3]), or in which nodes and edges are only added and never deleted (see, e.g., [2]). In [5,17], the problem is to mine spatio-temporal relationships between moving objects (the mined relationships are restricted to some predefined graphs like cliques, star graphs or sequences). In our approach, however, there is no information about the correspondence between the nodes in one graph (video frame) and those in the others.…”