Spatio-temporal Co-occurrence Pattern Mining in Data Sets with Evolving Regions

Abstract

Spatio-temporal co-occurring patterns represent subsets of event types that occur together in both space and time. In comparison to previous work in this field, we present a general framework to identify spatio-temporal cooccurring patterns for continuously evolving spatio-temporal events that have polygon-like representations. We also propose a set of measures to identify spatio-temporal co-occurring patterns and propose an Apriori-based spatio-temporal cooccurrence mining algorithm to find prevalent spatio-temporal co-occurring patterns for extended spatial representations that evolve over time. We evaluate our framework on real-life data to demonstrate the effectiveness of our measures and the algorithm. We present results highlighting the importance of our measures in identifying spatio-temporal co-occurrence patterns.

On visualization techniques for solar data mining

“…While not all of our work is directly visual, such as high-dimensional indexing techniques to facilitate similarity search and spatiotemporal frequent pattern mining (Pillai et al, 2012) towards possible predictive abilities, almost all of it is related to some sort of visualizable end result. In this section, we briefly highlight some interesting aspects of our research that are aided directly through visualizations.…”

Big Data New Frontiers: Mining, Search and Management of Massive Repositories of Solar Image Data and Solar Events

“…While not all of our work is directly visual, such as high-dimensional indexing techniques and spatio-temporal frequent pattern mining [7,11,12], almost all of it is related to some sort of visualizable end result. For example, with the help of visualization we can quickly analyze hundreds of solar events at once and validate a module's reporting effectiveness against known solar science, such as the confirmed distinct bands of active regions and coronal holes shown in Fig.2c.…”

Finding Spatiotemporal Co-occurrence Patterns of Heterogeneous Events for Prediction

“…Many algorithms for this approach can be found in [13], [18], [21] or [3]. For example, in [13], the authors proposed a prevalence measure called the participation index.…”

“…In stage 1, we build a set of spatiotemporal transactions from the set of instances of features of the co-occurrence combination, I = {i 1 , i 2 , ...i N }. Spatiotemporal transactions are created by using spatial and temporal overlap as in [18] and [14]. We are more interested in the stage 2 in which we want to mine co-occurrence patterns from a set of spatiotemporal transactions.…”

“…For example, in [13], the authors proposed a prevalence measure called the participation index. In [18], the authors used spatiotemporal overlap relation for mining spatiotemporal co-occurrence pattern in data sets with evolving regions. The transaction-based approach focus on defining transactions over space and time and then an association rule mining like [1] can be used.…”