Frequent subgraph mining in oceanographic multi-level directed graphs

ABSTRACT

We present an adaptation and application of frequent subgraph mining (FSM) in a time series of spatial multi-level directed graphs depicting probabilistic transitions of water masses between neighboring sea areas within a given time interval. The directed graphs are created from the results of the numerical model, the Mediterranean Ocean Forecasting System. We assign unique labels (geographical locations) to vertices of the multi-level directed graphs. Then, we add the edge labels as discretized values of the probabilities of transitions between vertices. This modification allows the use of the established algorithm gSpan to search for frequently directed subgraphs in the sequence of such directed graphs. Thus, we obtain both general and specific subgraphs, such as convergences, divergences, and paths of the ocean currents in the numerical model. The resulting substructures, revealed by directed subgraphs, match oceanographic structures (gyres, convergences/divergences, and paths) deduced from field observations, and can also serve as a tool for the validation of the numerical model of circulation in the sea.

KEYWORDS:

• Geographic knowledge discovery
• geovisualization
• spatio-temporal data modelling
• ocean dynamics

Acknowledgments

This research is supported by the Slovenian Research Agency (Contract no. P1-0237), EC project PERSEUS (Contract no. 287600), EC project MyOcean2 under the 7th Framework Programme and EC project MyOcean FO under HORIZON 2020.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Horizon 2020 Framework Programme [H2020-EU.2.1.6-MyOcean FO-193161]; Javna Agencija za Raziskovalno Dejavnost RS [P1-0237]; FP7 Space [FP7-SPACE-MYOCEAN2-102059];FP7 Environment [FP7-ENVIRONMENT-PERSEUS-102043].

Notes on contributors

Boris Petelin

Boris Petelin received his Ph.D. in 2014 in computer science from University of Ljubljana, Slovenia. He is a research associate at National Institute of Biology, Marine Biology Station Piran, Slovenia. His research includes the numerical modeling of coastal waters dynamics, quality control and spatial-temporal mining of oceanographic data. He is the (co)author of more than 100 publications of which 8 are the scientific papers published in international journals.

Igor Kononenko

Igor Kononenko received his Ph.D. in 1990 in computer science from University of Ljubljana, Slovenia. He is the professor at Faculty of Computer and Information Science in Ljubljana, and the head of Laboratory for Cognitive Modeling. His research interests include artificial intelligence, machine learning, neural networks and cognitive modeling. He is the (co)author of about 225 papers and 12 textbooks (two in English).

Vlado Malačič

Vlado Malačič received his PhD in physics in 1993, is experienced in physical oceanography of coastal waters. He erected coastal oceanography in Slovenia and as professor in meteorology (University of Ljubljana) is running courses on oceanography and physical oceanography. He was driving projects related to the setup of the information system, centered on the coastal buoy 'Vida' and established the oceanographic data centre registered at IOC-UNESCO. His research interest is focused on circulation and turbulence of coastal waters.

Matjaž Kukar

Matjaž Kukar recieved his Ph.D. in computer and information science from University of Ljubljana, Slovenia in 2001. He is associate professor at Faculty of Computer and Information Science in Ljubljana. His research interests include artificial intelligence, machine learning, spatial data mining and data management. He is the (co)author of about 100 papers and 1 textbook (in English).