Izvestiya Physics of the Solid Earth · 2016 · 41 citations · 23 references
Natural SciencesEarth SciencesGeophysicsEarthquake EngineeringGeospatial Artificial IntelligenceSeismologyEarthquake HazardsEarthquake LocationEngineering SeismologySeismic Hazard IdentificationSeismic MonitoringStrong-motion SeismologyBeltsEarthquake Detection AlgorithmsEarthquake Hazard MappingEarthquake Location EstimationEarthquake Signal ProcessingEarthquakesMicroseismic Event LocationSeismic Hazard Mapping
The fuzzy clustering and zoning method (FCAZm) of systems analysis is suggested for recognizing the areas of the probable generation of the epicenters of significant, strong, and the strongest earthquakes. FCAZm is a modified version of the previous FCAZ algorithmic system, which is advanced by the creation of the blocks of artificial intelligence that develop the system-forming algorithms. FCAZm has been applied for recognizing areas where the epicenters of the strongest ( M ≥ 7 3 /4) earthquakes within the Andes mountain belt in the South America and significant earthquakes ( M ≥ 5) in the Caucasus can emerge. The reliability of the obtained results was assessed by the seismic-history type control experiments. The recognized highly seismic zones were compared with the ones previously recognized by the EPA method and by the initial version of the FCAZ system. The modified FCAZm system enabled us to pass from simple pattern recognition in the problem of recognizing the locations of the probable emergence of strong earthquakes to systems analysis. In particular, using FCAZm we managed to uniquely recognize a subsystem of highly seismically active zones from the nonempty complement using the exact boundary.
23
Algorithms for Clustering Data
Warren S. Sarle, Anil K. Jain, Richard C. Dubes · Technometrics · 1990 · 7.8K citations
Algorithms for clustering data
Bernhard Flury · Journal of Statistical Planning and Inference · 1989 · 6.5K citations