AUTOMATED SEISMIC ANALYSIS USING SUPERVISED MACHINE LEARNING

THE SBIR PHASE I PROJECT PROPOSED CONSISTS OF THE TESTING OF THE CLASSIFICATION PERFORMANCE OF SEVERAL RELATED MACHINE LEARNING METHODS TO SEISMIC DISCRIMINATION. SEVERAL OTHER APPLICATIONS OF THESE METHODS ARE ALSO SUGGESTED. IN THE APPLICATION TO DISCRIMINATION, THE FOLLOWING DATA SETS WILL BE USED: REGIONAL ARRAY DATA FOR EARTHQUAKES AND QUARRY BLASTS IN THE SCANDINAVIAN REGIONS, IRIS DATA RECORDED IN THE USSR AND SELECTED DATA SETS USED PREVIOUSLY TO DEDUCE DISCRIMINANTS. THE PERFORMANCE OF THESE TECHNIQUES WILL BE EVALUATED BY VARIOUS DATA PARTITIONING AND RESAMPLING METHODS. THE OBJECTIVE OF THIS RESEARCH IS TO MAKE USE OF THE EXTENSIVE PARAMETRIC DATA GENERATED BY THE IMS. BY GENERATING NEW RULES FROM THE DATA BY THE MACHINELEARNING ALGORITHMS AND INCLUDING THEM IN THE SYSTEM, THE COGNITIVE CAPABILITIES OF THE IMS CAN BE CONTINUOUSLY UPGRADED. THE AUTOMATIC DISCRIMINATION SCHEMES TO BE TESTED AND DEVELOPED UNDER THIS PROJECT CAN BE INCORPORATED INTO THE IMS AND UPGRADED CONTINUOUSLY AS NEW DATA BECOME AVAILABLE. THIS WOULD LEAD TO THE ANTICIPATED BENEFIT OF CONTINUOUS IMPROVEMENT OF THE DISCRIMINATION CAPABILITY OF IMS.