APPLICATION OF PATTERN RECOGNITION TO THE STUDY OF THE DYNAMIC FACTOR OF OCCURRENCE OF LARGE EARTHQUAKES

A pattern recognition algorithm has been used to correlate the occurrence time of a large earthquake to intermediate-size earthquake activity, to the level of sunspot activity and the rotation of the earth. The characteristics given by pattern recognition suggest that, before the occurrence of a large earthquake when the level of intermediate earthquake activity increases to a definite level, may be an important characteristic for the occurrence of that large earthquake. As well around the extreme values of sunspot activity and the time interval of relatively large variations of the acceleration of rotation of the earth may also trigger large earthquakes.