Data Mining Techniques for Earthquake Frequency-Magnitude Analysis and Seismic Zone Estimation

摘要

India has had a number of the world's greatest earthquakes in the last century. In fact, more than 50% area in the country is considered prone to damaging earthquakes. The northeastern region of the country as well as the entire Himalayan belt is susceptible to great earthquakes of magnitude more than 8.0. The main cause of earthquakes in these regions is due to the movement of the Indian plate towards the Eurasian plate at the rate of about 50 mm per year. The earthquake zoning map of India divides India into 4 seismic zones (Zone 2, 3, 4 and 5). According to the present zoning map, Zone 5 expects the highest level of seismicity whereas Zone 2 is associated with the lowest level of seismicity. In this Paper, we have derived regression relations on Earthquake Magnitudefrequency Data Set using statistical tools like SAS (Statistical Analysis System) and Weka (Waikato Environment for Knowledge Analysis). The regression relations obtained are the first relations for this region. In Earthquake Disaster Management Data Set, Magnitude is considered as a Dependent variable. It depends on LocationRank, StateRank, Elevation, Population, Year, Month, Day, Hour, Minute, Sec, Latitude, Longitude and Depth which are considered as Independent Variables. Based on Dependent and Independent variables, the new ‘seismic zone’ can be analyzed at different earthquake locations. If magnitude is equal to 7 or more, large areas are damaged depending on their depth. If magnitude is equal to 3 or less, the probability of occurrence of an earthquake is weak.