Prediction of natural events and location using hybrid pso with fuzzy Logic and density based spatiotemporal Clustering with gps

Abstract

Spatial Data Mining can fulfill the existent requirements of numerous geographic applications for disaster management. It permits acquiring the benefit of enhancing the accessibility of geographically referenced data and their probable aspects. From natural events data, mining knowledge is the most significant concern due to the rapid improvement and the wide utilization of the data attainment method. In general, for natural events predictions, the conventional techniques have been employed by modeling estimations of laser beam atmospheric extermination and meteorological data from manned and unmanned soil, hills, aerospace vehicles, and ocean conditions. These conventional methods can be timeconsuming to the execution of process and more expensive along with the ambiguity of precise prediction of natural events and amongst different problems the disasters can also be a serious threat in the scenario of today. Therefore, in this research, the following methods are proposed to the prediction of natural events and find the location of the disaster. Hybrid PSO with Fuzzy logic Manhattan distance-Density based Spatio-temporal clustering (MD-DBSTC), enhanced decision tree (EDT) with GPS In the first proposed system, hybrid PSO and strong fuzzy rules are presented to predict the natural events and disaster management. This novel model utilizing the spatial data mining methods for predicting the disaster events and their place has been presented for improvement. In this proposed method, the pre-processing method improvement by which the newline