Research on location fusion of spatial geological disaster based on fuzzy SVM

Abstract

In order to effectively improve geological disaster response capacity, disaster tolerance capacity, reduce human and financial losses and other aspects of spatial data fusion plays a key role. Whether the location information can be effectively fused is able to monitor the occurrence of geological hazards and effectively identify the existing risks. In machine learning, the similarity and complementarity between support vector machine and fuzzy inference is the basis of their fusion. Support vector machine can achieve knowledge acquisition and learning, while fuzzy inference has the ability to infer knowledge rules. Aiming at the different attributes and dimensions of information spatial location data, a fuzzy fusion method based on support vector machine is proposed to describe Support vector machine related theories and models. Comparing the efficiency of support vector machine-fusion algorithm on GPLUS, OKLAHOMA and UNC with the other three algorithms, there is a great advantage in RMSE and time. The algorithm in this paper also has good performance in the three data sets on F1, which shows that the algorithm has a good effect.