Earthquake disaster loss prediction and evaluation method and system based on Softmax regression model

Abstract

The invention discloses an earthquake disaster loss prediction and evaluation method and system based on a Softmax regression model. The invention relates to the technical field of earthquake disasterloss prediction and evaluation. The problem of how to improve the precision and training speed of earthquake disaster loss prediction is solved. The method comprises the steps of: serving different earthquake disaster loss degree levels as classification labels of a Softmax regression classification model, selecting past earthquake disaster feature data for training, wherein the Softmax regression classification model is used for predicting past earthquake disaster loss; inputting new earthquake disaster feature data and classification labels into the Softmax regression classification model;judging that the input data belongs to the classification weight of each earthquake disaster loss degree level; determining a classification label corresponding to the input data according to the classification weight so as to determine the loss degree grade of the new earthquake disaster. Compared with a BP (Back Propagation) neural network and an SVM (Support Vector Machine), the method and thesystem provided by the invention have the advantages that the capability of distinguishing the earthquake disaster loss is stronger, the test precision is high, and the test time is short.

Description

technical field [0001] The invention relates to the technical field of earthquake disaster loss prediction and evaluation, in particular to a method and system for earthquake disaster loss prediction and evaluation based on a Softmax regression model. Background technique [0002] Earthquake, as one of the natural disasters that are difficult to predict accurately, has become the biggest security threat to human society. my country is one of the countries with the strongest seismic activity and the most severe earthquake disaster losses in the world. Earthquakes cause huge personal and property economic losses every year, and have a considerable impact on the sustainable development of economy, society and environment. Therefore, rapid and accurate assessment and prediction analysis of earthquake disaster losses is a key link in disaster risk management, especially emergency management. [0003] In the prior art, in the literature "Research on Earthquake Casualty Assessment...

AI Technical Summary

Problems solved by technology

There are many common algorithms for solving multi-classification problems, such as decision tree, naive Bayesian, backpropagation (BP) neural network, support vector machine (SVM), etc.; the disadvantages of BP neural network It is local minimization; algorithm convergence speed is slow; the choice of network structure is not easy to determine; the approximation and generalization ability of the network model is closely related to the typicality of the learning samples. It is a difficult problem to select typical sample instances from the problem to form the training set; the disadvantages of SVM The reason is that the classic SVM algorithm only gives the algorithm of the second-class classification, and cannot predict multi-level disasters. Therefore, it needs to be solved by combining multiple second-class SVMs, which increases the computational complexity.

Images