Construction And Application Of Structured Scenario Model Of Unconventional Emergency Based On Random Forest

In recent years,the scale of unconventional emergencies is gradually expanding,which seriously affects the rapid growth of our country economy and the stable development of society.All kinds of catastrophic events all over the world also break out frequently,and the scope of occurrence continues to expand,so that the traditional small-scale and untimely treatment of conventional emergencies gradually evolve into unconventional emergencies with wide coverage and extremely difficult recovery.Since unconventional emergencies are characterized by inexplicitness,extreme rarity,complexity,unpredictability,great destruction and urgent response time,when unconventional emergencies break out,their derivative disaster events are more difficult to deal with than conventional emergencies.At present,there is no welldeveloped emergency response to the outbreak of unconventional emergencies,which bring great difficulty to the emergency management department.Therefore,the emergency treatment of unconventional emergencies has become a hot research topic.In this thesis,python is used for simulation and random forest related theoretical knowledge,which is combined to study the structured system of unconventional emergency scenarios.Firstly,this article introduces the research background and significance as well as the current research results at home and abroad,based on this,the structured scenario system of unconventional emergencies is analyzed.Then,using the relevant theoretical knowledge of random forest and according to the logic of scenario evolution of unconventional emergencies,a structured scenario model of unconventional emergencies based on random forest is proposed.The emergencies are classified by using the random forest model.A comparative experiment based on Naive Bayes model and support vector machine model is proposed for different data sets under the same experimental environment,the results show that the average accuracy of the random forest model increased by at least 8.77%.Finally,practical cases are used to verify the random forest regression model,and the predicted results are very close to the real value,which has high application value.In order to further analyze unconventional emergencies,python is used to continue simulation and study the scenario evolution of unconventional emergencies.Firstly,the evolution model of unconventional emergencies is constructed by using fuzzy reasoning rules and scenario evolution path.Then,ID3 decision tree algorithm is used to predict the classification evolution results of events.To improve the accuracy of classification tree evolution,tree model pruning is carried out.Finally,the difflib model,fuzz model,edit distance model and cosine model are used to calculate the similarity of classification results,the results show that the similarity is very high,which verifies the reliability and accuracy of the evolution model.