{"title":"基于改进极值梯度增强的森林火灾面积预测模型","authors":"C. Ran, Lv Fang","doi":"10.1109/CCNS53852.2021.00011","DOIUrl":null,"url":null,"abstract":"The key state-owned forest areas in the Greater Khingan Mountains of Inner Mongolia are areas with a high incidence of forest fires. Accurate prediction of forest fire is necessary for forest fire prevention and effective control. This paper uses satellite fire and meteorological data in the Greater Khingan Mountains of Inner Mongolia as the experimental data set, and uses geographic information system software for data preprocessing. Temperature, air pressure, wind speed, elevation, etc. are selected as explanatory variables. The Extreme Gradient Boosting (XGBoost) is proposed to predicts the area of forest fire in the study area. Bayesian parameter adjustment method is used in the modeling process. The results show that the model is superior to traditional regression algorithms in terms of error parameters, training speed, and prediction accuracy.","PeriodicalId":142980,"journal":{"name":"2021 2nd International Conference on Computer Communication and Network Security (CCNS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Prediction model of forest fire area based on the improved Extreme Gradient Boosting\",\"authors\":\"C. Ran, Lv Fang\",\"doi\":\"10.1109/CCNS53852.2021.00011\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The key state-owned forest areas in the Greater Khingan Mountains of Inner Mongolia are areas with a high incidence of forest fires. Accurate prediction of forest fire is necessary for forest fire prevention and effective control. This paper uses satellite fire and meteorological data in the Greater Khingan Mountains of Inner Mongolia as the experimental data set, and uses geographic information system software for data preprocessing. Temperature, air pressure, wind speed, elevation, etc. are selected as explanatory variables. The Extreme Gradient Boosting (XGBoost) is proposed to predicts the area of forest fire in the study area. Bayesian parameter adjustment method is used in the modeling process. The results show that the model is superior to traditional regression algorithms in terms of error parameters, training speed, and prediction accuracy.\",\"PeriodicalId\":142980,\"journal\":{\"name\":\"2021 2nd International Conference on Computer Communication and Network Security (CCNS)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 2nd International Conference on Computer Communication and Network Security (CCNS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CCNS53852.2021.00011\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 2nd International Conference on Computer Communication and Network Security (CCNS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CCNS53852.2021.00011","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Prediction model of forest fire area based on the improved Extreme Gradient Boosting
The key state-owned forest areas in the Greater Khingan Mountains of Inner Mongolia are areas with a high incidence of forest fires. Accurate prediction of forest fire is necessary for forest fire prevention and effective control. This paper uses satellite fire and meteorological data in the Greater Khingan Mountains of Inner Mongolia as the experimental data set, and uses geographic information system software for data preprocessing. Temperature, air pressure, wind speed, elevation, etc. are selected as explanatory variables. The Extreme Gradient Boosting (XGBoost) is proposed to predicts the area of forest fire in the study area. Bayesian parameter adjustment method is used in the modeling process. The results show that the model is superior to traditional regression algorithms in terms of error parameters, training speed, and prediction accuracy.