期刊文献+

2种药剂防治麦叶蜂生测的SVM,TDM模型分析

SVM and TDM Model Analysis of Bioassay on Dolerus tritici Control with Two Pesticides
下载PDF
导出
摘要 支持向量机(SVM)模型与时间-剂量-死亡率(TDM)模型都是兼顾时间效应与剂量效应的毒理学生测数据处理方法,而吡虫啉和敌百虫则是有着不同杀虫机理的2种化学杀虫剂。对这2种药剂应用浸渍法进行室内防治麦叶蜂3龄幼虫的毒力测定,并利用SVM模型和TDM模型对其生测数据进行了分析。SVM回归分析结果表明,20 min敌百虫的LD50为20.995 mg/kg,吡虫啉为102.886 mg/kg;SVM模型的拟合效果优于TDM模型,且MSE(均方误差)均较小。所以,SVM模型有很好的鲁棒性,在击倒活性化学杀虫剂生测数据处理中所得的结果准确性较高。 Both support vector machine(SVM)model and time-dose-mortality(TDM)model are technologies of toxicological data analysis that take into account both time and dosage effects, furthermore dipterex and imidacloprid are two kinds of chemical insecticides of different mechanisms. The indoor virulence bioassay of these pesticides was carried out on Dolerus tritici control with maceration method, and the data were processed by SVM and TDM model. The result showed that the LD50 values of dipterex and imidacloprid in 20 minutes were 20.995 mg/kg and 102.886 mg/kg respectively, while the fit effects of two SVM models were better than two TDM models' because the MSE values of two SVM models were less than the values of two TDM models. Generally SVM model had a better Modeling Feasibility Robustness, so the veracity of this model was perfect in the bioassay data analysis of insecticides which had the knock-down effect.
出处 《山西农业科学》 2014年第9期1003-1006,共4页 Journal of Shanxi Agricultural Sciences
基金 国家现代农业产业技术体系建设专项基金项目(CARS-45-SYZ10) 河南科技学院重点资助基金项目(0500136) 新乡市农作物重大有害生物防控重点实验室开放基金项目
关键词 麦叶蜂 SVM模型 TDM模型 吡虫啉 敌百虫 Dolerus tritici support vector machine(SVM)model time-dose-mortality(TDM)model imidacloprid dipterex
  • 相关文献

参考文献18

  • 1NelloCristianini JohnShawe-Taylor 李国正 译.支持向量机导论[M].北京:电子工业出版社,2002..
  • 2罗正根.基于SVM的昆虫分类与TDM建模[D].长沙:湖南农业大学,2011.
  • 3王志明,谭显胜,周玮,袁哲明.基于支持向量回归的生物测定数据分析[J].昆虫学报,2010,53(12):1436-1441. 被引量:4
  • 4刘明涛,肖东峰.麦叶蜂综合防治技术[J].生物防治通报,1999,3(2):78-80.
  • 5赵善欢.植物化学保护[M].北京:农业山版社,1998:122-124.
  • 6陈年春.农药生物测定技术[M].北京:北京农业大学出版社,1996..
  • 7林毅,蔡福营,张光亚.苏云金杆菌杀虫晶体蛋白活性预测的支持向量机模型[J].生物工程学报,2007,23(1):127-132. 被引量:7
  • 8向昌盛,周子英,张林峰.支持向量机在害虫发生量预测中的应用[J].生物信息学,2011,9(1):28-31. 被引量:12
  • 9Tolambiya A, Kalra P K. Contrast sensitive epsilan-SVR and its application in image compression[C]. Singapore: IEEE Internation- al Conference on Systems, 2008 : 359-364.
  • 10林弘德.LibSVM[EB/OL].[2014-06-16].http://ntu.csie.org/-pi-aip/svm/svm_tutorial.html.

二级参考文献112

共引文献79

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部