决策树模型法评估外来医学媒介生物的输入风险  

The application of decision tree on the risk assessment of exotic medical-vector

作　　者：裘炯良[1] 孙志[1] 王军[1] 郑剑宁[1] 杨定波[1] 

机构地区：[1]宁波出入境检验检疫局,浙江宁波315012

出　　处：《中华卫生杀虫药械》2016年第2期137-141,144,共6页Chinese Journal of Hygienic Insecticides and Equipments

基　　金：国家质检总局科技计划项目(编号:2012B172)

摘　　要：目的应用决策树模型法评估研究入境国际航行船舶携带输入外来医学媒介生物的风险。方法以中国第二大港、世界第五大港的宁波港作为研究范围,以2014年到港的国际航行船舶为研究对象,对850艘媒介生物阳性船舶和2 183艘媒介生物阴性船舶的33项指标展开调查。应用决策树模型法建模训练与验证,并以所建模型预测新到港的1 364艘次船舶外来媒介携带率。结果模型筛选出船舶总吨、净吨、船龄、到达季节、船型、船员类型、SSC证书签发天数等7个有效预测变量。训练获得的决策树共有12个有效叶节点,对应12条分类规则。训练过程的正确分类率为73.74%,验证过程的正确分类率为72.30%。通过该模型预测船舶携带外来媒介风险与实际检疫结果的符合率达到82.70%,预测效果良好。结论针对高度不确定的非线性系统,应用决策树模型法可实现相对精确的预测功能,为国境卫生检疫风险评估及预警方面的研究提供理论基础。Objective To apply the decision tree on the risk assessment of exotic medical-vector captured in the international navigation ships based on SAS / EM. Methods Ningbo Port as the No. 2 in China and No. 5 in the world was taken as the researching area,and the arrival international navigation ships in 2014 as the researching objects. 33 indexes were surveyed for 850 vector-positive vessels and 2 183 vector-negative vessels. Decision tree model was employed for the training and verifying calculation. 1 364 new-arrival vessels were used for the prediction by the model. Results Seven predicting variables including gross tonnage,net tonnage,vessel age,arrival season,vessel type,crew category,and duration of SSC certificate issued were screened out as the effective variables. The trained decision tree had 12 leaf nodes,corresponding to 12 classifying rules. The correct rate of the training was73. 74%,while the validation rate was 72. 30%. The predictive condition was good as the according rate attained82. 70%. Conclusion The decision tree model can provide the theoretical base for the risk analysis and alert of health quarantine.

关 键 词：决策树模型 外来媒介生物 风险预测 SAS 

分 类 号：R384[医药卫生—医学寄生虫学] R184.3[医药卫生—基础医学]