摘要
针对使用当前方法检测建筑安全防护驱动因素时获取的数据与建筑安全防护驱动因素之间关联程度较低,导致在检测准确率低和检测效率低的问题,将数据挖掘算法应用到建筑安全防护驱动因素的检测中,提出数据挖掘下建筑安全防护驱动因素检测方法,根据数据相似程度获得建筑安全防护测试数据集。利用数据挖掘算法检测建筑安全防护驱动因素,通过预选候选支持向量机,降低训练样本数量,缩短训练时间,利用近邻样本密度设计隶属度函数,通过隶属度函数训练支持向量机,将建筑安全防护测试数据集输入支持向量机中,实现建筑安全防护驱动因素的检测。仿真结果表明,所提方法的检测准确率高、检测效率高。
At present, low correlation between the data obtained by the current method and the driving factors of building safety protection leads to low detection accuracy and low detection efficiency. Therefore, a data mining algorithm was applied to the detection of driving factors of building safety protection. In this article, a method to detect the driving factors of building safety protection under data mining was proposed. According to the similarity degree between data, the test data set of building safety protection was obtained. The data mining algorithm was used to detect the driving factors of building safety protection. After preselecting the candidate support vector machine, the number of training samples was decreased and the training time was reduced. Moreover, the density of adjacent samples was used to design the membership function, and then the support vector machine was trained by the membership function. Finally, the test data set of building safety protection was inputted into the support vector machine. Thus, the detection for the driving factors of building safety protection was achieved. Simulation results prove that the proposed method has high detection accuracy and high detection efficiency.
作者
徐建中
陈潜心
李冰洋
XU Jian-zhong;CHEN Qian-xin;LI Bing-yang(Harbin Engineering University,Harbin Heilongjiang 150001,China)
出处
《计算机仿真》
北大核心
2020年第6期430-433,共4页
Computer Simulation
关键词
数据挖掘
建筑安全
驱动因素
支持向量机
隶属度函数
Data mining
Building safety
Driving factor
Support vector machine
Membership function
