中文 | English

J. Hefei Univ. Tech. (Nat. Sci.)

合肥工业大学学报(自然科学版)

JOURNAL OF HEFEI UNIVERSITY OF TECHNOLOGY (NATURAL SCIENCE)

Competent Authorities: Ministry of Education of the People's Republic of China

Sponsor by: Hefei University of Technology

ISSN 1000-1646, CN 34-1001/N

Safety prediction model of building projects based on RS-PCA-SVM

Download Full Text

基于 RS-PCA-SVM 的建筑项目安全预测模型

Journal Information

J. Hefei Univ. Tech. (Nat. Sci.), Sept.2024, Vol.47 No.9: 1243-1247, 1261

DOI: 10.3969/j.issn.1003-5060.2024.09.013

Authors

LI Yongqing, MA Yabing, FENG Yahong

(School of Management, Xi'an University of Science and Technology, Xi'an 710054, China)

Abstract and Keywords

Abstract: In order to reduce the occurrence of safety accidents in building projects, a combined safety prediction model of building project based on RS-PCA-SVM is proposed. Rough set(RS) theory was adopted to perform attribute reduction for data, eliminating crossover and redundant information, reducing the dimension and computational complexity of input variables, and reducing the training time. On this basis, principal component analysis(PCA) was used for dimension reduction to remove the principal component with low contribution, and the principal component with high contribution was taken as the input variable of support vector machine(SVM). Particle swarm optimization(PSO) was used to optimize the parameters of SVM model to avoid the blindness of selecting parameters of SVM manually. The results show that the average prediction accuracy of this model is 93.78%. Compared with the traditional method, the prediction accuracy is higher and the calculation speed is faster.

Keywords: attribute reduction; principal component analysis(PCA); support vector machine(SVM); prediction model

作者信息

李永清，马亚冰，凤亚红

（西安科技大学管理学院，陕西西安710054）

摘要和关键词

摘要: 为了减少建筑项目安全事故的发生,文章提出一种基于RS-PCA-SVM建筑项目安全组合预测模型,采用粗糙集理论(rough set, RS)对数据进行属性约简,剔除交叉和冗余信息,降低输入变量维数和计算复杂度,减少训练时间;利用主成分分析(principal component analysis, PCA)法进行降维处理,除去贡献率较低的主成分,将剩余主成分作为支持向量机(support vector machine, SVM)的输入变量,并选择自适应权重粒子群优化算法(particle swarm optimization, PSO)优化SVM的参数,避免参数选择的盲目性。结果表明:该模型的平均预测准确率为93.78%,相比传统方法预测精度高、计算速度快。

关键词: 属性约简；主成分分析(PCA)法；支持向量机(SVM)；预测模型

Funding

陕西省科技厅软科学研究计划资助项目（2019KRM082）

Previous Next

Back to Contents