摘要
为及时发现智慧城市潜在信息安全风险,构建一种基于改进惯性权重的粒子群优化(IIWPSO)算法优化反向传播(BP)(IIWPSO-BP)神经网络算法的信息安全风险预测模型。首先,综合考虑信息拥有者、共享信息、联盟链技术、信息使用者、联盟链管理和安全措施6个一级指标,构建信息安全风险指标体系;其次,通过量化信息安全风险指标,训练并测试所构建的信息安全风险预测模型;最后,对比分析模型的鲁棒性、精确性和时间复杂度。结果表明:IIWPSO-BP预测模型的平均绝对误差(MAE)为0.1374,平均相对误差(MRE)为0.0385,拟合度为0.9720;与PSO-BP神经网络、BP神经网络相比,预测精度分别提升了37.6%、65.2%。
In order to find potential information security risks of smart cities in time,an information security risk prediction model was built based on IIWPSO algorithm optimized BP(IIWPSO-BP)neural network algorithm.Firstly,the information security risk index system was constructed by considering six aspects:information owner,shared information,alliance chain technology,information user,alliance chain management and security measures.Secondly,the information security risk prediction model was trained and tested by quantifying the information security risk index.Finally,the robustness,accuracy and time complexity of the model were compared and analyzed.The results show that the mean absolute error(MAE)of the IIWPSO-BP prediction model is 0.1374,the mean relative error(MRE)is 0.0385,and the fitting degree is 0.9720.The prediction accuracy is improved by 37.6%and 65.2%,respectively,compared with the PSO-BP neural network and the BP neural network.
作者
周新民
罗文敏
刘俊杰
谢宝
ZHOU Xinmin;LUO Wenmin;LIU Junjie;XIE Bao(Key Laboratory of Hunan Province for New Retail Virtual Reality Technology,Hunan University of Technology and Business,Changsha Hunan 410205,China;Computer College,Hunan University of Technology and Business,Changsha Hunan 410205,China;Frontier Cross College,Hunan University of Technology and Business,Changsha Hunan 410205,China)
出处
《中国安全科学学报》
CAS
CSCD
北大核心
2022年第8期52-60,共9页
China Safety Science Journal
基金
国家社会科学基金资助(21BGL231)。
