基于遗传算法的轨迹K匿名模型优化  

Anonymity Model Optimization of Trajectory K Based on Genetic Algorithm

作　　者：秦海涛 QIN Haitao(School of Information Engineering,Gansu Vocational and Technical College of Communications,Lanzhou 730070,China)

机构地区：[1]甘肃交通职业技术学院信息工程系,甘肃兰州730070

出　　处：《现代信息科技》2023年第15期63-68,共6页Modern Information Technology

摘　　要：常用轨迹隐私保护方法的得来离不开基于虚假轨迹的匿名研究,轨迹匿名方法生成虚假轨迹的不确定性及轨迹信息与背景知识之间的关联性,导致用户的真实轨迹隐私信息极易被识别。为此,文章提出基于遗传算法的轨迹k匿名模型优化算法。在用户真实轨迹的基础上,采用深度学习中有监督学习原理及幂律-对数函数解决分布函数中长尾数据问题,改进遗传算法中的变异操作和适应度函数,通过改进后的遗传算法来优化轨迹K匿名模型生成虚假轨迹的方法,并利用皮尔逊相关性计算轨迹相似性,调整轨迹中个体的位置,构建具有相同用户行为模式的k匿名轨迹集合。实验结果表明,该算法具有更好的适用性和隐匿性,降低了用户隐私披露风险。The development of commonly used trajectory privacy protection methods cannot be separated from anonymous research based on false trajectories.The uncertainty of false trajectory generated by trajectory anonymity methods and the correlation between trajectory information and background knowledge result in users'real trajectory privacy information being easily recognized.Therefore,this paper proposes a genetic algorithm-based trajectory k-anonymity model optimization algorithm.On the basis of the user's real trajectory,the principle of Supervised learning in deep learning and the power law logarithmic function are used to solve the problem of long tail data in the distribution function,the mutation operation and fitness function in the genetic algorithm are improved,and the improved genetic algorithm is used to optimize the method of generating false trajectory by the trajectory K anonymous model,and Pearson correlation is used to calculate the trajectory similarity and adjust the position of individuals in the trajectory,construct k anonymous trajectories set with the same user behavior patterns.The experimental results show that the algorithm has better applicability and concealment,reducing the risk of user privacy disclosure.

关 键 词：轨迹匿名 轨迹k匿名 遗传算法 虚假轨迹 用户行为模式 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]