基于遗传算法的用户行为数据生成  

User behavior data generation based on genetic algorithm

作　　者：方文辉 胡竹林[2] 朱欣娟[1] Fang Wenhui;Hu Zhulin;Zhu Xinjuan(School of Computer Science,Xi'an Polytechnic University,Xi'an 710000,China;Digital Resources Department of Shaanxi Provincial Library,Xi'an 710000,China)

机构地区：[1]西安工程大学计算机科学学院,西安710000 [2]陕西省图书馆数字资源部,西安710000

出　　处：《国外电子测量技术》2021年第9期154-159,共6页Foreign Electronic Measurement Technology

基　　金：国家重点研发计划(2019YFC1521405)项目资助。

摘　　要：用户使用互联网产生了大量的行为数据,这些大数据在分析用户行为,进行精准推荐,构造知识图谱和用户画像,改善用户体验方面具有重要的作用。由于种种原因,一直以来,用户及其行为数据的采集获取较为困难。提出了一种基于遗传算法的用户行为数据生成方法,该方法以图书馆借阅用户为例,首先对用户行为数据中的每个特征进行级联编码,将编码组合在一起,形成初始化种群;然后根据原始数据的概率分布得到适应度函数,对种群进行优化;得到最优解或达到迭代的结束条件,生成模拟数据。实验结果表明,该方法可以依据需求快速生成大量用户行为数据,且生成的数据基本符合原始数据特征分布。A large amount of behavioral data was generated when users use the internet.These data play an essential role in analyzing user behaviors,making accurate recommendations,constructing knowledge graph and user portraits,and improving user experience.For various reasons,it has been difficult to collect and obtain data on users and their behaviors.In this paper,the author proposes a genetic algorithm-based method for generating user behavior data.This method is based on the example of library checkout users.Firstly,cascade coding for each feature in the user behavior data,then combining codes together to form initialized populations.Then the fitness function is obtained from the probability distribution of the original data so that the population is optimized.The simulation date can be generated once the optimal solution is obtained or the end condition of the iteration is reached.The experimental results show that this method proposed in the paper can quickly generate a large amount of user behavior data based on specific demands.Moreover,the generated data basically matches the feature distribution of the original data.

关 键 词：用户及其行为数据 数据生成 遗传算法 级联编码 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]