数据耕耘技术支持下的大规模驾驶行为数据分析仿真技术  

Analysis and Simulation System of Large-scale Driving Behavior Data with Data Farming

作　　者：王磊[1] 彭敦陆[1] 

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《小型微型计算机系统》2016年第4期705-710,共6页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61003031)资助;上海市自然科学基金项目(10ZR1421100)资助;上海市研究生创新基金项目(JWCXSL1402)资助

摘　　要：大数据相关问题已成为当今数据科学领域的研究热点,在研究中需要有足够多的数据来验证相关研究成果.由于各种条件限制,研究人员难以得到足够体量的源数据,给研究带来了很大的困难.论文运用数据耕耘技术研究出一种用于大数据系统仿真的方法.该方法首先对被研究对象属性特征进行分析,确定属性之间的依附关系.然后选择合适的数据耕耘方法建立仿真系统,最后对仿真系统进行验证,并以大规模驾驶行为数据分析为例,说明如何利用该技术建立大数据分析及仿真的过程.At present,Big Data related problems have become the research hotspots in the field of data science.During the experimental study of Big Data we need enough data to verify the relevant research results.However,for some reason,it is usually hard for researchers to get enough data before the data collecting systems is put to use.This has brought great difficulties for the study.In this paper,a simulation system is provided for generating big volume of data by using the technique of data farming.It analyzes the dependent relationship between attributes before exploring the appropriate data farming methods to build the simulation system.Some approaches are also proposed to validate the simulation system.Besides,to illustrate the process of building the simulation,it takes the analysis of vehicle driving behavior with large scale simulated data as the example.

关 键 词：数据耕耘 大数据 仿真系统 驾驶行为 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]