一种基于本体的交通大数据分析框架.doc
一种基于本体的交通大数据分析框架 摘要:随着城市交通数据信息的日益增多,对交通数据的分析变得越加困难。现有的交通大数据分析面临对大量分散且异构的数据信息进行过滤、筛选以及整合等难题。因此,该文通过在
一种基于本体的交通大数据分析框架 摘要:随着城市交通数据信息的日益增多,对交通数据的分析变得越加困难。现有的交 通大数据分析面临对大量分散且异构的数据信息进行过滤、筛选以及整合等难题。因此,该 文通过在路网拓扑、道路交通对象和道路交通信息三个层次对交通数据相互关系进行描述, 并使之关联交通大数据存储信息,构建面向交通大数据的城市道路交通本体模型,提出了一 种基于本体的交通大数据分析框架。该框架中以城市道路交通本体为语义规范制定映射文件, 利用Jena开发工具构建城市道路交通本体库,为大数据分析的多源多维数据关联分析及知 识挖掘提供语义查询支持。该框架能够根据交通分析需求快速有效地找到目标数据,在大数 据分析与大数据存储之间起到了逻辑关联的作用,对现有交通数据分析具有重要意义。 中国论文网 http://www.xzbu.com/8/view-7241552.htm 关键词:道路交通;大数据;本体 中图分类号:TP391 文献标识码:A 文章编号:1009-3044(2016)01-0025-03 AKind of Traffic Data Analysis Framework Based on Ontology YAN Jun-wei, LING Qing-wei, WANG Jian (CIMS,Research Center,Tongji University, Shanghai 201804, China) Abstract: With the increasing of urban traffic data information, the analysis of traffic data becomes more difficult. Existing traffic data analysis faces problems of filtering, screening and integration of alarge number of dispersed and heterogeneous data information. Therefore, this paper proposes a traffic data analysis framework based on ontology by building ontology model for traffic data of urban road traffic .The ontology model describe the relationship between data through network topology, the object of road traffic and road traffic information and related storage information of transport large data.The framework uses urban road traffic ontology as mapping file for semantic specification, and uses Jena development tools make an ontology construction of urban road traffic, for big data analysis of multi-source multi-dimension data correlation analysis and knowledge discovery with semantic query support. The scheme can find the target data based on the analysis of traffic demand quickly and efficiently ,and have played an important role in logic between big data analysis and data storage.It is of great significance to the existing traffic data analysis. Key words: road traffic; Big data; ontology 1概述 随着城市规模的不断扩大以及城市聚集效应的加强,一方面城市交通数据日益增长,数 据种类和数量高速增长;另一方面市民的密集出行以及物流压力使城市交通面临着严峻的挑 战。针对上述现象,如何快速有效地对交通数据进行分析是目前亟待解决的重要问题。 由于交通信息源的多样性和交通信息量的不断增加等原因,信息交互共享的实现以及如 何对有价值地信息的进行检索变得日趋复杂,传统的基于关键字的搜索几乎不能实现语义 [1],很难满足大数据分析多源多维数据关联分析及知识挖掘的查询要求。而本体作为一种 可以在语义以及知识层次上描述信息系统的概念模型,提供了资源描述和查询所必需的元语、 可以为信息源提供必要的语义标注。同时由于交通数据具有多维多量的特性,通过构建本体

