Abstract: The data analysis and application of Internet of vehicles under the background of intelligent network has an important impact on improving traffic intelligence. In order to speed up the process of traffic intelligent, data redundancy in the data application of Internet of vehicles is studied. The data collected from the Internet of vehicles is taken as the research object, and the classification and identification of driving behavior characteristics is taken as the research object. Correlation analysis and principal component analysis were used for redundancy screening and dimensionality reduction of data, and k-means clustering algorithm was used for classification and identification of driving behavior characteristics. The results show that the data dimension reduction method can reduce the correlation redundancy of the Internet of vehicles data, and the characteristics of driving behaviors can be classified into three types of driving behaviors. The research improves the application value of the data of the Internet of vehicles, and also provides relevant support for traffic intelligence.

Key words: Intelligent network; Internet of vehicles; Principal component analysis; K-means algorithm

摘要： 智能网联背景下的车联网数据分析与应用对提升交通智能化有重要影响。为了加快交通智 能化进程，文章对车联网数据应用过程中存在的数据冗余问题进行研究。以采集的车联网数据为 研究对象，驾驶行为特点分类辨识为研究目标。采用相关分析与主成分分析方法对数据进行冗余 筛选与降维，使用 k-means 聚类算法对驾驶行为特点进行分类辨识。研究结果表明，使用数据降 维的方法可以降低车联网数据的相关冗余性，驾驶行为特点分类辨识结果表明其特点可分为三类 驾驶行为。研究提升了车联网数据的应用价值，也为交通智能化提供了相关的支持。

关键词: 智能网联；车联网；主成分分析；k-means 算法