Data analysis on train transportation data with nonnegative matrix factorization
Data analysis on train transportation data with nonnegative matrix factorization
复制标题
DOI：
10.1109/bigdata.2017.8258425
复制
发表时间：
2017-12
期刊：
影响因子：
--
通讯作者：
Kyoichi Ito;Masaki Ito;Kosuke Miyazaki;K. Tanimoto;K. Sezaki
中科院分区：
文献类型：
--
作者：
Kyoichi Ito;Masaki Ito;Kosuke Miyazaki;K. Tanimoto;K. Sezaki
鉴于人们认识到需要收集和分析数据以维持城市发展，“智慧城市”概念最近受到了广泛关注。传感和信息技术的发展促进了对城市流动性的分析，以更好地了解城市的特点。在可用于描述城市特征的信息和数据中，交通数据是最有用的，因为交通与人和城市流动的其他方面密切相关。在从自动收集的数据中提取特征时，最大的困难来自数据集的大小或复杂性，因为这些数据集通常具有太多的属性或索引来进行分析。本文讨论了使用非负矩阵分解（NMF）的智能卡票务认证日志的分析结果。结果提出提取的功能，适用于评估各种用户和电台的特点和动态。
In light of the recognized need to collect and analyze data to maintain urban development, the “smart city” concept has gained much attention recently. The development of sensing and information techniques has facilitated the analysis of urban mobility to better understand the characteristics of cities. Of the information and data that can be used to characterize cities, transportation data are among the most useful because transportation is so closely related to human and other aspects of urban mobility. In extracting features from automatically collected data, the greatest difficulty comes from the size or complexity of the data set, as these often have too many attributes or indices to analyze. This paper discusses the results of analyses of smart card ticketing authentication logs using nonnegative matrix factorization (NMF). The results present extracted features applicable to assessing various user and station characteristics and dynamics.