摘要:Abstract Nowadays, tremendous data, are continuously gathering from the smart card in public transport domain. Such data, conveying two viable distinct information, can ensue designing intelligent transportation. More specifically, users behavior in a public transport system, can be investigated, as one of the data mining and machine learning applications. The first component of the data, provides the spatial feature, indicates the geographical coordinates of bus stops or subway stations. The second component of the data, deals with the temporal feature, being the time of the trips that public transport is used. Hence, it is necessary to distill the data, in order to get the advantages of the data analysis techniques and extract the essential knowledge from the data. Due to the massive data storage and the diversity of the data analysis methods, various challenges are arisen during the process of exploiting the hidden patterns of the data. We review a couple of scenarios and suggest a solution to overcome a number of the raised challenges. Moreover, the other aspects of this problem, are remaining as the open problems for the future research.
关键词:KeywordsClusteringPublic transportSmart cardSpatial-Temporal data