Sensing the city with Instagram: Clustering geolocated data for outlier detection

作者：

Highlights：

• Instagram as geolocated data source for crowd detection.

• Obtaining spatio-temporal patterns of the distribution of crowds in the city.

• Establishing different reference days (Monday, Tuesday,...).

• On-the-fly detection of spatio-temporal outliers based on reference days.

• Validation using a real dataset in NYC on several special days.

摘要

•Instagram as geolocated data source for crowd detection.•Obtaining spatio-temporal patterns of the distribution of crowds in the city.•Establishing different reference days (Monday, Tuesday,...).•On-the-fly detection of spatio-temporal outliers based on reference days.•Validation using a real dataset in NYC on several special days.

论文关键词：Data mining,Location-based social network,Crowd detection,Instagram,Density-based clustering

论文评审过程：Received 3 June 2016, Revised 8 February 2017, Accepted 9 February 2017, Available online 14 February 2017, Version of Record 23 February 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.02.018