Automatic latent street type discovery from web open data

作者：

Highlights：

• Automatically learning latent street types using three Web data sources.

• The extended LDA model improves handling of multi-modal data.

• Learned street types matches traditional street type designs.

• The effect of learned street types is demonstrated on crime prediction.

摘要

•Automatically learning latent street types using three Web data sources.•The extended LDA model improves handling of multi-modal data.•Learned street types matches traditional street type designs.•The effect of learned street types is demonstrated on crime prediction.

论文关键词：Street classification,Web open data,Latent topic analysis,Urban computing

论文评审过程：Received 22 August 2019, Revised 30 March 2020, Accepted 19 April 2020, Available online 24 April 2020, Version of Record 29 April 2020.

论文官网地址：https://doi.org/10.1016/j.is.2020.101536