Context-aware personalized path inference from large-scale GPS snippets

作者：

Highlights：

• A context-aware personalized path inference model is proposed.

• The model incorporates both the road average speeds and the personal preferences.

• An EM based learning technique is proposed to discover the road average speeds.

• Matrix factorization is used to extract the personal preferences.

• Experiments were carried out on real world databases yielding better performance.

摘要

•A context-aware personalized path inference model is proposed.•The model incorporates both the road average speeds and the personal preferences.•An EM based learning technique is proposed to discover the road average speeds.•Matrix factorization is used to extract the personal preferences.•Experiments were carried out on real world databases yielding better performance.

论文关键词：Path inference,Context-aware,Personalization,Conditional random field,EM

论文评审过程：Received 1 November 2016, Revised 12 August 2017, Accepted 13 August 2017, Available online 17 August 2017, Version of Record 1 September 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.08.027