Efficient heterogeneous proximity preserving network embedding model
作者：
Highlights：
• Current proximity based network embedding models ignore the type information.
• Meta path is an effective concept to depict vertex proximity in HIN.
• Using meta path to guide embedding learning can preserve heterogeneous proximity.
• Exact proximity measurement is time consuming.
• Sampling based proximity measurement is more efficient for network embedding.
摘要
•Current proximity based network embedding models ignore the type information.•Meta path is an effective concept to depict vertex proximity in HIN.•Using meta path to guide embedding learning can preserve heterogeneous proximity.•Exact proximity measurement is time consuming.•Sampling based proximity measurement is more efficient for network embedding.
论文关键词：Network embedding,Heterogeneous information network,Random walk
论文评审过程：Received 15 May 2018, Revised 27 May 2019, Accepted 28 May 2019, Available online 29 May 2019, Version of Record 7 June 2019.
论文官网地址：https://doi.org/10.1016/j.eswa.2019.05.044
