Effective influence estimation in twitter using temporal, profile, structural and interaction characteristics

作者：

Highlights：

• Proposed two models Time Decay Features Cascade Model (TDF-C) and Time Decay Features Cascade Threshold Model (TDF-CT).

• Models integrate temporal, structural, profile and interaction characteristics of the social network in diffusion process.

• TDF-CT handles the limitations of the contemporary diffusion models, i.e., Independent Cascade and Linear Threshold.

• TDF-C and TDF-CT outperformed Independent Cascade, Time Constant Cascade, Time Decay Cascade & Time-Depth Decay Cascade.

• TDF-C and TDF-CT improve the influence propagation upto 39% with respect to contemporary models.

摘要

•Proposed two models Time Decay Features Cascade Model (TDF-C) and Time Decay Features Cascade Threshold Model (TDF-CT).•Models integrate temporal, structural, profile and interaction characteristics of the social network in diffusion process.•TDF-CT handles the limitations of the contemporary diffusion models, i.e., Independent Cascade and Linear Threshold.•TDF-C and TDF-CT outperformed Independent Cascade, Time Constant Cascade, Time Decay Cascade & Time-Depth Decay Cascade.•TDF-C and TDF-CT improve the influence propagation upto 39% with respect to contemporary models.

论文关键词：Influence diffusion,Temporal features,Interactions features,Structural features,Independent cascade,Linear threshold

论文评审过程：Received 29 January 2020, Revised 1 May 2020, Accepted 2 June 2020, Available online 30 June 2020, Version of Record 30 June 2020.

论文官网地址：https://doi.org/10.1016/j.ipm.2020.102321