BCMM: A novel post-based augmentation representation for early rumour detection on social media
作者：
Highlights：
• This paper proposes a novel post-based augmentation representation for early rumour detection.
• The network topology and metadata improve the performance of method.
• We construct backward compression mapping mechanism to enhance the semantic of posts after the vectorization processing.
• The proposed method and its variants are tested on four datasets.
• The method enables a significant increase in accuracy for early rumour detection.
摘要
•This paper proposes a novel post-based augmentation representation for early rumour detection.•The network topology and metadata improve the performance of method.•We construct backward compression mapping mechanism to enhance the semantic of posts after the vectorization processing.•The proposed method and its variants are tested on four datasets.•The method enables a significant increase in accuracy for early rumour detection.
论文关键词：Early rumour detection,Topology network,Metadata,Backward compression mapping,Semantic augmentation
论文评审过程：Received 15 May 2020, Revised 27 October 2020, Accepted 29 December 2020, Available online 12 January 2021, Version of Record 20 January 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.107818
