Towards an orthogonality constraint-based feature partitioning approach to classify veracity and identify stance overlapping of rumors on twitter

作者：

Highlights：

• An orthogonality constraint-based feature partitioning approach to classify veracity.

• The proposed approach does not rely on any unreliable auxiliary information.

• SeNoCe shows promising improvements over the baselines on two real world datasets.

摘要

•An orthogonality constraint-based feature partitioning approach to classify veracity.•The proposed approach does not rely on any unreliable auxiliary information.•SeNoCe shows promising improvements over the baselines on two real world datasets.

论文关键词：Rumor,Veracity classification,Deep Neural Networks,Multi-Task Learning,Auxiliary information,Orthogonality constraint

论文评审过程：Received 30 May 2021, Revised 2 July 2022, Accepted 14 July 2022, Available online 22 July 2022, Version of Record 28 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118175