Few-shot cross domain event discovery in narrative text

• Published in: Information processing & management Vol. 62; no. 1; p. 103901
• Main Authors: He, Ruifang, Huang, Fei, Ma, Jinsong, Zhang, Jinpeng, Zhu, Yongkai, Zhang, Shiqi, Bai, Jie
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2025
• Subjects: Event discovery; Few-shot domain adaptation; Parameter adapter; Positive–negative balanced sampling; Event discovery; Few-shot domain adaptation; Positive–negative balanced sampling; Parameter adapter
• ISSN: 0306-4573
• DOI: 10.1016/j.ipm.2024.103901

Cross-domain event detection presents notable challenges in the form of data scarcity, and existing few-shot algorithms only consider events whose types are predefined, resulting in low coverage or excessive trivial identification results. To address this issue, this paper proposes the task Few-shot Cross Domain Event Discovery, which includes two subtasks: Domain Event Discovery and Few-shot Domain Adaptation. The former aims to identify the type-agnostic event triggers, and the latter completes domain adaptation with only a few annotated domain samples. Additionally, we introduce a positive–negative balanced sampling mechanism and a novel domain parameter adapter for these two subtasks, respectively. Extensive experiments on the DuEE dataset and the ACE2005 dataset show that our proposed method outperforms the current state-of-the-art method by 6.3% in Mix-F1 score on average. Moreover, we achieve SOTA performance in all domains of the DuEE dataset. •Define and formulate the new task Few-shot Cross Domain Event Discovery.•Propose positive–negative balanced sampling mechanism and domain parameter adapter.•Experimental results show the effectiveness and application potential of our method.