CCL23-Eval任务6系统报告:基于原型监督对比学习和模型融合的电信网络诈骗案件分类(System Report for CCL23-Eval Task 6: Classification of Telecom Network Fraud Cases Based on Prototypical Supervised Contrastive Learning and Model Fusion)
Site Xiong (熊思诗), Jili Zhang (张吉力), Yu Zhao (赵宇), Xinzhang Liu (刘欣璋), Yongshuang Song (宋双永)
Abstract
“本文提出了一种基于原型监督对比学习和模型融合的电信网络诈骗案件分类方法。为了增强模型区分易混淆类别的能力,我们采用特征学习与分类器学习并行的双分支神经网络训练框架,并通过领域预训练、模型融合、后置分类等策略优化分类效果。最终,本文方法在CCL2023-FCC评测任务上取得了Macro-F1为0.8601 的成绩。”- Anthology ID:
- 2023.ccl-3.22
- Volume:
- Proceedings of the 22nd Chinese National Conference on Computational Linguistics (Volume 3: Evaluations)
- Month:
- August
- Year:
- 2023
- Address:
- Harbin, China
- Editors:
- Maosong Sun, Bing Qin, Xipeng Qiu, Jing Jiang, Xianpei Han
- Venue:
- CCL
- SIG:
- Publisher:
- Chinese Information Processing Society of China
- Note:
- Pages:
- 201–205
- Language:
- Chinese
- URL:
- https://aclanthology.org/2023.ccl-3.22
- DOI:
- Cite (ACL):
- Site Xiong, Jili Zhang, Yu Zhao, Xinzhang Liu, and Yongshuang Song. 2023. CCL23-Eval任务6系统报告:基于原型监督对比学习和模型融合的电信网络诈骗案件分类(System Report for CCL23-Eval Task 6: Classification of Telecom Network Fraud Cases Based on Prototypical Supervised Contrastive Learning and Model Fusion). In Proceedings of the 22nd Chinese National Conference on Computational Linguistics (Volume 3: Evaluations), pages 201–205, Harbin, China. Chinese Information Processing Society of China.
- Cite (Informal):
- CCL23-Eval任务6系统报告:基于原型监督对比学习和模型融合的电信网络诈骗案件分类(System Report for CCL23-Eval Task 6: Classification of Telecom Network Fraud Cases Based on Prototypical Supervised Contrastive Learning and Model Fusion) (Xiong et al., CCL 2023)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2023.ccl-3.22.pdf