该文针对当前法律智能体系可解释性差、低频易混淆法因预测效果不佳、民事纠纷研究过少的问题,设计了一种可解释性层次法因预测IHLCP模型,并将法因之间的层次依赖关系作为模型可解释性的来源进行了研究。模型首先基于案件的语义差异性对事实描述进行编码,然后通过改进的seq2seq-attention模块来预测法因路径,并利用法因内部的文本信息过滤事实描述中的噪声信息,以获得可靠的预测效果。该文设计的IHLCP模型在CIVIL、FSC和CAIL这三个大规模公开数据集上分别达到了当前最好的效果(CIVIL数据集: ACC-91.0%,PRE-67.5%,RECALL-57.9%,F1-62.3%。FSC数据集: ACC-94.9%,PRE-78.8%,RECALL-75.9%,F1-77.3%。CAIL数据集: ACC-92.3%,PRE-90.9%,RECALL-89.7%,F1-90.3%),其中ACC和F1值分别最高提升了6.6%和13.4%。实验结果表明,该设计能够帮助系统理解法因,弥补了当前法律智能体系在低频、易混淆法因预测上的不足,同时提升了模型的可解释性。
Abstract
To address such issues as the poor interpretability of current legal intelligence system, the unsatisfactory prediction of less-frequent and confusing legal causes and the insufficient research on civil disputes, an interpretable hierarchical legal causes prediction model (IHLCP) is proposed, taking the hierarchical dependence between legal causes as the source of interpretability. In IHLCP, the fact description is encoded by capturing the semantic differences of cases, and an improved attention-based seq2seq model is used to predict the cause path. Further, the inner text information of the cause is used to filter out the noise information in the fact description. Experiments show that the IHLCP model designed in this paper has achieved the state-of-art performance on three large-scale data sets: CIVIL (ACC-91.0%, Pre-67.5%, Recall-57.9%, F1-62.3%), FSC (ACC-94.9%, PRE-78.8%, RECALL-75.9%, F1-77.3%) and CAIL (ACC-92.3%, Pre-90.9%, Recall-89.7%, F1-90.3%), boosting the ACC and F1 by 6.6% and 13.4%, respectively. The experimental resuces show that this model can help the system to understand the law causes, make up for the start comings of current legal intelligence system in few-shot and confusing Causes of law prediction, make up for the deficiency of low frequency confusing cause prediction and improve the inter pretability of the model.
关键词
层次法因预测 /
可解释性 /
语义差异性 /
数据不平衡 /
低频类别预测
{{custom_keyword}} /
Key words
hierarchical legal cause prediction /
interpretability /
semantic differences /
data imbalance /
few-shot prediction
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] BOUTEL M R, LUO J, SHEN X, et al. Learning multi-label scene classification[J]. Pattern Recognition, 2004, 37(9): 1757-1771.
[2] SUNG F, YANG Y, ZHANG L, et al. Learning to compare: Relation network for few-shot learning[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 1199-1208.
[3] HU Z K, LI X, TU C H, et al. Few-shot charge prediction with discriminative legal attributes[C]//Proceedings of COLING, 2018: 487-498.
[4] DEVLIN J, CHANG M W, LEE K, et a1. BERT: Pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the NAACL,2019: 4171-4786.
[5] YANG Z, DAI Z, YANG Y, et al. XLNet: Generalized autoregressive pretraining for language understanding[C]//Proceedings of the NIPS, 2019: 5754-5764.
[6] LUO B,FENG Y,XU J, et al. Learning to predict charges for criminal cases with legal basis[C]//Proceedings of the EMNLP, 2017: 2727-2736.
[7] HENDRICKS LA, AKATA Z, ROHRBACH M, et al. Generating visual explanations[C]//Proceedings of the ECCV, 2016: 3-19.
[8] ZHONG H, WANG Y, TU C, et al. Iteratively questioning and answering for interpretable legal judgment prediction[C]//Proceedings of the AAAI, 2020, 34(1): 1250-1257.
[9] YE H, JIANG X,LUO Z C, et al. Interpretable charge predictions for criminal cases: Learning to generate court views from fact descriptions[C]//Proceedings of the NAACL,2018: 1854-1864.
[10] DAI Z,YANG Z,YANG Y, et al. Transformer-XL: Attentive language models beyond a fixed-length context[C]//Proceedings of the ACL, 2019: 2978-2988.
[11] VASWANI A,SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the NIPS, 2017: 6000-6010.
[12] JOHNSON R, TONG ZH. Deep pyramid convolutional neural networks for text categorization[C]//Proceedings of the ACL,2017: 562-570.
[13] KIM Y. Convolutional neural networks for sentence classification[C]//Proceedings of EMNLP, 2014: 1746-1751.
[14] IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the International Conference on Machine Learning. 2015: 448-456.
[15] GLOROT X, BORDES A, BENGIO Y. Deep sparse rectifier neural networks[C]//Proceedings of the AISTATS, 2011: 315-325.
[16] MISRA D. Mish: A self regularized non-monotonic neural activation function[J]. Arxiv prepvine arXiv: 2019: 1908.08681.
[17] SUTSKEVER I, VINYALS O, LE Q. Sequence to sequence learning with neural networks[C]//Proceedings of the NIPS,2014: 3104-3112.
[18] BAHDANAU D,CHO K,BENGIO Y . Neural machine translation by jointly learning to align and translate[J]. ArXiv preprint ArXiv: 2014: 1409.0473.
[19] ZHANG W,FENG Y,MENG F, et al. Bridging the gap between training and inference for neural machine translation[C]//Proceedings of the ACL, 2019: 4334-4343.
[20] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017(99): 2999-3007.
[21] XIAO C,ZHONG H,GUO Z, et al. CAIL2018: A large-scale legal dataset for judgment prediction[J]. ArXiv preprint Arxiv, 2018.
[22] YANG Z,YANG D,DYER C, et al. Hierarchical attention networks for document classification[C]//Proceedings of the NAACL,2016: 1480-1489.
[23] JOULINA,GRAVE E,BOJANOWSKI P, et al. Bag of tricks for efficient text classification[C]//Proceedings of the EACL,2017: 427-431.
[24] KUDO T, RICHARDSON J . SentencePiece: A simple and language independent subword tokenizer and detokenizer for neural text processing[C]//Proceedings of the EMNLP, 2018: 66-71.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(4355 KB)
