Abstract
The task of prison term prediction is to predict the term of penalty based on the charge and the seriousness of the sentencing plot. Most existing methods focus on improving prediction accuracy but disregard interpretability, which yields unreliable judgment results. To address this problem, we propose an interpretable prison term prediction method. First, the prison term is divided into intervals according to the charge and sentencing plot. Second, we propose a reinforcement learning principle representation model combined with an attention mechanism for regression prediction (PRRP), which extracts phrase-level principles representation as the explanatory basis of prediction results, uses the principle in conjunction with the charge semantics to predict the interval value, and extracts the interval keywords as the sentencing plot. Third, we design a novel multiangle attention mechanism to capture the distinguishing features of cases from different aspects, and a feature fusion network is employed to more effectively stitch multiple pieces of information to learn the feature-enhanced fact representation. Last, the feature-enhanced fact representation is used to predict the prison term. Experimental results on real-work datasets show the interpretability and effectiveness of our method.
Similar content being viewed by others
References
Medvedeva M, Vols M, Wieling M (2020) Using machine learning to predict decisions of the European court of human rights[J]. Artif Intell Law 28(2):237–266
Xiong Z, Shen Q, Wang Y (2018) Paragraph vector representation based on word to vector and CNN learning[J]. CMC-Comput Mater Contin 55:213–227
Dong H, Yang F, Wang X (2020) Multi-label charge predictions leveraging label co-occurrence in imbalanced data scenario[J]. Soft Comput 24:17821–17846
Guo XD, Zhang HL, Ye L, Li S (2021) TenLa: an approach based on controllable tensor decomposition and optimized lasso regression for judgement prediction of legal cases[J]. Appl Intell 51(4):2233–2252
Chao WH, Jiang X, Luo ZC (2019) Interpretable charge prediction for criminal cases with dynamic rationale attention[J]. J Artif Intell Res 66:743–764
Li XC, Kang XJ, Wang CW et al (2020) A neural-network-based model of charge prediction via the judicial interpretation of crimes [J]. IEEE Access 8:101569–101579
Li S, Zhang H, Ye L et al (2019) Prison term prediction on criminal case description with deep learning[J]. Comput Mater Contin 61(3):1217–1231
Yang Z, Yang D, Dyer C et al (2016) Hierarchical attention networks for document classification[C]. In: Proceedings of the 2016 conference of the North American chapter of the association for computational linguistics: human language technologies. San Diego, California, USA, June 12–17, 2016, pp 1480–1489
Xu N, Wang P, Chen L et al Distinguish confusing law articles for legal judgment prediction[C]. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, online, July 5–10, 2020, pp 3086–3095
Cheng X, Bi S, Qi G et al Knowledge-aware method for confusing charge prediction[C]. In: CCF International Conference on Natural Language Processing and Chinese Computing. Zhengzhou, China, October 14–18, 2020, pp 667–679
Zhong H, Guo ZP et al Legal judgment prediction via topological learning[C]. In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. Brussels, Belgium, November 4, 2018, pp 3540–3549
Yang WM, Jia WJ et al (2019) Legal judgment prediction via multi-perspective bi-feedback network[C]. International Joint Conference on Artificial Intelligence. Macao, China, August 10–16, 2019, pp 4085–4091
Ye H, Jiang X, Luo Z et al Interpretable Charge Predictions for Criminal Cases: Learning to Generate Court Views from Fact Descriptions[C]. Proceedings of the 2018 Conference of the north American chapter of the Association for Computational Linguistics: human language technologies, New Orleans, Louisiana, USA, June 1-6, 2018, pp 1854–1864
Zhong H, Wang Y, Tu C et al Iteratively questioning and answering for interpretable legal judgment prediction[C]. Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA, February 7–12, 2020, 34(01), pp 1250–1257
Li L, Zhao LY, Nai PR, Tao XH (2022) Charge prediction modeling with interpretation enhancement driven by double-layer criminal system[J]. World Wide Web-Internet AND Web Information Systems 25(1):384–400
Chen HJ, Cai D et al Charge-Based Prison Term Prediction with Deep Gating Network. [C] Proceedings of the 2019 Conference on empirical methods in natural language processing and the 9th international joint conference on natural language processing, EMNLP-IJCNLP 2019, Hong Kong, China, November 3-7, 2019, pp 6361–6366
Chen YS, Chiang SW, Wu ML (2022) A few-shot transfer learning approach using text-label embedding with legal attributes for law article prediction[J]. Appl Intell 52(3):2884–2902
Ranathunga D, Roughan M, Nguyen H (2022) Verifiable policy-defined networking using Metagraphs[J]. IEEE Trans Dependable Secure Comput 19(1):482–494
Guo S, Zhang X, Du Y et al (2021) Path planning of coastal ships based on optimized DQN reward function[J]. J Mar Sci Eng 9(2):210–233
Zhang T.; Huang M.; Zhao L. Learning structured representation for text classification via reinforcement learning[C]. Proceedings of the AAAI Conference on Artificial Intelligence. New Orleans, LA, USA, February 2–7, 2018, 32(1), pp: 6053–6060
Liu Z, Di XQ, Song W (2021) A sentence-level joint relation classification model based on reinforcement learning [J]. Comput Intell Neurosci
Zhu QN, Zhou XF, Tan JL, Guo L (2021) Knowledge base reasoning with convolutional-based recurrent neural networks[J]. IEEE Trans Knowl Data Eng 33(5):2015–2028
Le ML, Yi DW et al (2022) Deep reinforcement learning in computer vision: a comprehensive survey[J]. IEEE Trans Intell Transp Syst
Paternain S, Bazerque JA, Small A (2021) Ribeiro. A. Stochastic policy gradient ascent in reproducing kernel Hilbert spaces[J]. IEEE Trans Autom Control 66(8):3429–3444
Mikolov T, Chen K, Corrado G et al (2013) Efficient estimation of word representations in vector space[C]. In: 1st International Conference on Learning Representations, ICLR 2013, Scottsdale, Arizona, USA, May 2-4
Devlin J et al (2019) Bert: Pre-training of deep bidirectional transformers for language understanding[C]. In: Proceedings of the 2019 Conference of the north American chapter of the Association for Computational Linguistics: human language technologies, NAACL-HLT 2019, Minneapolis, MN, USA, June 2-7, pp 4171–4186
Yan CG, Hao YM et al (2022) Task-adaptive attention for image captioning[J]. IEEE Trans Circuits Syst Video Technol 32(1):43–51
Mee A, Homapour E, Chiclana F, Engel O (2021) Sentiment analysis using TF-IDF weighting of UK MPs' tweets on Brexit[J]. Knowl-Based Syst 228:107238
Zied HY, Sieg A, Deleris LA (2019) Towards Unsupervised Text Classification Leveraging Experts and Word Embeddings[C]. In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Florence, Italy, June 28–August 2, pp 371–379
Xiao SN, Li YM, Ye YA et al (2020) Hierarchical temporal fusion of multi-grained attention features for video question answering[J]. Neural Process Lett 52(2):993–1003
Sun C, Kong F (2018) The awarding ceremony of "China legal research cup" judicial artificial intelligence challenge (Cail 2018) was held [J]. Chin J inf 32(12):56
Sun MS, Chen XX, Zhang KX et al (2016) Thulac: An efficient lexical analyzer for chinese. Technical Report
Diederik K, Jimmy B (2015) Adam: A method for stochastic optimization. In: 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9
Nitish S, Geoffrey EH, Alex K, Ilya S, Ruslan S (2014) Dropout: a simple way to prevent neural networks from overfitting[J]. J Mach Learn Res 15(1):1929–1958
Cheng K, Lu ZZ (2021) Active learning Bayesian support vector regression model for global approximation[J]. Inf Sci 544:549–563
Funding
This work is supported by the National Key R&D Program of China (Grant No. 2018YFB1601502) and the LiaoNing Revitalization Talents Program (Grant No. XLYC1902071).
Author information
Authors and Affiliations
Corresponding authors
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wang, P., Zhang, X., Yu, H. et al. Interpretable prison term prediction with reinforce learning and attention. Appl Intell 53, 1306–1323 (2023). https://doi.org/10.1007/s10489-022-03675-1
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-022-03675-1