Recidivism Prediction: A Novel Machine Learning-based Imbalanced Learning Method Combined with the Differential Equation Algorithm

Fang  Chen; Huihui  Hou

doi:10.6180/jase.202603_29(4).0002

Recidivism Prediction: A Novel Machine Learning-based Imbalanced Learning Method Combined with the Differential Equation Algorithm

Research Categories

Fang Chen¹ and Huihui Hou²This email address is being protected from spambots. You need JavaScript enabled to view it.

¹School of General Education, Luzhou Vocational and Technical College, Luzhou 646000, China

²School of General Education and Humanities Education, Shijiazhuang Vocational College of Finance and Economics, Shijiazhuang 050000, China

Received: January 27, 2025
Accepted: June 22, 2025
Publication Date: July 17, 2025

Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: ||https://doi.org/10.6180/jase.202603_29(4).0002

We propose a new method to improve recidivism prediction by addressing hyperparameter tuning and class imbalance. Our model, the Imbalanced AUC-Proximal Support Vector Machine (ImAUC-PSVM), incorporates AUC optimization to enhance standard PSVM performance while reducing the need for extensive parameter tuning. The model opts for direct AUC optimization rather than sticking to the usual loss functions like hinge loss or cross-entropy. This choice kind of gives it a leg up in ranking performance and makes it more resilient when classes are imbalanced—a common pain point, really. While adapting to changing prediction settings, it retains core PSVM elements and uses the Differential Equation (DE) approach, which aids hyperparameter tuning in complex environments. The DE algorithm kind of fine-tunes performance more effectively than the usual grid or random search methods. It explores a wider range of possibilities and manages to steer clear of getting stuck in local optima. Because of that, you can be a bit more confident that it’ll actually find well-tuned hyperparameters in the end. We assess the model using data from Chinese correctional centers and North Carolina prison records. On Chinese and American samples, the model attained predictive scores of 88.125 and 90.809. Compared to the strongest baseline, the proposed method improved AUC by 7.1% on the Chinese and 6.7% on the American datasets. The results show strong performance on imbalanced data and confirm the method’s reliability for real-world recidivism prediction. It improves criminal justice modeling and provides a flexible framework for complex classification tasks in other domains.

Keywords: Recidivism Prediction; Imbalanced Class Distribution; Proximal Support Vector Machine; Hyperparameter

[1] D.Yukhnenko, L. Farouki, and S. Fazel, (2023) “Crim inal recidivism rates globally: A 6-year systematic review update" Journal of criminal justice 88: 102115. DOI: 10.1016/j.jcrimjus.2023.102115.
[2] C. Penate. “Rebuilding Risk Assessment Tools and Evaluating Their Effect on Recidivism in Illinois". (mathesis). Loyola University Chicago, 2023.
[3] O. Kovalchuk, M. Karpinski, S. Banakh, M. Kasianchuk, R. Shevchuk, and N. Zagorodna, (2023) “Prediction machine learning models on propensity convicts to criminal recidivism" Information 14: 161. DOI: 10.3390/info14030161.
[4] M. Cavus, M. N. Benli, U. Altuntas, M. Sari, H. Ayan, and Y. F. Ugurluoglu, (2025) “Transparent and bias resilient AI framework for recidivism prediction using deep learning and clustering techniques in criminal jus tice" Applied Soft Computing: 113160. DOI: 10.1016/j.asoc.2025.113160.
[5] D. Mu, S. Zhang, T. Zhu, Y. Zhou, and W. Zhang, (2024) “Prediction of recidivism and detection of risk fac tors under different time windows using machine learning techniques" Social Science Computer Review 42(6): 1379–1402. DOI: 10.1177/08944393241226607.
[6] K. M. A. Haque, S. Rao, B. Nasirian, V. C. P. Chen, C. Kan, and Y. Zhou. “Classification for Predicting Recidivism: Challenges and Scopes”. In: IISE Annual Conference. Proceedings. Institute of Industrial and Systems Engineers (IISE), 2024, 1–6. DOI: 10.21872/2024IISE_6000.
[7] S. Sultana, I. Jahir, M. Suukyi, M. M. R. Nabil, A. Waziha, and S. Momen. “Advancing Recidivism Prediction for Male Juvenile Offenders: A Machine Learning Approach Applied to Prisoners in Hunan Province”. In: Springer, 2023, 184–201. DOI: 10.1007/978-3-031-54820-8_16.
[8] M. L. Angondji, M. Chitauro, and F. B. Shava. “Evaluating Machine Learning Models Best Fit for Crime Prediction in Windhoek”. In: 2024 International Conference on Artificial Intelligence, Big Data, Computing and Data Communication Systems (icABCD). IEEE, 2024, 1–6. DOI: 10.1109/icABCD62167.2024.10645286.
[9] S. Etzler, F. D. Schönbrodt, F. Pargent, R. Eher, and M. Rettenberger, (2024) “Machine learning and risk assessment: Random forest does not outperform logistic regression in the prediction of sexual recidivism" Assessment 31: 460–481. DOI: 10.1177/10731911231164624.
[10] M. M. Motevalli, M. K. Sohrabi, and F. Yaghmaee, (2025) “Aspect-based sentiment analysis: A dual-task learning architecture using imbalanced maximized-area under the curve proximate support vector machine and reinforcement learning" Information Sciences 689: 121449. DOI: 10.1016/j.ins.2024.121449.
[11] O. Kovalchuk, R. Shevchuk, L. Babala, and M. Kasianchuk, (2024) “Support vector machine to criminal recidivism prediction" International Journal of Electronics and Telecommunications: 691–967. DOI: 10.24425/ijet.2024.149598.
[12] G. Wang, S. W. H. Kwok, D. Axford, M. Yousufuddin, and F. Sohel, (2023) “An AUC-maximizing classifier for skewed and partially labeled data with an application in clinical prediction modeling" Knowledge-based systems 278: 110831. DOI: 10.1016/j.knosys.2023.110831.
[13] P. Zareeihemat, S. Mohamadi, J. Valipour, and S. V. Moravvej, (2025) “Forecasting stock market volatility using housing market indicators: A reinforcement learning based feature selection approach" IEEE Access: DOI: 10.1109/ACCESS.2025.3554224.
[14] Y. Zhao, W. Zhang, and X. Liu, (2024) “Grid search with a weighted error function: Hyper-parameter optimization for financial time series forecasting" Applied Soft Computing 154: 111362. DOI: 10.1016/j.asoc.2024.111362.
[15] A. Mehdary, A. Chehri, A. Jakimi, and R. Saadane, (2024) “Hyperparameter optimization with genetic algorithms and XGBoost: a step forward in smart grid fraud detection" sensors 24: 1230. DOI: 10.3390/s24041230.
[16] S. V. Moravvej, S. J. Mousavirad, D. Oliva, G. Schae fer, and Z. Sobhaninia. “An improved de algorithm to optimise the learning process of a bert-based pla giarism detection model”. In: 2022 IEEE congress on evolutionary computation (CEC). IEEE, 2022, 1–7. DOI: 10.1109/CEC55065.2022.9870280.
[17] A. A. Aghaei and K. Parand, (2025) “Hyperparameter optimization of orthogonal functions in the numerical solution of differential equations" Mathematical Methods in the Applied Sciences 48: 605–625. DOI: 10.1002/mma.10346.
[18] G. Maheswari, S. Felixa, A. J. Vidhya, M. Sambath, and K. Remya. “Forecasting of Crime Hotspots and Analysis using Machine Learning”. In: 2024 Interna tional Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS). IEEE, 2024, 1–5. DOI: 10.1109/ADICS58448.2024.10533569.
[19] A. Dubey, N. Venu, D. Bisen, P. Garg, K. Srinivas, and H. Arya. “A Clustering-Based Method for Hotspot Recognition in Crime Forecasting”. In: 2024 IEEE 8th International Conference on Information and Communication Technology (CICT). IEEE, 2024, 1–6. DOI: 10.1109/CICT64037.2024.10899713.
[20] R. Ahmad, A. Nawaz, G. Mustafa, T. Ali, M. Tlija, M. A. El-Meligy, and Z. Ahmed, (2024) “CHART: Intelligent Crime Hotspot Detection and Real-Time Track ing Using Machine Learning." Computers, Materials &Continua 81: DOI: 10.32604/cmc.2024.056971.
[21] A. Morin-Martel, (2024) “Machine learning in bail decisions and judges’ trustworthiness" Ai & Society 39: 2033–2044. DOI: 10.1007/s00146-023-01673-6.
[22] L. Yang, J. Guofan, Z. Yixin, W. Qianze, Z. Jian, R. Alizadehsani, and P. Pławiak, (2024) “A reinforcement learning approach combined with scope loss function for crime prediction on Twitter" IEEE Access: DOI: 10.1109/ACCESS.2024.3473296.
[23] Z. Yixin, L. Yang, J. Guofan, Z. Jian, J. Yang, R. Al izadehsani, R. Tadeusiewicz, and P. Pławiak, (2025) “An off-policy deep reinforcement learning-based active learning for crime scene investigation image classifica tion" InformationSciences710:122074. DOI:10.1016/j.ins.2025.122074.
[24] A. Sedik, H. Kolivand, and M. Albeedan, (2024) “An efficient image classification and segmentation method for crime investigation applications" Multimedia Tools and Applications: 1–25. DOI: 10.1007/s11042-024 19773-w.
[25] S. Sandhya, A. Balasundaram, and A. Shaik, (2023) “Deep Learning Based Face Detection and Identification of Criminal Suspects." Computers, Materials & Continua 74: DOI: http: //dx.doi.org/10.32604/cmc.2023.032715.
[26] J. H. Jhee, M. J. Kim, M. Park, J. Yeon, and H. Shin, (2023) “Fast Prediction for Criminal Suspects through Neighbor Mutual Information-Based Latent Network" International Journal of Intelligent Systems 2023: 9922162. DOI: 10.1155/2023/9922162.
[27] M.Barash, D. McNevin, V. Fedorenko, and P. Giverts, (2024) “Machine learning applications in forensic DNA profiling: A critical review" Forensic Science International: Genetics 69: 102994. DOI: 10.1016/j.fsigen.2023.102994.
[28] Y. Lee, B. Bradford, and K. Posch, (2024) “The effectiveness of big data-driven predictive policing: systematic review" Justice Evaluation Journal 7: 127–160. DOI: 10.1080/24751979.2024.2371781.
[29] E. Ingram, F. Gursoy, and I. A. Kakadiaris. “Accu racy, fairness, and interpretability of machine learning criminal recidivism models”. In: 2022 IEEE/ACM International Conference on Big Data Computing, Appli cations and Technologies (BDCAT). IEEE, 2022, 233–241. DOI: 10.1109/BDCAT56447.2022.00040.
[30] Y. Ma, K. Nakamura, E.-J. Lee, and S. S. Bhat tacharyya. “Eadtc: An approach to interpretable and accurate crime prediction”. In: 2022 IEEE International Conference on Systems, Man, and Cybernetics (SMC). IEEE, 2022, 170–177. DOI: 10.1109/SMC53654.2022.9945130.
[31] O. Kovalchuk, S. Banakh, M. Masonkova, V. Burdin, O. Zaverukha, and R. Ivanytskyy. “A Scoring Model for Support Decision Making in Criminal Justice”. In: 2022 12th International Conference on Advanced Com puter Information Technologies (ACIT). IEEE, 2022, 116 120. DOI: 10.1109/ACIT54803.2022.9913182.
[32] T. Babad and S. A. Chun. “Enhancing accuracy and explainability of recidivism prediction models”. In: The International FLAIRS Conference Proceedings. 36. 2023. DOI: 10.32473/flairs.36.133382.
[33] Y. Lee, S. O, and J. E. Eck, (2025) “Improving Recidivism Forecasting With a Relaxed Naïve Bayes Classifier" Crime & Delinquency 71: 89–117. DOI: 10.1177/00111287231186093.
[34] A. Singh and S. Mohapatra. “Automatic Criminal Recidivism Risk Estimation in Recidivist Using Classification and Ensemble Techniques”. In: Springer, 2023, 279–289. DOI: 10.1007/978-981-19-6525-8_22.
[35] C. Wang, B. Han, B. Patel, and C. Rudin, (2023) “In pursuit of interpretable, fair and accurate machine learning for criminal recidivism prediction" Journal of Quantitative Criminology 39: 519–581. DOI: 10.1007/s10940-022-09545-w.
[36] S. Etzler, F. D. Schönbrodt, F. Pargent, R. Eher, and M. Rettenberger, (2024) “Machine learning and risk assessment: Random forest does not outperform logistic regression in the prediction of sexual recidivism" Assessment 31: 460–481. DOI: 10.1177/10731911231164624.
[37] S. Bentos, E. Bailis, F. Oikonomou, S. Spirou, E. Mavrikos, S. Chatzistamatis, K. Kotis, and G. E. Tsek ouras. “Evaluation of Fairness in Machine Learning Based Recidivism Predictions: the Case of Greek Fe male Prison System”. In: 2024 4th International Conference on Innovative Research in Applied Science, Engi neering and Technology (IRASET). IEEE, 2024, 1–8. DOI: 10.1109/IRASET60544.2024.10548202.
[38] J. K. Athota, K. K. Parimi, M. K. Teja, M. A. Bhavani, and M. M. Y. Devi. “Fairness in Predicting Recidi vism Score”. In: International Conference on Smart Data Intelligence. Springer, 2024, 239–253. DOI: 10.1007/978-981-97-3191-6_18.
[39] Y. Lee and S. O, (2025) “Redefining Recidivism Prediction: The Impact of Race and Geographic Lo cation in Machine Learning Models" Crime & Delinquency: 00111287251316515. DOI: 10.1177/ 00111287251316515.
[40] G. M. Circo and A. P. Wheeler, (2025) “An open source replication of a winning recidivism prediction model" International Journal of Offender Therapy and Comparative Criminology 69: 438–453. DOI: 10.1177/0306624X221133004.
[41] Y. Lee, S. O, and J. E. Eck, (2025) “Improving Recidivism Forecasting With a Relaxed Naïve Bayes Classifier" Crime & Delinquency 71: 89–117. DOI: 10.1177/00111287231186093.
[42] M. Saberian, V. Samadi, and I. Popescu, (2024) “Probabilistic Hierarchical Interpolation and Interpretable Configuration for Flood Prediction" Hydrology and Earth System Sciences Discussions 2024: 1–41. DOI: 10. 5194/hess-2024-261.
[43] H. Najafi, K. Savoji, M. Mirzaeibonehkhater, S. V. Moravvej, R. Alizadehsani, and S. Pedrammehr, (2024) “A Novel Method for 3D Lung Tumor Reconstruc tion Using Generative Models" Diagnostics 14: 2604. DOI: 10.3390/diagnostics14222604.
[44] M. Merikhipour, S. Khanmohammadidoustani, and M. Abbasi, (2025) “Transportation mode detection through spatial attention-based transductive long shortterm memory and off-policy feature selection" Expert Systems with Applications 267: 126196. DOI: 10.1016/j.eswa.2024.126196.
[45] A. Kokhahi and D. Li, (2024) “GLHAD: A group lasso based hybrid attack detection and localization framework for multistage manufacturing systems" Journal of Computing and Information Science in Engineering 24: DOI: 10.1115/1.4063862.
[46] S. M. S. Mohammadabadi, M. Entezami, A. K. Moghaddam, M. Orangian, and S. Nejadshamsi, (2024) “Generative artificial intelligence for distributed learning to enhance smart grid communication" Interna tional Journal of Intelligent Networks 5: 267–274. DOI: 10.1016/j.ijin.2024.05.007.
[47] J. Xiang, (2025) “The duration and high-hazard factors of food crimes in China: an empirical study with automatic text analytics" Crime, Law and Social Change 83: 1 20. DOI: 10.1007/s10611-025-10214-9.
[48] A. N. Tamar, M. Karbasi, M. R. Khani, T. Hamzehlouyan, and B. Shokri, (2023) “Response sur face methodology (RSM) for optimizing ozone-assisted process parameters for formaldehyde removal" Journal of Environmental Health Science and Engineering 21: 475–484. DOI: 10.1007/s40201-023-00873-y.
[49] A. Shafiee, H. R. Moghadam, M. Merikhipour, and J. Lin. “Analyzing Post-Pandemic Remote Work Accessibility for Equity through Machine Learning Analysis”. In: International Conference on Transportation and Development 2024. 2024, 453–462. DOI: 10.1061/9780784485521.041.
[50] S. Thundiyil, S. S. Shalamzari, J. Picone, and S. McKenzie. “Transformers for Modeling Long-Term Dependencies in Time Series Data: A Review”. In: 2023 IEEE Signal Processing in Medicine and Biology Symposium (SPMB). IEEE, 2023, 1–5. DOI: 10.1109/SPMB59478.2023.10372632.
[51] B. Thai, S. McNicholas, S. S. Shalamzari, P. Meng, and J. Picone. “Towards a More Extensible Machine Learning Demonstration Tool”. In: 2023 IEEE Signal Processing in Medicine and Biology Symposium (SPMB). IEEE, 2023, 1–5. DOI: 10.1109/SPMB59478.2023.10372731.
[52] M. A. L. Khaniki, A. Golkarieh, H. Nouri, and M. Manthouri, (2024) “Enhanced fault detection and cause identification using integrated attention mechanism" arXiv preprint arXiv:2408.00033: 1–36. DOI: 10.48550/arXiv.2408.00033.
[53] M. Entezami, S. Basirat, B. Moghaddami, D. Bazman deh, and D. Charkhian, (2025) “Examining the Im portance of AI-Based Criteria in the Development of the Digital Economy: A Multi-Criteria Decision-Making Approach" Journal of Soft Computing and Decision Analytics 3: 72–95. DOI: 10.31181/jscda31202555.