MTCT: Using AI and machine learning to help Police Forces automate Out of Court Order administration, streamline the restorative justice process and help reduce the court backlog.

MTCT：利用人工智能和机器学习帮助警察部队实现庭外命令管理自动化、简化恢复性司法流程并帮助减少法庭积压案件。

• 批准号: 10042299
• 金额: $ 44.56万
• 依托单位: MAKE TIME COUNT TODAY LTD
• 依托单位国家: 英国
• 项目类别: Collaborative R&D
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=10042299
• 关键词: MTCT; Using; AI; machine; learning

With criminal justice cuts, legal aid issues, barristers' strikes and COVID, the UK has seen court backlogs spiral to over 400,000 cases and left victims waiting years for justice. Majority of cases will be for low-level offences from repeat offenders,* Only 10% of cases result in custodial sentences.* 90% of all cases brought against people who already have one conviction or more,Alternatives are drastically needed to reduce court burden and address reoffending.An Out of Court Disposal (OOCD) is a method of resolving low-level crime when an offender is known and admits offence. A driving principle for OOCDs is to reduce re-offending by consultation with the victim and enabling diversion and restorative justice (RJ). UK's largest OOCD trial recently completed in London. Result show that users processed via OOCD:* 30% less likely to reoffend* 58% less for those aged 18-21* 34% reduction in offending for users from BAME groupsThis would have transformative results if applied at-scale. However the OOCD process is time-consuming and there's no technology to support Officers implementing OOCDs, nor assess effectiveness as a recidivism deterrent. This puts significant administrative and moral pressure on Officers, resulting in a high-level of subjectivity and huge regional and ethnic disparities in OOCD use.Working with the London MET police and Cambridge Centre for Evidence Based Policing, Make Time Count Today intend to develop a highly innovative AI data analytics platform to help police forces:1. Review current court backlog and determine which cases could be dealt with as an OOCD;2. Predict effectiveness of OOCD process as a mechanism to prevent reoffending vs court,3. Streamline and automate the administrative pinch-points in the OOCD process;4. Embed RJ frameworks into OOCD system to improve outcomes;And ultimately:* Reduce number of low-level offences processed through courts* Reduce re-offending through use of OOCD and RJ.

随着刑事司法削减，法律的援助问题，大律师罢工和新冠肺炎，英国法院积压的案件螺旋上升至40多万起，受害者等待多年的正义。大多数案件将是惯犯的低级罪行，* 只有10%的案件导致监禁判决。90%的案件是针对已经有一次或多次定罪的人提起的，迫切需要替代方案来减轻法庭负担并解决重新犯罪。庭外处置（OOCD）是一种解决低级别犯罪的方法，当罪犯是已知的并承认犯罪时。OOCD的一个驱动原则是通过与受害者协商减少重新犯罪，并实现分流和恢复性司法。英国最大的OOCD试验最近在伦敦完成。结果显示，通过OOCD处理的用户：* 18-21岁的用户再次犯罪的可能性降低30%* 58%* 来自BAME组的用户犯罪减少34%如果大规模应用，这将产生变革性的结果。然而，OOCD过程是耗时的，没有技术支持官员实施OOCD，也没有评估作为累犯威慑的有效性。这给警官们带来了巨大的行政和道德压力，导致OOCD的使用存在高度的主观性和巨大的地区和种族差异。Make Time Count Today与伦敦警察局和剑桥循证警务中心合作，打算开发一个高度创新的人工智能数据分析平台，以帮助警察部队：1.审查当前法院积压的案件，并确定哪些案件可以作为OOCD处理;2.预测OOCD程序作为预防再次犯罪的机制对法院的有效性。简化和自动化OOCD过程中的管理关键点;4.将RJ框架嵌入OOCD系统，以改善结果;最终：* 减少通过法院处理的低级犯罪数量 * 通过使用OOCD和RJ减少重新犯罪。