A systematic literature review of machine learning methods in predicting court decisions
<span>Envisaging legal cases’ outcomes can assist the judicial decision-making process. Prediction is possible in various cases, such as predicting the outcome of construction litigation, crime-related cases, parental rights, worker types, divorces, and tax law. The machine learning methods can function as support decision tools in the legal system with artificial intelligence’s advancement. This study aimed to impart a systematic literature review (SLR) of studies concerning the prediction of court decisions via machine learning methods. The review determines and analyses the machine learning methods used in predicting court decisions. This review utilised RepOrting Standards for Systematic Evidence Syntheses (ROSES) publication standard. Subsequently, 22 relevant studies that most commonly predicted the judgement results involving binary classification were chosen from significant databases: Scopus and Web of Sciences. According to the SLR’s outcomes, various machine learning m
<span>Envisaging legal cases’ outcomes can assist the judicial decision-making process. Prediction is possible in various cases, such as predicting the outcome of construction litigation, crime-related cases, parental rights, worker types, divorces, and tax law. The machine learning methods can function as support decision tools in the legal system with artificial intelligence’s advancement. This study aimed to impart a systematic literature review (SLR) of studies concerning the prediction of court decisions via machine learning methods. The review determines and analyses the machine learning methods used in predicting court decisions. This review utilised RepOrting Standards for Systematic Evidence Syntheses (ROSES) publication standard. Subsequently, 22 relevant studies that most commonly predicted the judgement results involving binary classification were chosen from significant databases: Scopus and Web of Sciences. According to the SLR’s outcomes, various machine learning methods can be used in predicting court decisions. Additionally, the performance is acceptable since most methods achieved more than 70% accuracy. Nevertheless, improvements can be made on the types of judicial decisions predicted using the existing machine learning methods.</span>
Executive Summary
The article presents a systematic literature review (SLR) of machine learning methods used to predict court decisions. It identifies 22 relevant studies from significant databases, focusing on binary classification tasks. The review finds that various machine learning methods achieve acceptable accuracy (over 70%) in predicting judicial outcomes. However, it notes that improvements are needed in the types of judicial decisions that can be predicted. The study adheres to the ROSES publication standard, ensuring a rigorous and systematic approach.
Key Points
- ▸ Machine learning methods can predict court decisions with acceptable accuracy.
- ▸ Binary classification is the most common prediction task in the reviewed studies.
- ▸ Improvements are needed in the diversity of judicial decisions that can be predicted.
Merits
Comprehensive Review
The study provides a thorough and systematic review of existing literature, adhering to the ROSES standard, which enhances the credibility and reliability of the findings.
High Accuracy
The review demonstrates that machine learning methods can achieve high accuracy in predicting court decisions, making them valuable tools for judicial decision-making.
Clear Methodology
The methodology is clearly outlined, including the use of significant databases and a systematic approach to selecting relevant studies.
Demerits
Limited Scope
The review focuses primarily on binary classification tasks, which may not capture the full complexity of judicial decision-making processes.
Generalizability
The findings may not be generalizable to all types of judicial decisions, as the review is limited to specific databases and a select number of studies.
Data Quality
The quality and representativeness of the data used in the reviewed studies are not extensively discussed, which could impact the validity of the findings.
Expert Commentary
The systematic literature review conducted in this article provides a valuable contribution to the field of legal technology by demonstrating the potential of machine learning methods in predicting court decisions. The finding that most methods achieve over 70% accuracy is significant, as it indicates that these tools can be reliable aids in judicial decision-making. However, the focus on binary classification tasks is a limitation, as it does not fully capture the nuanced and complex nature of many legal decisions. Future research should explore the application of machine learning to a broader range of judicial decisions and consider the ethical implications of integrating these technologies into the legal system. Additionally, the quality and representativeness of the data used in these studies should be a focus of future research to ensure the validity and generalizability of the findings. Overall, this study highlights the potential of machine learning in the legal field while also identifying areas for further exploration and improvement.
Recommendations
- ✓ Future studies should expand the scope of machine learning applications to include more complex and diverse types of judicial decisions.
- ✓ Researchers should pay closer attention to the quality and representativeness of the data used in machine learning models to ensure the validity of the findings.