A CASE BASED SYSTEM FOR SUPPORTING COURT DECISION MAKING: IN THE CASE OF ETHIOPIA

Abstract

Case based reasoning is a computer technique that combines the philosophy of knowledge-based support with the simulation of human reasoning using experience. It focuses on using explicit and documented experience to solve novel problems. Real-world decision problems are complex, unstructured, and uncertain with potential risks associated with inaccurate decisions. This may necessitate applying the previous solution to current constraints. Therefore, case based problem-solving can be more effective than starting from scratch when solving a new problem. However, Ethiopia's legal system faces challenges in establishing legal precedents, resulting in uncertainty and lack of legal certainty. Different judgments were made in similar cases, leading to ambiguity and a lack of legal certainty, resources, and overburdened caseloads, resulting in a delay in the delivery of justice. Case based technology, which draws on prior and analogous cases to address novel situations, is particularly helpful in decision-making in cassation systems, especially in managing these processes and minimizing delays. Case based systems have facilitated a more consistent and uniform approach to decision-making, improving the decision-making process and making legal information more accessible to practitioners. In addition, it aids in structuring and retrieving legal information. In this study, a design science approach was employed to design and develop a case-based system to address court decision making. Design science research addresses and motivates problem awareness. This study involves artifact development using FreeCBR and a demonstration of solutions with relevant attributes extracted through document analysis. Statistical analysis techniques and user-acceptance tests were used to evaluate the performance of the proposed system. This study achieved a recall of 57.1% and precision of 87.5%. Additionally, the evaluation of the prototype by experts showed 90% user acceptance. This study achieved promising results with room for improvement to enhance its capabilities and increase its sophistication in handling court problems.