Application of Case Based Reasoning in Legal Knowledge Based System: A Prototype on Children Criminal Cases in Ethiopia

No Thumbnail Available



Journal Title

Journal ISSN

Volume Title


Addis Ababa University


The target of Artificial Intelligence is to make computer behave intelligent. This is causing the development of many subfields; among which Case -Based Reasoning (CBR) is one of the hottest subfield of AI. Case-Based Reasoning (CBR) proposes the reuse of past experiences and cases to solve new problems. Law is an excellent domain for studying case-based reasoning that enables the reuse of the information contained in the cases to solve problems from diverse areas. The goal of this thesis is to design prototype, case based reasoning in legal knowledge based system that advises in the investigation of children criminal cases. This paper describes Judicial Application, a case-based system that is able to adjust its retrieval and adaptation metrics, in addition to storing cases. Standard case representation to the original knowledge source (legal cases) has been used to store legal cases. The prototype is developed using the Object-Oriented framework jCOLIBRI1.0. The CBR model has been effectively used to retrieve analogues previous cases in reaching a potential solution for decision making and also a means to adopt the current case of a problem for the future use. To enhance the effectiveness of the legal argument, local similarity matching and weight of the attributes are adopted to compare the equivalent values of a problem of a case and source cases. The performance of the system is evaluated using statistical analysis which results with average precision value 82.75% and user satisfaction which shows the suitability of CBR for designing legal knowledge based system. To improve the performance of the system the researcher mainly recommends the need for upgrading the capacity of the case using integrated domain knowledge to have a precise and up-to-date decision processing.



Based Reasoning in Legal Knowledge Based System