Application of machine learning algorithms for predicting outcomes of accident cases in Moroccan courts

Abstract

Due to the large number of legal cases, the processing of them by the courts is generally very slow. Among these cases, we find accidents cases, which require a great speed of judgment to compensate the victims of those accidents. To this end, we thought of exploiting the possibilities offered by machine learning in order to simulate the work of judges and contribute to speeding up the time of decision. Further, we applied different machine learning algorithms, such as linear regression, decision trees, and random forests. According to the results achieved, the Random Forest is the most perfect model for with the utmost accuracy about 91.05%.