The economic courts in Egypt play a crucial role in resolving disputes related to economic and commercial matters. These cases cover a wide range of issues that require timely and cost-effective resolutions to ensure the effective management of huge construction projects in Egypt. Machine learning plays a crucial role in the cases handled by Egyptian Economic Courts. It helps in the analysis of large amounts of data, identifying patterns and trends, and making predictions based on historical data. The research delves into the application of various algorithms to identify similar court rulings within Arabic texts, using a dataset comprising 3633 court rulings from the Economic Court in Egypt. The employment of various algorithms has facilitated the efficient selection of comparable court rulings, thus expediting the process of deriving effective conclusions within a minimal timeframe. The evaluation of classifier performance was conducted using metrics such as accuracy, precision, recall, and the F1 score. Notably, the feature extraction done after asking expert judges for the judicial dataset yielded promising results, with an accuracy rate of 95.32%, an F1 score of 95.05%, a recall rate of 95.32%, and a precision rate of 95.32% when employing the Random Forest Classifier algorithm. Conversely, the Support Vector Classification algorithm yielded slightly inferior results. These findings underscore the efficacy of employing the Random Forest Classifier algorithm in conjunction with the count vectorizer feature extraction method for judicial datasets, particularly within the context of court rulings from the Economic Court in Egypt.