In the legal context, court rulings can exhibit variability due to differing interpretations of the law by judges. This paper presents a software solution for recommending court rulings, which contributes to the standardization of judicial practice through the automated extraction of information from previous rulings and the identification of similar legal cases. Using regular expressions, key attributes are extracted from textual documents, and the k-Nearest Neighbors algorithm is employed to find rulings similar to the current case. The system is trained on a dataset containing first-instance court rulings in cases of forest theft and forest devastation in Montenegro. A total of 34 court rulings are analyzed, with 17 out of 19 attributes successfully extracted with a high degree of accuracy. The recommended rulings are manually verified to determine whether they are meaningful and relevant to the current case. The proposed software solution is integrated into a software system for reviewing and recommending court rulings, enabling judges to quickly access relevant information about individual rulings without the need to read the entire text or analyze individual documents. Additionally, by providing insight into previous decisions that are most similar to the current case, the system enhances consistency and maintains uniformity in legal practice.