<![CDATA[The increasing number of traffic violations in Pagar Alam City has led to a yearly rise in ticketing case data at the Pagar Alam District Attorney’s Office. This accumulation of data creates difficulties in effective data management and hinders the extraction of meaningful insights. The current classification process for ticketing cases remains limited in its accuracy and efficiency, making it difficult to identify patterns or trends. This study aims to address this issue by developing a classification model for traffic ticket cases using data mining techniques, specifically the C4.5 algorithm. The model classifies cases based on attributes such as the relevant article of law, type of vehicle, evidence submitted, and the fine imposed. The CRISP-DM framework is used to guide the process through six phases: business understanding, data understanding, data preparation, modeling, evaluation, and deployment. RapidMiner is used as the primary tool for data processing, and the model is evaluated using the X-Cross Validation technique. The results show that the C4.5 algorithm achieves a high classification accuracy of 99.75%. The “Article” attribute emerged as the most influential factor with the highest gain ratio value. These findings can support law enforcement and policymakers in identifying the most frequent violations and developing more targeted strategies to improve traffic law enforcement and public safety.]]>
Copyrights © 2025
