The process of selecting a major is a critical stage for students because it can influence their motivation and learning outcomes while attending school, especially at Vocational High Schools (SMK). This challenge is becoming more significant with the emergence of many new schools in various cities and districts in Indonesia, especially in DKI Jakarta Province. Prospective students often choose majors not based on personal interests, which can then result in lower grades, especially in productive subjects or certain competencies. To overcome this problem, a major suitability system is needed that can provide recommendations based on student abilities through certain attributes. In this research, a department suitability classification process was carried out using the Naive Bayes and k-Nearest Neighbor methods using data from 238 tenth grade (X) students for the 2023/2024 academic year, which included 9 relevant attributes. The testing process was carried out with a composition of training data and test data in five comparisons, namely 90:10, 80:20, 70:30, 60:40, and 50:50. The research results show that the 80:20 composition provides the best results, with k-Nearest Neighbor achieving recall, accuracy and precision levels of 100%. On the other hand, the Naive Bayes Classifier produces a recall rate of 61%, with an accuracy of 73%. These results indicate that k-Nearest Neighbor is superior in predicting major suitability compared to Naive Bayes under these conditions.