<![CDATA[The transition of Vocational High School (SMK) students to higher education requires objective data-driven guidance services. This research aims to develop and implement a web-based classification system using the Naive Bayes algorithm to help map the higher education specialization of students of SMK Al-Ikhlas Losari. A quantitative approach is used with data obtained through questionnaires, interviews, observations, and school supporting data. The operational dataset consisted of 556 students, namely 203 Office Management (MP) students and 353 Motorcycle Engineering (TSM) students. The data was divided into 444 training data (80%) and 112 test data (20%), with categorical attributes transformed through one-hot encoding. The classification model is built using Multinomial Naive Bayes and is integrated into the Streamlit interface. The results of the evaluation showed 100.00% accuracy with a confusion matrix that showed that all test data was classified correctly. The system presents tables, graphs, classification reports, and recommendations in the field of higher education according to the major. MP students are directed to the fields of management, administration, business, accounting, and economics, while TSM students are directed to the fields of mechanical engineering, automotive, industry, and engineering. However, there is a label inconsistency between the implementation of the system and the conclusions of the study. Therefore, perfect accuracy needs to be interpreted as performance against operational targets, while subsequent research needs to explicitly define labels of interest to make the conclusions more valid.]]>
