T The increasing number of scientific publications by lecturers in the field of informatics engineering has led to difficulties in identifying dominant research trends and patterns, particularly when relying solely on manual exploration. universitas trisakti has many informatics lecturers who actively publish scientific works across various research areas; however, to date, there is no automated system capable of classifying lecturers’ research based on the titles of scientific articles. in fact, article titles are considered capable of representing research topics and directions in a concise and specific manner. this study aims to classify the titles of scientific articles authored by informatics engineering lecturers at universitas trisakti into information technology research fields using the multinomial naïve bayes method. the research data consist of scientific article titles obtained from google scholar and the science and technology index (sinta) covering the period from 2012 to 2023. the research stages include data collection, labeling based on scientific fields and subfields, text preprocessing (demojization, case folding, url filtering, tokenization, and stopword removal), feature extraction using tf-idf, as well as classification and model evaluation. the results show that after data balancing was performed, the classification model was focused on four main categories, namely information technology engineering, information technology, web software applications, and information systems and technology. the developed multinomial naïve bayes model was able to classify lecturers’ research titles with good performance, as indicated by satisfactory values of accuracy, precision, recall, and f1-score. the classification results can assist universities, lecturers, and students in understanding lecturers’ research tendencies and facilitate the determination of research directions in the field of informatics engineering.