Cataloging is an essential part of the mission of the libraries, as the collection serves as a carrier of knowledge, so the users can effectively retrieve and utilize this knowledge. Automatic classification technologies have been introduced to the library technical services to enhance efficiency and improve inconsistency in cataloging. To address the actual cataloging needs and problems in libraries, this study used 620,217 titles from the National Taiwan Normal University Library as experiment datasets and trained with the BERT distilbert-base-multilingual-cased model on different combinations of call number, titles, and authors’ data to make multiple subject cataloging predictions in both Chinese and English languages. The prediction results were evaluated in terms using indicators such as accuracy, precision, and recall. The results showed that the combination of the number and title had the best accuracy in predicting subject cataloging, indicating that the term of occurrences of the theme had a significant influence on the prediction performance. The study also found two key factors that can improve the prediction, the one is a number the glossary terms as the training data sets, and the similarity of the meaning represented by the training data and the glossary terms.