Journal of Engineering, Project, and Production Management, 2026, 16(6), 2026-0055
Intelligent Classification of Chinese Books in University Libraries Based on BERT and PLM-LCN
Librarian, Library, Qufu Normal University, Qufu, 273100, China, E-mail: BoHongbh@outlook.com
Project Management
Received May 11, 2026; revised June 25, 2026; accepted August 3, 2026
Available online August 15, 2026
Abstract: With the expansion of the collection scale of university libraries, the intelligent classification of Chinese books has become the key to improving efficiency. However, traditional classification methods have significant shortcomings in handling the complexity of Chinese semantics and interdisciplinary issues. Therefore, this study proposes an intelligent classification model based on the bidirectional encoder representations from pre-trained models and the fusion of neural network features. Firstly, this study utilizes a bidirectional encoder to represent the text's global semantic features in the pre-trained model. Secondly, a text classification model based on recurrent convolutional neural networks is utilized to capture local contextual information. Next, a multi-head attention mechanism is introduced to enhance and optimize text features, and finally, an efficient classification is achieved through a Softmax classifier. The experiment showed that the model achieved classification accuracy of 96.8% and 93.3% on the Chinese Library Classification Dataset and the National Science and Technology Library Literature Center Chinese Book Dataset, respectively. In the interdisciplinary book classification test, the model's AUC reached 83.7%, significantly improving retrieval accuracy and user satisfaction in real-world library systems. Research has shown that the proposed model can effectively address semantic understanding and feature fusion in Chinese book classification, providing an efficient and accurate solution for intelligent library management.
Keywords: Bidirectional encoder representation transformers (BERT); pre-trained language model and long short-term memory-convolutional neural network (PLM-LCN); multi-head attention (MHA); Chinese book classification; interdisciplinary semantic understanding.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Hong, B. (2026). Intelligent Classification of Chinese Books in University Libraries Based on BERT and PLM-LCN. Journal of Engineering, Project, and Production Management, 16(6), 2026-0055.
