Article Details

Abstract

In library service work, when facing local characteristic literature with a small data volume, library managers need to spend a great deal of time and effort manually organizing such local characteristic literature. In order to achieve automated pre-classification of characteristic literature, this paper proposes the CSGB model, which is an automated classification model for literature with a small data volume. Taking the characteristic literature dataset of Guizhou Province as the experimental object, the model conducts pre-training through GloVe and BERT, fuses the generated vectors, extracts and represents features through TextCNN, and classifies characteristic literature of different data scales. Experimental results indicate that the accuracy of the model with fused word vectors is at least 4% higher than that of the benchmark model.

Keywords

local characteristic literature; text classification; text vectorization

References

1. [1] ZHU B，PAN W. Chinese Text Classification Method Based on Sentence Information Enhancement and Feature Fusion [J/OL].Heliyon，2024，10（17）：e36861[2024-09-25].https://doi.org/10.1016/j.heliyon.2024.e36861.
2. [2] LI J，TANG C，LEI Z，et al. KRA: K-Nearest Neighbor Retrieval Augmented Model for Text Classification [J/OL].Electronics，2024，13（16）：3237[2024-09-25].https://doi.org/10.3390/electronics13163237.
3. [3] SARIN G，MUKUND P K M. Text Classification Using Deep Learning Techniques: A Bibliometric Analysis and Future Research Directions [J].Benchmarking: An International Journal，2024，31（8）：2743-2766.
4. [4] Wang Qinchen, Duan Liguo, Wang Junshan, etc A Short Text Semantic Matching Strategy Based on BERT Sentence Vector and Differential Attention [J]. Computer Engineering and Science, 2024, 46 (7): 1321-1330
5. [5] Sun Qinghua, Deng Cheng, Gu Zhenyu Design Material Recommendation System Combining Word Vectors and Self Attention Mechanism [J]. Journal of Computer Aided Design and Graphics, 2024, 36 (1): 63-72
6. [6] Zhang Yuhui Research and Implementation of Automated Deep Classification of Documents Based on the Chinese Library Classification System [J]. Library Journal, 2024, 43 (3): 61-74
7. [7] Lv Qi, Shangguan Yan Hong, Li Rui Research on Interdisciplinary Measurement Based on Reference and Text Content Discipline Classification [J]. Journal of Intelligence, 2024, 43 (8): 976-991
8. [8] SAJID N，AHMAD M，RAHMAN U A，et al. A Novel Metadata Based Multi-Label Document Classification Technique [J].Computer Systems Science and Engineering，2023，46（2）：2195-2214.
9. [9] Tan Keren, Lan Yunshi, Zhang Yang, etc Chinese Text Readability Grading Model Based on Multi level Language Feature Fusion [J]. Chinese Journal of Information Science, 2024, 38 (5): 41-52
10. [10] Shen Si, Chen Meng, Feng Shuyang, etc ChpoBERT: A Pre trained Model for Chinese Policy Texts [J]. Journal of Intelligence, 2023, 42 (12): 1487-1497
11. [11] Yi Ming, Li Huoran, Liu Jiyue Research on Online Seminar Information Classification Model Based on GloVe BiLSTM [J]. Intelligence Theory and Practice, 2022, 45 (9): 173-179
12. [12] Zhou Yan Text classification method based on GloVe model and attention mechanism Bi LSTM [J]. Electronic Measurement Technology, 2022, 45 (7): 42-47
13. [13] DEVLIN J，CHANG M W，LEE K，et al. BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding [C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies，Volume 1 （Long and Short Papers）.Minneapolis：ACL，2019：4171-4186.
14. [14] LI X，HU L L. Chinese Long Text Similarity Calculation of Semantic Progressive Fusion Based on Bert [J].Journal of Computational Methods in Sciences and Engineering，2024，24（4-5）：2213-2225.
15. [15] Lu Jiali Multi label classification method for open source threat intelligence text based on Bert TexCNN [J]. Information Security Research, 2024, 10 (8): 760-768
16. [16] AZIZ K，JI D，CHAKRABARTI P，et al. Unifying Aspect-Based Sentiment Analysis BERT and Multi-Layered Graph Convolutional Networks for Comprehensive Sentiment Dissection [J].Scientific Reports，2024，14（1）：14646[2024-09-06].https://www.nature.com/articles/s41598-024-61886-7.
17. [17] Shao Yibo, Qin Yuhua, Cui Yongjun, etc User portrait generation method integrating multi granularity information [J]. Computer Application Research, 2024, 41 (2): 401-407

Funding

Guizhou University of Finance and Economics 2022 Campus level Research Fund Project (2022KYYB14); 2024 Guizhou Province University Humanities and Social Sciences Research Project (2024RW313)

About the authors

Chen Lan (1997-), female, Han ethnicity, Chongqing native, assistant librarian, master's degree, research direction: information management and information services, text mining; Zhou Jie (1992-), male, Han ethnicity, from Hubei and Anhui provinces, associate research curator, master's degree, research direction: data mining, intelligent retrieval; Corresponding author: Yang Fan (1969-), male, Han nationality, from Guiyang, Guizhou, professor, PhD, research direction: knowledge organization and knowledge engineering, text mining.