Article Details

Abstract

In the face of the challenges in traditional literary research, where intertextuality analysis relies on personal accumulated knowledge and is difficult to systematize or quantify, this research proposes a Text Historical Analysis Knowledge Base Question Answering System based on the DeepSeek model. This system utilizes technologies such as text chunking, vectorization, knowledge graph, and Retrieval-Augmented Generation, to construct a historical knowledge base using historical literature materials. It vectorizes the query text, conducts relevant retrieval in the knowledge base, and uses a Large-Language Model to organize and output comprehensive analysis results. Experiments demonstrate that this system can accurately identify the historical content in the query text and output reasonable analysis, which helps readers better understand the compositional background and writing intention of the text.

Keywords

DeepSeek; Knowledge Graph; Retrieval-Augmented Generation; knowledge base; Large Language Model

References

1. [1] Singhal A. Official Google Blog: Introducing the Knowledge Graph: Things, not Strings [EB/OL].[2025-12-11].https://www.mendeley.com/catalogue/ca10852b-1f48-3172-9b4f-8b521878b39d/.
2. [2] Zhao Yingying, Zhu Lingrate A Review of Knowledge Graph Construction Technology Research [J/OL]. Computer Engineering, 2025:1-21 https://doi.org/10.19678/j.issn.1000-3428.0252971.
3. [3] Bordes A，Usunier N，Garcia-Durán A，et al. Translating Embeddings for Modeling Multi-relational Data [C]//Proceedings of the 27th International Conference on Neural Information Processing Systems，2013：2787-2795.
4. [4] Wang Z，Zhang J W，Feng J L，et al. Knowledge Graph Embedding by Translating on Hyperplanes [C]//Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence，2014：1112-1119.
5. [5] Sun Z Q，Deng Z-H，Nie J-Y，et al. RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space [PP/OL].https://arxiv.org/abs/1902.10197.
6. [6] Schlichtkrull M，Kipf T N，Bloem P，et al. Modeling Relational Data with Graph Convolutional Networks [PP/OL].https://arxiv.org/abs/1703.06103.
7. [7] Vashishth S，Sanyal S，Nitin V，et al. Composition-based Multi-relational Graph Convolutional Networks [PP/OL].https://arxiv.org/abs/1911.03082.
8. [8] Wang B Y，Xu L Y，Wang J Y，et al. DeepSeek and Qwen in Healthcare: Pioneering Multimodal Large Language Models for Nextgeneration Disease Prediction [C]//2025 10th International Conference on Computer and Communication System （ICCCS），2025：100-105.
9. [9] Ren Haiyu, Liu Jianping, Wang Jian, etc A review of research on intelligent question answering systems based on big language models [J]. Computer Engineering and Applications, 2025, 61 (7): 1-24
10. [10] Lewis P，Perez E，Piktus A，et al. Retrieval-Augmented Generation for Knowledge-intensive NLP Tasks [PP/OL].https://arxiv.org/abs/2005.11401.
11. [11] Li Zixuan, Bai Long, Ren Weicheng, etc Overview of Key Technologies for Knowledge Graph Empowered by Code Language Model [J]. Big Data, 2025, 11 (2): 3, 19-28

Funding

2025 Anhui Province Quality Engineering Project Teaching Research Project "Research on the Status Quo and Improvement Strategies of Digital Literacy of Vocational College Teachers from the Perspective of Digital Transformation" (2025jyxm0958)

About the authors

Wang Xiaomei (born on July 11, 1987), female, Han ethnicity, from Chuzhou, Anhui Province, is a teaching assistant and master's student. Her research interests include linguistics and applied linguistics theory and practice, reform of Chinese language curriculum teaching in higher vocational education, and innovation of Chinese language teaching models and methods in higher vocational education; Liu Jun (February 2002-), male, Han ethnicity, from Hefei, Anhui, is currently pursuing a master's degree in wireless sensing technology.