Article Details

Abstract

This research aims to deal with the problems such as individual differences among students, limited teaching resources, and inadequate teacher-student interaction in education, and introduces an innovative education and teaching knowledge question and answering system based on Large Language Model. The system integrates classroom knowledge and network resources, builds a comprehensive and rich teaching knowledge base, and improves data quality through meticulous data cleaning and classification processing. To optimize performance, the research team designs and implements an efficient multi-model retrieval mechanism, which ensures the system could respond quickly in different hardware environments. This system not only improves response speed but also reduces infrastructure costs. It is suitable for multiple fields such as school education and party class training, and has a profound impact on the digital transformation and innovation of education.

Keywords

Large Language Model; Artificial Intelligence education; efficient retrieval model

References

1. [1] Zhu Weiping, Lin Hai, Xie Rong, etc Application of Intelligent Question Answering System in College Curriculum Teaching [J]. Computer Education, 2019 (10): 23-26
2. [2] Hu Ping, Pu Xiaoqiong Research and Application of Interactive Autonomous Learning Teaching Mode Based on "Internet plus" [J]. Journal of Higher Education, 2022, 8 (17): 80-83
3. [3] Zhang Chunhong, Du Longfei, Zhu Xinning, etc Research on Educational Question Answering System Based on Large Language Model [J]. Journal of Beijing University of Posts and Telecommunications: Social Sciences Edition, 2023, 25 (6): 79-88
4. [4] He Chaobo, Lin Xiaofan, Cheng Junwei, etc Empowering the Construction of a Strong Education Country with Learning Communities: From the Perspective of Online Learner Relationship Network Analysis [J]. China Electronic Education, 2024 (6): 38-45
5. [5] Feng Xingli, Hong Dandan, Luo Junfeng, etc Research on Automated Acceptance Testing Technology for Unified Identity Authentication in Universities Based on Selenium+Python [J]. Modern Electronic Technology, 2019, 42 (22): 89-91+97
6. [6] Fu Qiang, Xu Zhenping, Sheng Wenxing, etc End to end Chinese speech recognition method combining byte level byte pair encoding [J/OL]. Computer Applications, 1-8 (May 17, 2024) http://kns.cnki.net/kcms/detail/51.1307.TP.20240515.0955.004.html.
7. [7] Bai Xingzhen, Sui Shuting, Ge Leijiao, etc Wind speed power anomaly data cleaning method based on sliding quartile and feasible search circle algorithm [J]. Journal of Shandong University of Science and Technology: Natural Science Edition, 2023, 42 (6): 106-116
8. [8] Jiang Haoda, Zhao Chunlei, Chen Han, etc A method for constructing a domain sentiment dictionary based on improved TF-IDF and BERT [J]. Computer Science, 2024, 51 (S1): 162-170
9. [9] Zhao Zichen, Yang Feng, Guo Yuhui, etc Design of Accelerator Big Data Secure Storage and Efficient Analysis System Based on Hadoop Technology [J]. Modern Electronic Technology, 2024, 47 (8): 9-17
10. [10] Yang Binxia, Luo Xudong, Sun Kaili The latest progress in machine translation based on pre trained language models [J]. Computer Science, 2024, 51 (S1): 50-57
11. [11] MORALES-CHAN M，AMADO-SALVATIERRA H R，MEDINA J A，et al. Personalized Feedback in Massive Open Online Courses: Harnessing the Power of LangChain and OpenAI API [J].Electronics，2024，13（10）：1960.
12. [12] Guo Shun Research on Word and Text Representation and Text Classification [D]. Dalian: Dalian University of Technology, 2020
13. [13] Wu Yongliang, Zhao Shuliang, Li Changjing, etc Text classification method based on TF-IDF and cosine similarity [J]. Chinese Journal of Information Science, 2017, 31 (5): 138-145
14. [14] FERREIRA R，CANESCHE M，JAMIESON P，et al. Examples and Tutorials on Using Google Colab and Gradio to Create Online Interactive Student-learning Modules [J].Computer Applications in Engineering Education，2024，32（4）：e22729.

Funding

Research Project on Education Information Technology in Sichuan Province (kt202309286459124); 2023 Party Building Research Project of Hope College, Southwest Jiaotong University (19)

About the authors

Yi Yunheng (1998-), male, Han nationality, from Dazhou, Sichuan, full-time teacher, master's degree, research direction: natural language processing; Corresponding author: Pan Ji (1998-), female, Han ethnicity, from Huludao, Liaoning Province, full-time teacher, master's degree, research direction: undergraduate teaching of computer science.