基于数学模型分析教师使用生成式大语言模型的影响因素研究

A Study on the Influencing Factors of Teachers Using Generative Large Language Models Based on Mathematical Model Analysis

摘要: 随着教育界对人工智能技术的广泛应用和深度探索，尤其以文心一言、ChatGPT等生成式大语言模型为代表的先进技术，在高等教育环节中展现出了愈发显著的应用潜力。本文通过建立数学神经网络模型分析高校教师在采纳生成式大语言模型进行教学的影响因素。研究团队通过发放问卷的形式，利用Pearson相关系数矩阵对数据进行一致性与相关性分析。然后，建立GA-BP神经网络预测模型，分析发现感知信任、感知专业及感知智能，均在不同程度上能够正面预见他们对人机协同教学模式持有的肯定立场。为高校教师进一步利用生成式大语言模型提高教学质量提供了理论指导。

Abstract: With the extensive application and in-depth exploration of artificial intelligence technology in the educational world, especially the advanced technologies represented by the generative big language models such as ERNIE Bot, ChatGPT, etc., have shown more and more significant application potential in the link of higher education. This article analyzes the influencing factors of college teachers adopting generative language models in teaching by establishing a mathematical neural network model. The research team conducted consistency and correlation analysis on the data by distributing questionnaires and using the Pearson correlation coefficient matrix. Then, a GA-BP neural network prediction model was established, and analysis revealed that perceived trust, perceived professionalism, and perceived intelligence can all positively predict their positive stance on the human-machine collaborative teaching model to varying degrees. This provides theoretical guidance for university teachers to further utilize generative language models to improve teaching quality.