Article Details

Abstract

The Bayesian formula is characterized by high abstractness. Traditional teaching modes focus on symbolic derivation, which makes it difficult for students to achieve “understanding-transfer” in Artificial Intelligence scenarios. This paper constructs a three-level progressive teaching framework of “concept image-multiple representation-cognitive transfer”. The first level uses a “disease screening” magnetic puzzle activity to activate students’ intuitive cognition of prior probability, likelihood and posterior probability. The second level develops web interactive animations to dynamically demonstrate the belief updating process in spam classification and reinforce the understanding of the Bayesian iterative mechanism. The third level relies on the classic mushroom dataset to guide students to manually build a Naive Bayes classifier and compare it with the classification results of 22-dimensional features implemented based on scikit-learn, so as to complete the far transfer to high-dimensional real-world tasks.

Keywords

Bayesian formula; visualization; concept image; multiple representation; cognitive transfer

References

1. [1] Su You, Liang Xin, Hu Zhijun, etc Exploration of Teaching Innovation in the Course of Probability Theory and Mathematical Statistics: Taking Total Probability Formula and Bayesian Formula as Examples [J]. Education and Teaching Forum, 2022 (9): 65-68
2. [2] Huang Shiwen, Liu Chaohui, Luo Lingyun, etc Research on Bayesian Knowledge Tracking Model Integrating Behavior and Forgetting Factors [J]. Computer Application Research, 2021, 38 (7): 1993-1997
3. [3] Cao Jianmei, Feng Chenjiao, Wang Fengxiang Case study on introducing mathematical modeling ideas into the teaching of Bayesian formulas [J]. Jiangsu Science and Technology Information, 2020, 37 (34): 57-59
4. [4] Gu Lijuan, Sun Xijing, Liu Xiaolei Probability and Statistics Teaching Based on "Scenario based+Task driven" - Taking Bayes Formula as an Example [J]. Advanced Mathematics Research, 2023, 26 (4): 6-10
5. [5] Xu Daojun, Sun Dasheng, Wu Suqin Course Ideological and Political Design of Total Probability Formula and Bayesian Formula [J]. Advanced Mathematics Research, 2023, 26 (6): 81-83, 92
6. [6] Cheng Yan, Yan Yang, Lu Chou Li Exploration of Bayes Formula Teaching Method under the Concept of Course Ideology and Politics [J]. College Mathematics, 2023, 39 (5): 113-118
7. [7] Xuanzu Xing, Zhang Lixin, Yuan Anfeng Curriculum Ideological and Political Teaching Design in Bayesian Formula [J]. College Mathematics, 2022, 38 (2): 104-111
8. [8] TALL D，VINNER S. Concept Image and Concept Definition in Mathematics with Particular Reference to Limits and Continuity [J].Educational Studies in Mathematics，1981，12：151-169.
9. [9] AINSWORTH S，GRIMSHAW S. Evaluating the REDEEM Authoring Tool: Can Teachers Create Effective Learning Environments? [J].International Journal of Artificial Intelligence in Education，2004，14（3-4）：279-312.
10. [10] PERKINS D N，SALOMON G. Transfer of Learning [M]//HUSÉN T，POSTLETHWAITE T N. The International Encyclopedia of Education.2nd ed.Oxford：Pergamon，1994：425-441.
11. [11] UCI Machine Learning Repository. Mushroom [DB/OL].（1987-04-26）[2025-07-27].https://doi.org/10.24432/C5959T.

Funding

Cultivation of characteristic courses for the integration of digital technology at the National University of Defense Technology in 2024

About the authors

Wang Shuaige (1991-), female, Han nationality, from Pingdingshan, Henan Province, teaching assistant, master's student, research direction: university mathematics teaching and mathematical modeling.