In response to the problems of lagging feedback, strong subjectivity, and insufficient process guidance in traditional internship teaching support assessment, this article designs a set of ability dynamic assessment software system based on practical teaching feedback. The system has constructed a three-dimensional assessment system covering teaching design, classroom implementation, and professional development, tracking the process from three dimensions: teaching behavior, language expression, and teaching content. This system focuses on key indicators such as teacher-student interaction, questioning level, language clarity, and knowledge logic. By integrating multiple sources of data such as classroom videos, lesson plan texts, and online behavior, the system achieves continuous analysis and correlation mining of teaching performance, and generates personalized feedback reports containing ability weakness identification, specific improvement suggestions, and growth trajectories. The experiment shows that the system can significantly shorten the feedback cycle, reduce assessment errors by about 65%, and form a closed-loop support mechanism of "collection analysis feedback improvement". In the 12 week teaching practice, the key indicators such as classroom interaction quality and teaching reflection depth of the internship teaching students improved by an average of 1.4-1.7 points, promoting their teaching behavior improvement and continuous development of professional abilities.

Yaoyao Li. Architecture and Implementation of Dynamic Assessment Software for Internship Teaching Students' Abilities Based on Practical Teaching Feedback. Frontiers in Educational Research (2026), Vol. 9, Issue 4: 78-83. https://doi.org/10.25236/FER.2026.090412.

[1] Liu Xiaohu, Tang Bin, Wang Xuesong, et al. 2024. Research on the Teaching Quality Assessment Model of Elastic Mechanics Course Based on PSO-BP Algorithm. Journal of Changchun Normal University 43, 8, 86–92.

[2] Evstigneev M N. 2024. A model of language and methodological pre-service teachers' training based on artificial intelligence technologies. Tambov University Review. Series: Humanities 29, 5, 1222–1238.

[3] Luo Shijie, Han Shuzhen. 2023. Teaching Quality Analysis Algorithm Integrating Random Forest and Neural Network. Small Microcomputer System 44, 8, 1686–1692.

[4] Tilton Z, LaVelle J M, Ford T, et al. 2023. Artificial intelligence and the future of assessment education: Possibilities and prototypes. New Directions for Assessment 2023, 178-179.

[5] Khalid P Z M, Sulaiman S, Sufi M K A, et al. 2023. Systematic Literature Review: Integrating Artificial Intelligence (AI) in Teaching and Learning of Language. AJELP: Asian Journal of English Language and Pedagogy 11, 1, 108–119.

[6] Wang Bing, Zheng Shangnan, Li Panchi, et al. 2023. Online teaching quality assessment method based on the concept of diversified assessment subjects and convolutional neural networks. Modern Electronic Technology 46, 8, 91–98.

[7] Pang Shiyan, Hao Jingjing, Hu Hanchun, et al. 2023. Teacher Teaching Behavior Recognition Method Based on Spatiotemporal Graph Convolutional Neural Network. Journal of Central China Normal University (Natural Science Edition) 57, 5, 715–723.

[8] Gao Ting. 2023. Quality Assessment Method for Flipped Classroom Teaching Reform Based on Convolutional Neural Networks. Journal of Shenyang Medical College 25, 5, 552–556, 560.

[9] Tidake V M, Das U, Bhatti K K, et al. 2024. Vocational education skill assessment and intelligent assistance: A study on the application of machine learning algorithms in the assessment of vocational information literacy teaching ability. Journal of Theoretical and Applied Information Technology 102, 13, 5237–5252.

[10] Ni H, Zhang F, Ye J, et al. 2024. Research on assessment of college-classroom teaching quality based on pentapartitioned neutrosophic cubic sets and machine vision. Journal of Advanced Computational Intelligence and Intelligent Informatics 28, 5, 1132–1143.

[11] Zhang Xiaoheng, Gong Youlong. 2024. Research on Classroom Teaching Quality Assessment Based on Artificial Intelligence Technology. Science and Technology Information 22, 8, 215–219.

[12] Wei Peiwen, Zhu Ke, Ye Haizhi, et al. 2024. Construction of a Precision Teaching Ability Assessment Model for College Teachers Based on BP Neural Network. Journal of Henan Normal University (Natural Science Edition) 52, 5, 108–116.