Research on Paradigm Reconstruction and Practical Path of Interdisciplinary Teaching Evaluation Empowered by Artificial Intelligence

Abstract

In the era of artificial intelligence, AI technology has become a core driving force for educational innovation and development, especially in the field of interdisciplinary learning. Against the backdrop of digital education transformation, this paper analyzes the limitations of traditional interdisciplinary evaluation paradigms from the dimensions of evaluation objectives, data collection and evaluator participation. It proposes a four-dimensional reconstruction of interdisciplinary teaching evaluation: shifting from knowledge-centered objectives to competency-oriented objectives, transforming terminal summative evaluation to full-process tracking evaluation, upgrading static testing methods to dynamic diagnostic methods, and expanding single-subject evaluation to multi-stakeholder collaborative evaluation. From a practical perspective, this paper constructs a whole-process evaluation system covering pre-class diagnosis, in-class collaboration and post-class extension, and innovates intelligent application paths for portfolio assessment, group cooperative evaluation and concept map evaluation. The study emphasizes that AI-enabled evaluation must adhere to three principles: student-demand orientation, the irreplaceable dominant role of teachers, and technical practicability, so as to provide a complete theoretical framework and operable implementation schemes for the digital transformation of interdisciplinary teaching evaluation.