人工智能时代机器辅助教学:能力向度及发展进路

Machine-assisted Instruction in the Era of Artificial Intelligence:Ability Dimensions and Development

人工智能技术变革下的机器辅助教学为教育教学发展注入新的动能。机器辅助教学的内涵与外延在被人工智能技术重塑的同时,其结构也在智慧性、主体性和协同性等方面有新发展。然而,随着教学机器服务能力不断增强,范畴不断拓展,如何对其教学成效展开合理评测成为亟待解决的重要议题。机器辅助教学的能力向度是指机器赋能教学过程的指向目标与能力层级,对其剖析并构建能力向度模型,能以全新视角全面认知教学机器的客观概况与内部逻辑,又可从智能教学机器应用中提取可观测、可测量、可解释的关键因素,借以解决能力评测问题。本研究基于对机器辅助教学基本内涵的探讨,确定机器辅助教学的价值取向、系统框架与关键要素,构建并验证了以教育主体、场景、资源和数据为承载的机器辅助教学能力向度模型。这一模型研究需要秉持教育伦理先导的理念,坚持数据与理论双向驱动,推动人机协同并进,结合因果推断增强数据解释,从而为机器辅助教学能力评测标准的制定与实践提供理论参照和方向指引。

Machine-assisted instruction under the evolution of artificial intelligence technologies has provided more and more power into the new round of education and teaching development.The intension and extension of machine-assisted instruction is being reshaped by artificial intelligence technologies,and its structure also has new changes in terms of wisdom,subjectivity,and coordination.However,with the continuous extension of the teaching machine ability,how to evaluate the instructional effect of teaching machines has become a challenging and urgent issue at present.The ability dimension of machine-assisted instruction includes all dimensions and levels of resources and services provided by teaching machines.By analyzing and constructing the ability dimension model,it is possible to gain a new perspective and comprehensive understanding of the objective profile and internal logic of teaching machines and extract observable,measurable,and interpretable key elements to solve the problem of ability assessment.The construction of the ability dimension model should be based not only on the basic connotation but also on a certain paradigm.This paradigm includes:defining the value orientation,specifying the system framework and key elements,and constructing,validating,and evaluating the ability dimension model.Under the guidance of this paradigm,a theoretical deconstruction of the ability dimensions is achieved by analyzing the internal logic of machine-assisted teaching systems and instructional activity systems.Thus the construction procedure of the ability dimension model of machine-assisted instruction is performed.In the future,the development and evolution of machine-assisted instruction also need to construct educational ethics,stick to two-way integration between theory and data,promote human-machine collaboration,and augment data interpretation with causal inference.Thus,this model will provide a theoretical reference and direction for the development and practice of machine-assisted instruction ability assessment standards.