以评促教:基于事理图谱的计算思维水平评价方法
Assessing for Teaching:A Matter-of-Fact Mapping-based Approach to Evaluating Computational Thinking Levels
詹泽慧 1钟煊妍 1邹萱萱 1骆丽霞1
作者信息
摘要
计算思维是人工智能时代教育发展的核心目标,而监测计算思维状况、评价计算思维水平并据此提出计算思维培养的教学策略是当前"教-学-评"一体化教学改革下实现计算思维培养的重要路径.然而,由于计算思维的复杂性、抽象性和过程性,教师往往难以全面掌握和准确评价学生计算思维应用过程,更难以个性化了解学生的计算思维发展状况并据此提出教学干预策略.为此,从测评的视角出发,基于计算思维的本质内涵,结合新兴的事理图谱概念,构建基于事理图谱的计算思维评价方式.首先,设计面向计算思维的事件编码框架,利用iB-CM算法实现事件关系抽取,生成事理图谱.然后,利用随机森林根据事理图谱对计算思维水平进行分类,并采用基于规则的事理图谱优化方法,完善iBCM提取出的特征模式,生成可视化的事理图谱辅助教学策略的调整与个性化的指导.基于此方法开展的实证研究表明,设计的模型在各项指标上都具有良好的性能效果.其结合了事理图谱所具备的事件逻辑深层次分析与推理的优势,为理解学习者的思维过程、抽取思维特征、探究思维模式提供了支撑和条件,助力教师培养学习者的计算思维提供路径指导.
Abstract
Computational thinking is the core goal of education development in the era of artificial intelligence,and how to monitor the status of computational thinking,evaluate the level of computational thinking,and propose teaching strategies for computational thinking de-velopment accordingly is an important path to realize the development of computational thinking under the current teaching and evaluation in-tegration teaching reform.However,due to the complexity,abstractness and process of computational thinking,it is often difficult for teach-ers to achieve a comprehensive grasp of the process of students'computational thinking application and accurate evaluation,and even more difficult to personalize the understanding of the development of students'computational thinking and propose teaching intervention strategies accordingly.In this study,from the perspective of assessment,based on the essential connotation of computational thinking,combined with the emerging concept of event mapping,this study constructs a computational thinking assessment method based on event mapping.Firstly,we design the event coding framework for computational thinking,and utilize iBCM to realize event relationship extraction and generate the matter-of-fact map.Then,we use random forest to classify computational thinking levels according to the event map,and adopt the rule-based event map optimization method to improve the feature patterns extracted by iBCM,and generate visual event maps to assist the ad-justment of teaching strategies and personalized instruction.The empirical research based on this method shows that the designed model has good performance in all indicators.It combines the advantages of deep analysis and reasoning of event logic possessed by event mapping to provide support and conditions for understanding learners'thinking process,extracting thinking features,exploring thinking patterns,and assisting teachers in cultivating learners'computational thinking to provide path guidance.
关键词
以评促教/事理图谱/计算思维/评价方法Key words
Assessing for Teaching/Event Graph/Computational Thinking/Evaluation Methods引用本文复制引用
基金项目
国家自然科学基金面上项目(62277018)
教育部人文社会科学研究项目(22YJC880106)
广东省学位与研究生教育改革研究项目(2023JGXM＿046)
出版年
2024
NETL
NSTL
万方数据