摘要
知识建构以“观点”为核心,对在线协作知识建构过程中的观点进行实时精准分类、监控和干预能够支持学习者对知识构建过程有觉知的反思,促进观点改进。为避免人工编码低效和深度学习的可解释性不足,文章设计了包含六个环节的机器学习驱动的在线协作知识建构观点分类方法,采用Word2Vec对特征词进行向量表征,以SMOTE过采样技术解决样本量不均衡问题,采用随机森林(RF)、K近邻(KNN)、支持向量机(SVM)和逻辑回归(LR)四种机器学习算法模型构建观点分类器,特别是为区分不同特征的重要性,本文自定义TF-IDF叠加增强权重算法,提高了机器学习观点分类的准确性。通过对比实验和消融实验,验证SVM算法表现最佳,分类准确率达到92%,F1值达到87%。该方法面向在线协作知识建构教学的过程性评价、调控学习者知识建构进程、促进观点改进等方面具有较高的应用价值。
Knowledge building is centered on“ideas”,real-time accurate classification and monitoring of ideas in online collaborative knowledge building process which can support learners to reflect on the knowledge building process with awareness and promote the improvement of ideas.To avoid the inefficiency of manual coding and the lack of interpretability of deep learning,the paper designs a machine learning-driven online collaborative knowledge building idea categorization method that includes six stages.Word2Vec is used to vectorize feature words,while the SMOTE oversampling technique addresses the sample size imbalance.Additionally,four machine learning algorithms—Random Forest(RF),K-Nearest Neighbors(KNN),Support Vector Machines(SVM),and Logistic Regression(LR)—are employed to construct the idea classifiers.In order to distinguish the importance of different features,this paper customizes the TF-IDF superimposed enhancement weighting algorithm to improve the accuracy of machine learning idea classification.The SVM algorithm has been verified to perform the best with 92%classification accuracy and 87%F1 value through comparison and ablation experiments.The method is oriented to the process evaluation of online collaborative knowledge building,regulating the process of learners'knowledge building,and promoting the improvement of ideas with high application value.
作者
陈洁
李佳政
CHEN Jie;LI Jia-Zheng(Teachers College,Shihezi University,Shihezi 832003,Xinjiang,China)
出处
《兵团教育学院学报》
2026年第1期41-49,共9页
Journal of Bingtuan Education Institute
基金
2025年度自治区教育科学规划项目“AI智能体赋能‘教—学—评’的探索与应用研究”(HES2025008)
石河子大学2024年度高层次人才科研启动项目“人机智能协同情境下学习者高阶思维能力发展机制研究”(RCSK202437)。
关键词
在线协作知识建构
机器学习
观点分类
online collaborative knowledge building
machine learning
idea classification
