基于K-means聚类分析算法的大学生在线学习行为分析

K-Means Clustering Analysis Algorithm Applied in College Students Online Learning Behavior Analysis Based on Data Mining Technology

本文利用基于Python平台的机械学习算法库scikit-learn提供的无监督学习的机械学习算法K-Means聚类分析模型,对在学习通平台上进行在线学习的学生产生的学习行为数据进行聚类分析.通过对学习行为数据的抽取、清洗、转换和K-means模型的训练,最终将在线学习的学生分为4种类型的学习者,并且给出每名在线学习学生的学习者类型.本文重点分析了每种类型学习者的线上学习行为数据,结合教师实践教学经验对4种类型的学习者的学习行为特征进行分析和总结,并针对每种类型在线学习者的学习行为特征制订了科学、合理和个性化的教学方案,有效提升了学生的在线学习效率,也为疫情结束后有针对性地制订个性化、合理、有效的线下教学方案提供了决策依据.

In this paper,we use k-means clustering analysis model of unsupervised learning algorithm provided by scikit learn,which is a mechanical learning algorithm library based on Python platform,to cluster and analyze the learning behavior data generated by online learning students on the learning pass platform.Through the extraction,cleaning and transformation of learning behavior data and the training of K-means model,the online learning students are finally divided into four types of learners,and the learner type of each online learning student is given.This paper focuses on the analysis of online learning behavior data of each type of learners,analyzes and summarizes the learning behavior characteristics of four types of learners combined with the practical teaching experience of teachers,and formulates a scientific,reasonable and personalized teaching scheme for each type of online learners’ learning behavior characteristics,which effectively improves the online learning efficiency of students,and also after the end of the epidemic Making personalized,reasonable and effective offline teaching programs provides decision-making basis.