基于Android平台的移动学习系统大数据挖掘技术研究

摘要

The big data mining technology of mobile learning system based on Android platform is studied. User clustering was performed by means of the expectation maximization (EM) algorithm. The personalized resource recommendation model of mobile learning system is used to select neighbour users and predict the score. The CRISP-DM model is used to establish the mining model of ARTXP algorithm according to the time series data of learner′s downloading resource. The downloading of Eng-lish courseware,law courseware and computer courseware in seven days is forecasted,which shows that the demand of English mobile learning resources is declined,and the demands of law courseware and computer courseware are increased. The model can predict other courses′ learning resource demand of mobile learning system,and create and upload the corresponding mobile learning resources according to the changes in demand.%基于Android平台对移动学习系统大数据挖掘技术进行研究.通过期望最大化EM算法进行用户聚类,利用移动学习系统的个性化资源推荐模型进行近邻用户的选取和评分预测,采用CRISP-DM模型,根据学习者下载资源的时间序列数据建立ARTXP算法挖掘模型,通过对英语类课件、法律类课件、计算机类课件在7天后的下载预测,表明英语类的移动学习资源需求有所下降,法律类与计算机类课件需求有所增加,同样,可对移动学习系统的其他类学习资源需求进行预测,根据需求的变化情况制作并上传相应移动学习资源.