基于学情数据的用户画像推荐算法研究OA

In order to achieve intelligent recommendation of personalized and accurate teaching resources,this paper has deep-ly studied the fusion model of algorithms.It combines educational administration and learning situation data with personalized recom-mendation,excavates feature information from student portraits,and achieves more accurate and personalized independent learning recommendation.This paper focuses on the research of personalized recommendation algorithms.The three categories of binary algo-rithms,namely SVM,LR and RF,are considered as three single machine learning methods.They are fused into a new strong learn-er,and a differential joint voting algorithm based on integrated learning is obtained.The user learning situation is personalized ana-lyzed,so that the user's learning difference tags are voted and classified according to the data of different users,thus a chieving per-sonalized learning recommendation guidance.Finally,it extractes features from the learning situation data,builts an evaluation sys-tem with the fusion of new voting algorithms,establishes feature portraits and user interest models,and achieves more accurate and effective recommendations,which has certain practical significance for improving personalized learning recommendation guidance.