In the era of online learning,online learning resources provide users with more choices for resource search and learning.However,a large amount of network information and resources make users face a "data overload" situation when conducting online learning.Personalized recommendation can effectively solve this problem and provide users with learning resources that meet their own needs.Judging from the current research,domestic and foreign scholars rarely focus on a specific subject in the research of basic education,and develop targeted and personalized learning resources according to the specific laws of the subject and the learning characteristics of learners.Research.Based on this,this research proposes personalized recommendations for junior high school mathematics learning resources.Through the collection of basic student information and behavioral data,the model constructs students and learning resources,and recommends appropriate personalized after-school learning for students Resources to allow students to learn more efficiently.This article analyzes the requirements for junior high school mathematics education under the guidance of the new curriculum standards,and combines the characteristics of current junior high school learners and mathematics learning resources to construct a junior high school mathematics learning resource library model and a learner model.After discussing and comparing the personalized recommendation algorithms used in the current personalized recommendation,this paper introduces the Slope one algorithm into the user-based collaborative filtering algorithm to solve the cold start problem caused by the sparse scoring data in the system.Hybrid recommendation algorithm.After analyzing the proposed algorithm and personalized model,each module is designed in detail,and a personalized recommendation system for junior high school mathematics learning resources is constructed,and the entire system is tested in an experimental manner.Finally,in the form of a questionnaire survey,statistics and analysis of the use effect of the personalized resource recommendation system are carried out to provide a certain reference for the subsequent improvement of the system. |