Recommendation Research And System Implementation Of Educational Resources For Elementary And Middle Schools On Social Networks

With the development of social informatization and Internet technology,the way people receive education is gradually enriched.the ways of obtain educational resources are more and more diversified,and the trend of educational resources expanding from offline to online becomes inevitable.In the background of"Internet + education",the rapid development of social networks and the continuous iterative updating of social networking products have attracted a large number of educators,teachers,parents and children to break through the traditional methods of education and teaching,using social networks for educational resource information.In exchange,people are more willing to seek educational resource information from social networks.Therefore,how to obtain more effective educational resources from a large amount of information resources in a limited time is an urgent problem to be solved.This paper takes the social network primary school education resources as an example,and proposes educational resources services for primary school students,parents and other user groups concerned with primary school education resources,and designs a good recommendation algorithm,this algorithm solve how users can quickly acquire and collect the primary education resources needed,and tap into the user community that focuses on primary education resources and recommend useful educational resources.This paper mainly studies from the following parts:The first part is a collaborative filtering recommendation algorithm based on mixed similarity.Aiming at the sparse similarity calculation method of traditional collaborative filtering algorithm,the data set sparse problem is easy to occur.The calculation method of keyword table is used to indirectly calculate the content similarity.At the same time,the relationship between author,user and resource text is started from multiple angles.In the traditional collaborative filtering algorithm,the author similarity and user similarity are calculated respectively,then the similarity degree is calculated by the mixed similarity weighting,and the recommendation is based on the comprehensive similarity value.The second part is fusion recommendation similarity and eigenvalue recommendation algorithm.Considering the characteristics of educational resources from three aspects:content richness,resource timeliness and information authority,extracting the feature calculation methods of educational resources in social networks,and then proposing an improved recommendation algorithm,The eigenvalue are fusioned to calculate the recommendation level for recommend.The experiment results show that our algorithm can improve the recommendation satisfaction.The third part is mainly to design the recommendation method for resource content and the recommendation method for user groups.The recommendation algorithm is used to give the resource text with the highest recommendation satisfaction,and the recommendation method for resource content is proposed.Then,the user groups with high resource attention are further explored,and the recommendation method for user groups is proposed.The fourth part is mainly to recommend system development and design implementation.The recommendation system mainly provides the user with an interface that can directly contact the recommended function,and guides the user to use the recommend function.This article aims to help primary school students and parents quickly and accurately access relevant educational resources,saving students and parents limited spare time.At the same time,it can help the school's education managers to conduct centralized research and management for educational researchers,and provide them with corresponding primary education resources.In addition,I hope to contribute to the work of building an educational resource platform for social networks through the research in this paper.