Design And Implementation Of Personalized Recommendation System For Crowdfunding Educational Resources

With the improvement of Internet transmission technology,the growth of network data resources is explosive.People's demand for self-learning and information acquisition is increasing,and the ways of learning to improve themselves are also increasing.This kind of trend also makes the traditional way of education develop towards digitalization and network.On the one hand,the scale of digital education resources is also expanding,on the other hand,the education mode of crowdfunding and mass creation has also attracted more attention.In this context,the online education mode based on crowdfunding and mass creation mode,as the mainstream of online education mode,has a strong cyclical and development value.The data scale of digital education has far exceeded the traditional model.Because of the wide sources of digital resources,the storage structure of data has a certain complexity,which is not easy to classify and find.In order to mitigate the impact of educational data overload and increase the utilization rate of digital resources,it is necessary to be able to understand users' interests and tastes,accurately locate resources,and make personalized suggestions.This dissertation studies the digital resources recommendation technology under the crowdfunding and crowdfunding mode,and designs and implements the personalized recommendation system of educational resources on this basis.The main research contents of this dissertation are as follows:1)In this dissertation,a method of knowledge reasoning based on vector representation is proposed,which improves the accuracy of knowledge reasoning by dealing with the relationship among triples.2)In this dissertation,an improved collaborative filtering recommendation algorithm is proposed,which uses the association search of knowledge map to replace the traditional clustering algorithm for resource selection,and uses a fusion neural network model to recommend and predict the target users,and achieves better recommendation results.3)In this dissertation,a set of personalized recommendation platform system for educational resources is built,which integrates the acquisition and integration of crowdfunding and crowdfunding educational resources,complements the knowledge reasoning of knowledge map,and makes a unified integrated scheduling of personalized recommendation module,and finally presents it to users,improving the use value of digital educational resources.