Research on Personalized Recommendation Method of Academic Resources based on Hadoop

Facing the huge amount of academic resources, it is difficult for users to find the needed resources accurately through keyword search. On the basis of collaborative filtering, a personalized recommendation method for academic resources is proposed by fusing user attributes, academic resource attributes, and users' scores on academic resources, effectively alleviate the scarcity of scoring and cold start problems, improve the accuracy of the recommended method. Combined with the advantages of Hadoop distributed computing, design and implementation the highly effective academic resources personalized recommendation method which based Hadoop, the validity and of the design are proved by experiments.