The electronic information of equipment involves a wide range of data sources and various types. It is necessary to accurately extract useful information from massive data. Therefore, an accurate recommendation algorithm of electronic information resources of equipment based on knowledge graph is put forward. The knowledge graph of the equipment electronic information resources based on the text and structure. CNN(Cellular Neural Network) is used to complete the knowledge graph, so that the algorithm covers the resources more comprehensively. The user's interests and preferences ares analyzed, and the characteristics of the device's electronic information resources are extracted. Finally, a collaborative filtering recommendation algorithm is used to obtain the resource similarity matrix, predicting the user's retrieval behavior, so as to obtain the recommendation list. The experiment proves that the average coverage of the proposed algorithm is 94. 5% , the average hit rate is 96. 7% , and the cumulative gain of normalized loss reaches 0. 91, which can accurately recommend the required information resources for users.

knowledge graph, equipment electronic information resources, complete the knowledge graph, user interests, collaborative filtering recommendation algorithm