Research On The Information Resource Recommendation Based On User's Tagging Interest Model

The value of the Internet comes from the knowledge and creativity of online users,and the explosive growth of network information is driven by users.In the social tagging system,the user adds personalized tags to the interested resource to implicitly reflect the individual's interests and hobbies.A user interest model represented by tag features is constructed.The similarity calculation calculates the similarity user of the target user,and the similar user's knowledge resource is used as the candidate recommendation resource set of the target user.According to a certain algorithm,the most relevant resource is selected to generate the target.The user's recommended resource list can realize the user's personalized information resource recommendation.This kind of collaborative user interest mining and utilization is the main way for the current social tagging system and some e-commerce platforms to personalize information or recommend products for users.This paper breaks through the user interest model of the planar structure in previous studies,builds a user interest model of a hierarchical tree structure,uses hierarchical labels to express user interests,and uses watercress network as a data source to capture user-centered annotation data to users.The tag set characterizes the user features,establishes a planar structure and a hierarchical structure model that reflecting the user's interests,and uses the Pearson correlation coefficient algorithm to calculate Top-N neighboring readers that have the greatest similarity with the target user's interests;and then reads and thinks about the books according to neighboring readers.The four states of read,want to read,reading and unread are given different weights,and the book's recommendation score is calculated.Finally,the top M book with a higher rating is recommended to the target user.By comparing the recommendation results based on user interest model based on planar structure and hierarchical tree structure,it is found that the hierarchical structure model of user interest can optimize the target user's resource recommendation and discover more new interested resources for the target user and achieve better personalized information recommendation.