A method for constructing a resource recommendation model and a resource recommendation method based on the model

Abstract

The present invention provides a method for constructing a resource recommendation model and a resource recommendation method based on the model, including: A. obtaining each resource information, and obtaining each keyword tag and word frequency weight of each resource information; B. obtaining each User's behavior data, and obtain each user's attention weight of each resource information; C, establish a graph database model, and each resource and its corresponding keyword tags and the weight of each keyword tag, and each user and their attention Each resource and the attention weight of each resource are correspondingly stored in the graph database model, so as to construct the association among the resources, tags and users. From the above, this application does not require a preprocessing clustering process, and can flexibly construct tags, which is conducive to efficient and real-time recommendation of resources.

Description

technical field [0001] The invention belongs to the technical field of computer applications, and in particular relates to a method for constructing a resource recommendation model and a resource recommendation method based on the model. Background technique [0002] Database technology has become the core and foundation of information systems. However, with the expansion of data scale and the increase of data complexity, the relational model can no longer meet the needs of the field. Taking social networks as an example, the use of relational databases will lead to data redundancy, and cannot adapt to the dynamics of social data, nor can it be very good It supports multi-layer complex queries such as "friends of friends". Aiming at the problem of complex and dynamic internal relationships between data, people once again turn their attention to graph databases. Graph databases can effectively store, manage, and update data and their internal relationships, and can efficient...

AI Technical Summary

Problems solved by technology

Usually these data are stored in the graph database, but this method has an obvious disadvantage that the recommendation results cannot be calculated in real time. Generally, the clustering process needs to be preprocessed, and the tags cannot be flexibly constructed, and efficient calculations cannot be completed, which is not conducive to resources for efficient recommendations

Images