Knowledge graph data mining and recommending method based on dynamic suboptimal minimum spanning tree

Abstract

The invention discloses a knowledge graph data mining and recommending method based on a dynamic suboptimal minimum spanning tree, which comprises the following steps of: 1) performing isomorphic sub-graph matching in a knowledge graph according to a sub-graph corresponding to a query condition to obtain a plurality of query results; abstracting each query result into a node; 2) constructing a minimum spanning tree of the query results by taking the importance degree of each query result as the sum of the importance degrees of the entities in the query result as the weight of a corresponding node and taking the dissimilarity degree between the query results as an edge weight; and (3) cutting edges in the minimum spanning tree to form a plurality of different connected branches, and respectively returning a node with the highest importance degree in each connected branch to serve as a query result to be output. According to the method, the importance and diversity of the promotion content can be considered at the same time, and the waiting time of the user can be shortened.

Description

technical field [0001] The invention belongs to the field of knowledge map data mining, and in particular relates to a recommendation method that uses a dynamic suboptimal minimum spanning tree to simultaneously consider the importance and diversity of search content. Background technique [0002] With the diversification of online social platforms and entertainment platforms, it has become the main goal of the recommendation system to provide users with high-quality recommended content or search results according to user preferences or query instructions entered by users. Some well-known platforms such as Zhihu, Bie Bilibili has its own recommendation algorithm. Although it is a relatively mature technology, the recommendation system is still one of the research hotspots today. However, the existing recommendation algorithms only pay attention to the quality and popularity of the recommended content, and often appear The recommended content is similar but not the same, that...

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the problem of repetitive and delayed content recommended by existing recommendation algorithms, the purpose of the present invention is to provide a knowledge map data mining and recommendation method based on dynamic suboptimal minimum spanning tree, which can simultaneously promote the importance and variety of content revolutionary method, and can reduce the user's waiting time

Images