Non-coal mine literature association recommendation method based on knowledge graph

Abstract

The invention discloses a non-coal mine literature association recommendation method based on a knowledge graph, and belongs to the field of non-coal mine literature association recommendation. In order to solve the problems that in the prior art, associated recommendation cannot be achieved through TFIDF, the number of model topics obtained through an LDA model is small, and data sparsity is large, the method includes: obtaining non-coal mine data, especially equipment information through distributed multi-thread crawlers and manual collection; processing the acquired data, then constructinga knowledge graph by adopting an LDA model, and respectively constructing initial knowledge models on the basis of the knowledge graph; and then introducing an activation diffusion model and adoptingan association distance to obtain a recommendation result. According to the method, the knowledge graph and the activation diffusion model are combined, data sparsity can be effectively relieved, multi-directional association recommendation accuracy is greatly improved, recommendation accuracy is improved according to recommendation selection result positive feedback recommendation calculation, and a non-coal mine literature recommendation method is optimized.

Description

technical field [0001] The present invention relates to the technical field of document association recommendation for non-coal mines, and more specifically, relates to a method for association recommendation of non-coal mine documents based on knowledge graphs. Background technique [0002] With the advent of the information technology era, smart construction centered on informatization, automation, and intelligence has achieved great success in the non-coal mining industry. It has been widely used, and also accumulated a large amount of equipment data, literature data, etc. How to store and utilize them is a problem that must be solved. Traditional relational databases can handle structured data well, but they are messy but interrelated. In this form, knowledge graph is an effective way to solve the above problems. [0003] The knowledge map is essentially a graph-based semantic network, which represents the relationship between entities and entities, with the purpose of...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the problems existing in the prior art that the use of the TF-IDF model cannot solve the association recommendation between models, the number of model topics obtained by using the LDA model is scarce, and the data is sparse, the present invention provides a non-coal mine literature based on knowledge graphs. The association recommendation method uses the LDA model to build a knowledge map; and introduces the activation diffusion model in the modeling process to solve the problem of data sparsity; in the process of calculating similarity, the method based on association distance is used to improve the accuracy of document association recommendation

Images