Product configuration recommendation method and system based on graph relation mining

Abstract

The invention discloses a product configuration recommendation method and system based on graph relation mining, and the method comprises the steps: obtaining customer-product-material entity information, inputting the information into a trained FM model, and outputting a product configuration result; construction of an FM model: completing knowledge extraction of products, materials and components by using an actual business database and completing graph construction; recall of materials and components forming a product is completed through map path analysis; calculating the frequency of product configuration in an off-line manner, and substituting the frequency into the updated relationship among the product, the material and the component to obtain a knowledge graph after relationship optimization; training is carried out through an FM model based on the knowledge graph after relation optimization, model training of the product configuration recommendation method is completed, and a trained FM model constructed based on an enterprise business database is obtained. The method has the advantages that the materials are recalled based on the knowledge graph, the cold start problem can be relieved, the related materials and components of the product can be recalled to the maximum degree, and the recall range is controllable.

Description

technical field [0001] The invention relates to a product configuration recommendation method and system based on map relationship mining, which belongs to the technical field of recommendation systems and machine learning. Background technique [0002] Product configuration refers to the automatic and intelligent configuration of product core materials and components based on user needs or application scenarios. Enterprises such as Huawei and ZTE have established a relatively complete 5G product configuration line to quickly meet the needs of customers in multiple scenarios and customization, and provide digital support for enterprise technology standardization, production inventory, and R&D management. Realizing product automation and intelligent configuration has become a key capability indicator for the digital transformation of enterprises. [0003] The recommendation system is an information filtering system that uses machine learning and other technologies to predict...

AI Technical Summary

Problems solved by technology

Only when the knowledge map reaches a certain scale can it reflect the role in the recommendation system, but the introduction of a certain scale of knowledge map will increase the complexity of the recommendation system. Typical other problems include the mining of features in the knowledge map, etc.

Images