Supplier selection method based on knowledge graph

Abstract

The invention discloses a supplier selection method based on a knowledge graph. The method comprises the following steps: S1, constructing a purchase data set; S2, constructing a knowledge graph G; S3, mapping the knowledge graph G to a low-dimensional embedding space to obtain a low-dimensional feature vector of each entity in the knowledge graph G; S4, calculating relevancy scores of the new purchasing demand u'and other entities in the knowledge graph G based on the low-dimensional feature vectors; and S5, summing the relevancy scores of the purchasing demand u'and other entities in the knowledge graph G to obtain global relevancy scores, and sorting the global relevancy scores to obtain N suppliers before sorting. According to the supplier selection method based on the knowledge graph,the knowledge graph is applied to supplier selection, so that selection of suppliers is more scientific and accurate, selection of the suppliers is guided, dependence on manual experience is reduced,and supplier selection efficiency is improved.

Description

technical field [0001] The present invention relates to the technical field of deep learning, and more specifically, to a supplier selection method based on a knowledge graph. Background technique [0002] In the network chain structure model of the supply chain, a large number of suppliers are gathered upstream of the core enterprise. Suppliers play an important role in improving product quality, service quality, and customer satisfaction for core enterprises. A cooperative relationship of "sharing weal and woe" has been formed between core enterprises and suppliers. Therefore, the selection and evaluation of suppliers is at the core of the enterprise's procurement business activities and even the entire supply chain management. [0003] The selection of suppliers under the traditional model is mainly based on the comprehensive consideration of two factors: hard indicators and soft indicators. Among them, hard indicators are individual indicators, which mainly examine obj...

AI Technical Summary

Problems solved by technology

[0005] Due to the lack of scientific basis in the process of supplier selection and evaluation by qualitative methods, there are a large number of subjective factors in the process of supplier selection; it is difficult to influence certain indicators in supplier selection and evaluation in quantitative methods, such as brand influence, Cooperative spirit, quality management level, etc., are difficult to describe with quantitative objective data; and the method of combining qualitative and quantitative methods often pursues the perfection of the model and the skills of solving, which makes the selection and evaluation of suppliers too cumbersome. Ineffective phenomenon

Images