Supplier recommendation based on knowledge graph embedding
Cixing Lv, Yao Lu, Xiaohui Yan, Wei Lü, Hua Tan
Abstract
Selection optimal suppliers is an important issue for supply chain management. Cloud manufacturing and other new digital manufacturing paradigms pose challenges to supplier selection due to high dynamic characteristics, but also in turn provide new opportunities for improving supplier selection by usage of data. Knowledge graph has been widely researched in recommendation system and achieved remarkable results. And knowledge graph will also play an important role in improving supply chain management. In this work, a novel approach is proposed to learning purchase demand-procurement records property specific and global relatedness from supply knowledge graph based on knowledge graph embedding. And then we use the relatedness features to predict top-N procurement records which are most related to purchase demand. Finally, we conduct a numerical example to demonstrate the practicality and effectiveness of our approach.