Link Prediction with Supervised Learning on an Industry 4.0 related Knowledge Graph

Abstract

Industry 4.0 requires the integration of many actors to provide correct, personalized, and quick answers to customers. In order to meet this integration, data coming from different actors demand to be semantically integrated and harmonized. In these settings, knowledge graphs have proven to be successful in the task of semantic data integration of distinct data silos. Despite the increasing adoption of knowledge graphs in the Industry 4.0 domain for integrating and harmonizing data, still, all the power of the integrated data is not exploited. In this article, we tackle the problem of knowledge graph completion presenting an approach that applies supervised machine learning algorithms on top of the knowledge graph. In general, observed results indicate that supervised machine learning algorithms perform with an AUC of more than 88%. These outcomes suggest that knowledge graph completion enables to unveil new relations by connecting entities in the knowledge graph. Thus, the discovered relations in the knowledge graph bring added value to the Industry 4.0 domain.