Construction of Visual Question and Answering System Based on Knowledge Graph for Specific Objects

As China’s industry enters the 4.0 digital era, the segmentation of industrial applications is diversified. The knowledge of specific objects in the industrial scene is scattered that brings the pain points for management and operation. Knowledge graph (KG) has a powerful ability for relationship expression and real-time feedback that can integrate multi-source heterogeneous knowledge and provide data support for the Question Answering (Q&A) system. In this paper, a semi-automatic construction scheme for cupping machine knowledge graph is proposed and a visual Q&A system based on KG by the rule-based method is constructed. In the implementation of the KG, we design an ontology schema of the cupping machine and obtain entities and relationships from the network. This work solves the problem of the data island of cupping machines and provides a new technical method for the knowledge management of specific objects in the industrial equipment field.