Intelligent EMC Management Fabric Based on Knowledge Base and Machine Learning

摘要

In this paper, we propose a new knowledge fabric for electromagnetic compatibility management of mobile phone. It combines the knowledge graph and machine learning to achieve intelligent management. The fabric leverages knowledge base to store entities and relational triples of entities. Moreover, we firstly propose the conception of ”function triple”. By using recurrent neural network methods, the model recognizes entities and their corresponding attributes to reflect the real query intention. When evaluated on a mobile phone task, the accuracy of the fabric is 68.3%. It is practical useful to help engineers to carry out circuit design.