Construction Of Business Knowledge Graph Based On Deep Learning

Knowledge graph, as a visual tool to display the structure of knowledge relationship, can use a visual way to show people knowledge by processing the knowledge resources. With the technology innovation, it has been involved in business area.However, the application of knowledge graph in business area is limited into the aspects of search engine and FAQ systems. It has not been widely used in other business areas. The reason can be find in the construction process of knowledge graph. There are three main steps in the construction of knowledge graph:the extraction of knowledge unit, the extraction of relations between knowledge units and the drawing of knowledge graph. And the former two steps are especially important. This article will focus on the knowledge extraction of knowledge graph and use deep learning to solve these key issues. This article also use graphical database to provide reference for drawing commercial knowledge graph.Deep learning is a hot topic in the area of machine learning, it can better simulate the human brain and achieve better learning effect relative to the shallow learning models. Deep learning algorithm usually has more hierarchical structure, it can realize the approximation of complex function by using deep and nonlinear learning structure. Besides, it reduces the effects of artificial feature selection on the accuracy of the model, which makes it has strong generalization ability.This article involves deep learning algorithm in machine learning area and use it to solve the difficulty of extracting knowledge unit. In addition, this paper also introduces graphical tools to construct knowledge graph and provides reference for drawing commercial knowledge graph.