The Design And Implementation Of Knowledge Extraction Service For Constructing The Knowledge Graph Of The Financial Domain

Knowledge Graph is an important technology for knowledge management and application in the era of big data,it has become a key technology for search engine and knowledge-based reasoning and decision-making in various fields.As an important member of the Semantic Network,Knowledge Graph makes large-scale knowledge storage more standardized and search more efficient.In recent years,the rapid development of Knowledge Graph in information retrieval has brought new opportunities for financial domain tasks such as enterprise relationships search and enterprise risk events transmission.Using unstructured text to construct Knowledge Graph requires the technology on Knowledge Extraction.Therefore,the thesis proposes a topic research on the Knowledge Extraction Service for constructing the financial domain Knowledge Graph.Based on the existing technologies and algorithms,this thesis studies the knowledge extraction method of constructing financial domain Knowledge Graph.The thesis expounds the development background of the Knowledge Extraction Service System for constructing the Knowledge Graph of the financial domain,explains the importance of Knowledge Extraction Service in the whole Knowledge Graph construction process,does a detailed analysis of the design of each module of Knowledge Extraction Service,introduces the concrete implement of each module,finally it is packaged into Knowledge Extraction Service to solve the problem of knowledge extraction from unstructured text data.Based on the model design and specific usage scenarios,the thesis combines two different Named Entity Recognition model applications and one Relation Extraction model application,packages the models and applications into a Docker container and deploys to the Kubernetes cluster to build an Knowledeg Extraction Service System.The Knowledge Extraction Service System implemented by the project has been officially launched in the company's Knowledge Graph product,the program of this thesis can not only be used to construct the Knowledge Graph of the financial domain,but also plays an important role in the Knowledge Graph update.