Design And Implementation Of Financial Service System Based On Knowledge Graph

As a highly information-sensitive industry,financial practitioners are in urgent need of comprehensive industry information so as to make important decisions in intelligent investment research and risk control.The rapid growth of financial information and a large number of unstructured financial announcement make it difficult for the financial research team to sift through such a large amount of information quickly.Based on this situation,it is important to provide high quality financial information service.Knowledge graph can integrate scattered information into structured high-value knowledge,and explore the relationship between information.It helps us to realize the intelligent domain information service.Therefore,based on knowledge graph technology,this thesis designs and implements a financial service system with information query function and financial announcement extraction function.The following studies are mainly made:1.This thesis studies the construction of knowledge graph in the financial field,and puts forward a complete knowledge graph construction scheme.According to the information requirements,a new schema layer of financial knowledge graph is designed.The financial knowledge graph is constructed by using the data extracted from the Internet.This research provides the foundation for the system implementation and the subsequent research of the financial knowledge graph.2.This thesis studies the information extraction algorithm of financial announcement.A financial announcement information extraction model is constructed by using the BiLSTM-Attention-CRF network.Due to the lack of public data sets,this thesis puts forward an process of data set construction and model training.The performance of traditional BiLSTM-CRF model is also tested to ensure the advantages and effectiveness of the proposed model.Finally,a financial announcement information extractor is constructed by encapsulating the trained model and complete the whole process.3.Based on the knowledge graph and the announcement information extractor constructed above,an information service system in the financial field is designed and implemented.For the need of processing structured and unstructured data,this system provides knowledge inquiry service and automatic announcement information extraction service.During the system implementation,the response results and system page are designed.The knowledge graph is visualized in the front end.In general,this thesis proposes a construction scheme of knowledge graph in the financial field,and uses deep learning technology to put forward an announcement information extraction algorithm in financial field.Combined with those two,this thesis puts forward the construction scheme of financial information service system for users to obtain financial information.In the end,the model and function of the system are tested,and the effectiveness of the system is verified.