Research on Computer Natural Language Processing Intelligent Question Answering System Based on Knowledge Graph

The knowledge graph question answering algorithm is deeply discussed, which provides strong technical support for improving the accuracy of knowledge graph question answering algorithm. Based on the semantic similarity of the problem and the relation, a problem embedding model combining the two characteristics of words and words is established. Then, in order to improve the inference performance of this method on long links, a convolutional neural network for extracting higher-order vector information with embedded vectors is established. The medical knowledge graph question answering system is taken as a case study. This project intends to use Apache jena storage, RDF triadic representation, Jieba segmentation, Echarte and other methods to build knowledge graph. A pharmacologic knowledge answering system based on Apache jena was established by combining Python Django and Apache fuseki library. The experiment proves that this system can realize the interactive answer between doctors and patients, effectively improve the diagnosis and treatment effect of patients, and also bring great convenience to patients.