Research on medical automatic Question answering model based on knowledge graph

With the advent of the era of big data, knowledge interconnection has received extensive attention. As a core of this era, knowledge graph has broad applicationprospects. The application research in the intersection of big data and medical care will play an important role in solving the contradiction between the insufficient supply of high -quality medical resources and the continuous increase in the demand for medical services. Finding the medical -related knowledge you need from a large amount of data has become the core of this technology. At present, the research on medical knowledge graph is still in the exploratory stage. In the case of insufficient clinical data, how to obtain the data set is difficult. For the strong professionalism of knowledge in the medical field, the construction of knowledge graph should be targeted. How the system needs to accurately collect the questions that users want to ask is a big difficulty in the question answering system of deep learning. To solve the above problems, the data collection source is from the Tianchi Chinese data set Toyhom, the bert model is used for word segmentation and vocabulary construction, Neo4j realizes the organization and storage of knowledge, and the naive Bayesian machine learning method is used for intention recognition. Based on the above technology, the medical knowledge graph is built, and the visual Q&A window is completed. Finally, this kind of medical automatic question answering robot was realized.