Knowledge graph is a new data storage theory and architecture in big data era, which is expected to provide more efficient underlying data support for artificial intelligence, data mining and other key technologies. However, the work about open source knowledge graph on Chinese and distributed system is limited at present. Besides, building a knowledge graph with high quality requires a lot of labor and financial cost. Therefore, a modeling system that can improve efficiency and quality of the knowledge graph construction is urgently needed.
This graduation project has completed a knowledge graph modeling system based on distributed systems, providing technical support for building knowledge graph, including automated scripts, auxiliary editing platform, knowledge graph data visualization platform, distributed storage and intelligent question-and-answer platform. A series of measures have been taken to ensure the work of distributed systems, including communication protocols, monitoring systems, and distributed search strategies. Basing on the above knowledge graph, an intelligent question-and-answer robot is developed which is able to reply questions about medical advice. The evaluation results for the robot are used to assess the quality of knowledge graph in return. All the work in this thesis will eventually form a complete knowledge graph modeling system from the bottom to the top. The system will be optimized and tested.