Research And Application Of Knowledge Graph For Achievements Transformation

In recent years,the output value of China's scientific and technological achievements has been continuously increasing.However,the conversion rate of achievements is still not high.Most of the achievements of universities and research institutes have not been effectively transformed.The crux lies mainly in the asymmetric information between the two sides.Enterprises as demand side hope to find experts to solve technical problems,but suffer from unable to know the research ability of experts.The lack of relevant information of these experts led to enterprises unable to make an objective evaluation and judgment on an expert,so it is difficult to find suitable scientific and technical personnel.In this paper,Knowledge Graph technology is used to realize the deep excavation of big data of science and technology,provide more information to the demand-side of the achievement transformation,and further promote the transformation of scientific and technological achievements.Knowledge Graph describes the real-world concepts,entities and their relationships in a structured way.In recent years,Knowledge Graph have been widely used in semantic search,knowledge quizzes,big data analysis and decision-making.Their strong ability of knowledge management has gradually been highlighted.The research and application of Knowledge Graph have gradually spread from general fields to various vertical fields.The Knowledge Graph of this paper is collect various types of scientific and technological literature data,use the data sorting and correlation,knowledge mining,visualization,make it possible to provide a one-stop search for diversified literature and corresponding information display,improve the efficiency of enterprises find experts,speed up the transformation of achievements,and contribute to the strategic goal of building an innovation-oriented country in our country.The main research work of this paper is as follows:(1)Completed the construction of Knowledge Graph based on massive scientific and technological literature and put forward a set of feasible construction methods of Knowledge Graph oriented to achievement transformation.On this basis,according to the characteristics of the knowledge calculation,the knowledge calculation is abstracted into three categories: real-time approximation knowledge calculation,offline full-quantity data batch processing knowledge calculation,and graph structure knowledge calculation.(2)For the real-time approximation knowledge calculation,the Hot research of regional research is taken as an example to illustrate how to achieve real-time approximate knowledge calculation use online aggregation.At the same time,an optimized sampling algorithm is proposed to further improve the online aggregation processing efficiency.For the offline full-quantity data batch processing knowledge calculation,the Expert multidimensional ability profiling is used as an example to illustrate how to use Hadoop framework to implement offline batch processing knowledge calculation.Finally,for the graph structure knowledge calculation,an Expert relational network mining is used as an example to show how to use Neo4 j to implement knowledge calculation based on graph database.(3)The above knowledge graph construction and knowledge calculation research results were applied to the search and recommendation platform of scientific and technological talents to verify the validity and feasibility of the research results in this paper,and to display multi-dimensional scientific and technological resource information for users in a visualization,and to promote the transformation of achievements.