The knowledge graph management platform for big data in technology consulting is a tool for quickly assisting experts and users to complete consulting services.Different from general big data,big data in technology consulting has the characteristics of massive,multi-source,and heterogeneous.The data includes technical achievement such as patent,papers and fund projects;technological entities such as enterprises,institutions,and technological personnel,and their activities;the relationship between the upstream and downstream of the industry chain,etc.For the above big data,a knowledge graph for technology consulting is constructed to support the services such as knowledge information acquisition,enterprise-related technical capability analysis,and industrial chain insight analysis.Although researchers have done a lot of research on the knowledge graph construction and platform,they still cannot meet the needs of technology consulting graph-based insight analysis.There are problems and challenges:1)Multi-source and heterogeneous data management problems:the loosely organized big data in technology consulting is difficult to share and directly utilize;2)The construction of knowledge graph:traditional construction methods are difficult to convert,save and integrate the structured or semi-structured data of technology consulting from various databases;3)The efficient query of large-scale graph data in technology consulting:because of the complex correlation logic and topology between the big data,the query calculation complexity is very high.In response to the above problems and challenges,this paper focuses on the research of the unified data resource management,knowledge graph construction,and the query optimization in large-scale knowledge graph,and the completion of the knowledge graph management platform for big data in technology consulting.The main research as follows:1)Design and implement a unified data resource management framework for technology consulting,solving the problems of multi-source and heterogeneous of the data;and design and implement the technology or industry label hierarchical system in the hotspots,which helps the multi-dimensional statistical analysis of science achievements.2)Propose and implement the method of domain knowledge graph construction based on property graph model,solve the problem of schema construction and knowledge graph data transformation for structured and semi-structured data in specific fields.3)Propose and implement query tasks and query optimization strategies in the large-scale graph of technology consulting,carry out reasonable query task from the three domains of organization,talent and technology,and design corresponding efficient calculation skills based on multiple optimization strategies to solve the query complexity in the big data environment.4)Design and implement a knowledge graph management platform for big data in technology consulting,use the data as the cornerstone,and integrate the multi-source data resource management,knowledge graph construction,storage,query and optimization.The platform provides simple and portable services for science and technology consultants.Finally,this platform was applied to the national key R&D subject"Technology consulting Data Resource System Research and Resource Construction".Based on the big data in technology consulting,the platform has constructed and applied a knowledge graph,which verified the effectiveness and value of the method and platform proposed in this paper. |