| The database of research experts is the foundation of research evaluation activities.However,during its construction,the diverse channels for data collection and varying data quality often lead to confusion when using expert information due to incomplete information.Additionally,a significant amount of research expert information exists in business information systems such as research project applications and technology award reviews.This thesis focuses on the key problem of how to extract this information and use it to supplement incomplete information in the research expert database while creating multidimensional portraits of experts.Based on an analysis of existing research expert information from multiple sources,this thesis designs an expert information completion algorithm,establishes a research expert portrait model,and develops a practical application system.The main accomplishments of this work are as follows:(1)To address the issue of incomplete information in existing expert databases,this thesis designs a set of expert information completion methods that extract expertrelated information from various business activities such as research project applications and technology award reviews.This method matches this information with research expert information in the database,not only completing missing information for existing research experts but also collecting a group of candidate experts.(2)The user portrait model is applied to the display of research expert information,constructing a research expert portrait label system.Using data from the expert database,multidimensional portraits of research experts are generated from basic information,academic information,research performance,and research direction.The research expert portrait space model is established,and the research expert portrait is visualized by calculating label weights and utilizing a tag cloud.(3)A research expert information service system is developed.Using the research expert information completion algorithm and portrait construction method proposed in this thesis and incorporating user needs,the system is designed and developed to manage research expert information,analyze portraits,and complete information.In summary,this thesis makes a positive contribution to improving the information in research expert databases.Additionally,the expert portrait function can effectively assist research management departments in selecting and recommending experts and conducting related work.The developed system has also achieved good results in practical applications. |
