Research And Implementation Of Research Project Team Recommendation System Based On Expression Learning

In daily work,the team is the basic unit of work for completing the project.As the project work becomes more and more rapid and diversified,it requires a good social relationship between the team members,which can solve the problems in the work more efficiently and orderly.The traditional team generation problem has been extensively studied.With the advancement of technology,the popularity of various social platforms and the rapid development of communication networks,how to find an efficient team on the social network has become a research hotspot.By studying the strength of the relationship between heterogeneous information networks and team nodes,this paper proposes a team generation algorithm based on heterogeneous information network to better solve the problem of team generation.Firstly,this paper proposes a research on the relationship strength algorithm based on meta path(RSMP).The algorithm first obtains the vector representation of the nodes in the heterogeneous network according to the random walk of the meta-path,and then calculates the relationship strength between the nodes by calculating the degree of similarity between the nodes.Based on this,the team generation algorithm SkillsFirst is proposed.SkillsFirst transforms the heterogeneous network into a complete graph,and then builds the minimum spanning tree according to the skills required for the project work,and finally obtains the required team.Then the experiment is carried out on the DBLP dataset and the Wanfang dataset created by ourselves.It compares with some previous team generation algorithms in terms of correctness,communication cost and running time.The experimental results show that the proposed algorithm is effective and reasonable.Finally,the algorithm proposed in this paper is applied to practice,and the research project team recommendation system is designed and implemented.