Resaerch And Application Towards User Professional Portrait With Social Network

Social networking platforms have become an integral part of Internet services.With the explosive growth of the number of users,the data scale of social networks continues to expand,which makes user portraits a research hotspot.At present,the research on user portrait technology has already had very rich results in the fields of ecommerce and news media.However,the existing research work lacks the research and application of user occupational portraits.This thesis analyzes the user's occupational adaptability,occupational skill clustering characteristics and related applications based on professional social network data.The main contents are as follows:(1)Based on professional social network data,the user's career adaptability is analyzed.This article first collected 64,442 profiles of Linked In members,each of which contains the user's personal interests,professional skills,work experience and other information.Then all the interest and skill data filled out by the user are standardized,and 214 high frequency interests and 4764 high frequency skills are screened by synonym clustering.Establish an association analysis model and mine multi-dimensional association rules which reveals the relationship and dependence between different interests and skills.The relationship between the human interest and skills and career adaptability is discussed,and the impact of these association rules on user's occupational adaptability is verified.Finally,the use scenarios of occupational adaptability analysis are listed,and the application value and development prospects of the experimental results are discussed.(2)The clustering characteristics of the occupational skill life curve are studied.The thesis refines the collected user data and counts the degree of recognition of occupational skills in the user groups of all age groups,thus generating the life curve of each skill.Then,a hybrid clustering algorithm combining density peak clustering algorithm and K-Means clustering algorithm is used to cluster the occupational skill life curves,and the vocational skills with similar development trajectories are mined.Two sets of experiments were designed to verify the effect of the hybrid clustering algorithm.(3)Explore the application of social network user portrait technology.This thesis introduces the character portrait system based on massive social network data,and explains the project background,overall structure ad some principles of the system.The application of occupational adaptability research in the user portrait system is expounded,and the partial visualization effect of the system is demonstrated.The association rules of interests and professional skills and the results of occupational skill clustering are applied to the professional recommendation algorithm.The comparison experiments prove that the association rules and clustering results can solve the data sparsity problem of the recommendation algorithm to a certain extent,and improve the accuracy of the career recommendation algorithm.The research results in this thesis can prove that there are a lot of associations and clustering features between human interests and professional skills.These data features can be applied to practical application scenarios such as professional recommendation systems and user portrait system based on social network big data.In addition,the research results of this thesis can be used as reference for the research of social network user attribute mining.