CSDN User Portait Research

With the era of big data,the basic information and behavior information of users are flooding the entire network,such as user addresses,user shopping information,and user comments.Through collating,analyzing and mining a large number of user data,in order to obtain useful hidden information from them,the value of the data is the focus of current research.At present,some enterprises have used existing user data to analyze user commonality,form user tags and construct user portraits.Help enterprise achieve user research,precision marketing,personalized service,and business decision-making through user portraits.CSDN is a Chinese IT technology exchange platform,which provides a way for domestic Internet practitioners and enthusiasts to spread knowledge.The platform has a large number of users,and most of them are IT technology practitioners.At present,it is very difficult to recruit IT technicians in enterprises,such as the lack of suitable resumes for recruiters,and the failure of interviewers to meet the expectations of enterprises.The user profile has little research in the field of corporate recruitment,and the CSDN platform has a large number of user data of IT practitioners.In order to ease the status quo of hiring IT staff,from the perspective of corporate recruitment,this article builds a user portrait based on CSDN user data.The purpose is to provide a reference for corporate IT staff recruitment.The main work of this article is as follows:(1)Data collection and processing: Through the analysis of CSDN web pages,the user data crawling process is designed,and the crawler program is implemented.Store the collected user data in database tables and perform data processing to ensure the accuracy and standardization of user data and provide a data basis for user portraits.(2)User label mining: Based on the collected user data and analysis from the perspective of corporate recruitment,two labels required by the enterprise were designed,including user type labels and learning category labels.The user type label is a comprehensive evaluation of users.The K-means ++ clustering algorithm based on Mahalanobis distance is used to mine the label and obtain the user type label.The users of the label are divided into general,good,and excellent users.The learning category label is a label that identifies the user's technical direction.The graph is classified by the graph convolutional neural network and the long-short-term memory network algorithm,according to the categories of articles published by users,the users are marked with learning category tags,used to determine the level of the user's category.The mined user tags will provide the basis for tagging user portraits.(3)Establish a user portrait and visualization platform: Based on the label of the enterprise's needs,combined with the user's basic attribute label to create a user portrait.According to the established user portraits,design and implement a visualization platform to provide services and references for corporate recruitment.