Research On Employer Portrait Of Campus Recruitment Based On Big Data

In recent years,with the rapid development of China's higher education,the number of college graduates soared year by year and employment pressure is becoming more and more serious.As an important bridge connecting universities with society,campus recruitment plays an important role as well as facing more and more challenges and deficiencies.On the one hand,there is great information asymmetry among universities,students and employers.The complexity of the data has increased the cost of acquiring information for universities and students in the era of big data,as a result,universities and students can not get the full range of information from employers.On the other hand,universities can not provide targeted individualized employment services,failing to achieve a precise match between students and jobs.In this paper,we studied the problems of campus recruitment,by building an employer portrait.Firstly,we study the characteristics of data and user portraits in the context of big data,and discuss the necessity of constructing employer portraits and the feasibility of applying user portraits to the field of campus recruitment.Then,we study the overall frame of building employer portraits,and design employer portrait dimension and label system.Based on the practical application of user portraits,we put forward the method and process of employer portraits building.Finally,we take the rating label of employer portraits as an example to introduce the details of the principles,techniques and steps of extracting tags by means of machine learning modeling.In addition,we validate the effect of machine learning modeling to label extraction in the form of simulation experiments,and optimize the model.The thesis studies the feasibility,necessity and urgency of portraying employer portrait of campus recruitment in the background of big data,and the construction of employer portrait is of great significance to the solution of the two major problems existing in campus recruiting.We proposed the method based on improved fuzzy clustering and generalized regression neural network to extract the employer's value rating label,which not only implements the modeling process of machine learning to solve the problem,but also resolves the subjectivity of traditional dependence on expert experience to determine customer value.