Construction of sustainable development professional cluster in private universities based on data mining model

Abstract

Abstract Professional cluster construction is the change of professional development mode in the context of current social development, and it is the innovation of internal construction and intensive management of private colleges and universities. This paper analyzes the advantages and disadvantages of the maximum expectation algorithm in big data mining, proposes to optimize the parameter estimation iteration by using the Monte Carlo method for the problem that the stable point of the EM algorithm does not necessarily reach the global optimum, uses the sample mean instead of the overall mean, and further proposes MCEM acceleration algorithm to improve the convergence speed of the parameter estimation iteration. The evaluation score of the audit professional group construction in internal business process management of school A is 3.21, that of the software technology professional group construction in the learning and growth dimension of school B is 3.45, that of the Chinese medicine professional group construction in developing core competitiveness dimension of school C is 2.87. The comprehensive evaluation score of the construction of the industrial robotics professional group in school D is 3.33. The mining evaluation of professional cluster construction based on big data can precisely locate the key factors of professional cluster construction in private colleges and universities, which is of reference significance for the sustainable development of professional cluster construction in colleges and universities.