Research On The Analysis And Prediction Model Of Human Resource Turnover In China’s High-Tech Enterprises

High-tech enterprise is a kind of intellectual technology-intensive economic entities,which is an important engine for realizing the high-quality development of China’s economy.Human resource(HR)turnover is an important issue facing by Chinas high-tech enterprises.The explicit and invisible cost and risk are important constraints of the high-quality development of China’s high-tech enterprises,industry,and the overall economy.HR retention is a key point for solving the problem of HR turnover.The retention decisions driven by big data can improve the insight,foresight,and wisdom of HR retention management in an enterprise,thereby improving the efficiency and reducing the turnover rate.The models of HR turnover prediction and retention analysis based on machine learning are a kind of management and decision support tools of HR retention based on big data,which can provide decision-making reference for retention business problems about who retained and how to retain.However,most of the relevant research on HR turnover prediction focuses on forward-looking and accurate identification of potential HR turnover,rather than targeted and differentiation retention strategies.Therefore,this article is of great theoretical and practical significance to be carried out on the basis of the existing research on HR turnover prediction modeling,drawing on the classic theories and research results related to HR turnover analysis,and closely combining the characteristics of China’s high-tech enterprises.First of all,this article sorts out the relevant research background and refine the actual problems of this study.On this basis,aiming to solve the realistic problems,a wide range of investigation and analysis of related theoretical research is conducted to focus on scientific issues.The main research content and technical route of this article is thus concluded.Then,combining with the reality and scientific issues,the values and significance of this article is analyzed from the aspects of practice and theory.Following,according to the CRISP-DM(cross-industry standard process for data mining),the research framework of corporate HR turnover analysis and prediction based on data mining is constructed,solve the matching problem of the retention strategy from the level of modeling process and path planning.Secondly,under the leadership of the research framework,the PCA(primary component analysis method)and CLARA(clustering large application)methods are applied to design the portrait label refining methods of high-tech enterprise HR turnover,solving the matching problem of the retention strategy from the level of model design.On this basis,the portrait-based analysis model of high-tech enterprise HR turnover is investigated to solve the development problem of the retention strategy.Moreover,on the basis of portrait labels,A portrait-based prediction model of high-tech enterprise HR turnover is constructed applying SMOTE-ENN(synthetic minority oversampling technique-evolutionary neural networks)and CART(classification and regression tree)to solve the problem of pre-identification of potential 1 HR turnover.Finally,the above research results are experimented,tested and verified using a high-tech enterprise public data set as a carrier.The results show that the relevant theoretical results this article raised are feasible and effective.Theoretical analysis and experimental studies show that this study is positive for enriching the relevance theoretical of HR turnover analysis,prediction,and retention strategy development,can provide scientific decision basis and intelligent support for the precise identification of potential HR turnover,the targeted development of retention strategies,and the different matching for high-tech enterprises.Moreover,this article is closely combined with related research results such as interpretation-oriented HR turnover analysis and inductive-oriented HR turnover prediction,and conducts related research according to the data mining process of data understanding,data preparation,model construction,model evaluation,and model application.In terms of theoretical aspects,this article aims to provide a case reference to research paradigms that combine the interpretation and inductive orientation,promoting the in-depth development of big data HR research,therefore promoting the improvement in the maturity of corporate HR data analysis in practice,and the development process of digital transformation of China’s enterprises.