The Theoretical Method And Application Study Of Evaluation Of Small And Medium-sized Enterprises Growth

The comprehensive evaluation research tries to solve the problem of unified evaluation of complex phenomena reflected by different indicators.This method is widely used in government policy-making,business decision of enterprise and related academic evaluation research.In recent years,the rapid development of big data information technology has provided important support for the progress of comprehensive evaluation theoretical method in information diversification and data processing.Some societies are highly concerned,and the in-depth study on evaluation with extremely complex problems has become a hot topic again.The growth evaluation problem of small and medium-sized enterprises elaborated in this paper is one of them.Small and medium-sized enterprises are an important part of national economy and an important foundation for social and economic steady development.Especially after our country entered the economic new normal,it is an important part of the supply side structural reform and the government introduces a series of relevant policies to improve the small and medium-sized enterprise growth environment and support its innovation and development.Regardless of the government's implementation of the policy of support for small and medium-sized enterprises,the recognition of the policy effect,and the utilization of financial capital investment to support the development of small and medium-sized enterprises,it needs to be evaluated for its growth.Relevant research has important practical significance.The high complexity of enterprise growth evaluation is reflected in:the growth process is the result of many internal and external factors.The mechanism of each growth influence factor is different,and some factors may be decisive.The factors that affect growth and its mechanism are dynamically changing.At present,with the rapid progress of information technology and the globalization of market resource allocation,the complexity of enterprise growth problem is also increasing.Its related research has the significance of promoting the enterprise growth theory exploration.This study focuses on fully reflecting the small and medium-sized enterprise growth environment changes,taking serving the government and fostering financial capital market for small and medium-sized enterprises as the goal.It introduces relevant big data technology and develops theoretical method and application study of small and medium-sized enterprise growth evaluation,to further promote the development of comprehensive evaluation theoretical method.The relevant research framework is arranged like this.Chapter 1 Introduction.The background and significance of the research are given.Comb relevant domestic and foreign literatures;Put forward research idea,main method and frame;Be definite about innovative work.Chapter 2 The basis of theoretical method of the growth evaluation of small and medium-sized enterprises.Mainly include:1.Define the concept of small and medium enterprises;Interpret the external and internal factors that determine the growth of enterprises and conclude the growth mechanism.2.Give a statistical comprehensive evaluation method for index measurement integration and index weight distribution.3.Introduce web crawler,text mining technology based on naive bayes classification,two machine learning algorithms,Random forest and Adaboost and relevant big data technology;provide new ideas in order to solve the problems such as index measuring and evaluation result test in the growth evaluation method of small and medium-sized enterprises.Chapter 3 Research and construction on growth evaluation model of small and medium-sized enterprises.1.Establish an evaluation theoretical model based on the five elements including innovation ability,entrepreneur quality,product competitiveness,operational efficiency and environmental risk.2.Put forward the basic principles of evaluation index selection and the main idea of its index selection.3.Based on the rationality and completeness standard of measurement,construct the evaluation index system of growth value of small and medium-sized enterprises.Chapter 4 The research on the measurement and integration method of growth evaluation information of small and medium-sized enterprises.Specifically include:1.Further improve the expert group evaluation method based on two-stage learning to achieve efficient measuring of evaluation index.2.Establish an index measuring method based on text mining technology,and provide a new idea for index measuring.3.Further promote the dimensionless information integration technology of evaluation index.4.Establish an index weight distribution method based on the subjective and objective information optimization model.5.Establish the growth prediction classification model of small and medium enterprises based on machine learning by utilizing two machine learning algorithms,Random forest and Adaboost.Chapter 5 An empirical analysis of growth evaluation based on the data of small and medium-sized listed companies.Chapter 6 A review and a study prospect of the whole content in the form of a conclusion.This research seeks to carry out innovative work in the following three aspects.First,establish the growth theoretical model of small and medium-sized enterprises and the corresponding evaluation index system.Taking relevant enterprise growth theories and its current business environment changes as the foundation,set up a kind of growth evaluation model of small and medium-sized enterprises based on the five core elements and make every effort to fully reflect the growth trend and potential of small and medium-sized enterprises.Secondly,propose two evaluation index measuring methods.Provide technical choices in order to solve the completeness and operability of the growth evaluation index system of small and medium-sized enterprises,and the availability problem of some evaluation index data.Among them,1.The improvement of expert group evaluation method based on two-stage study can be a part of qualitative index data source which is difficult to directly measure through further effective description and integration of subjective preference information of expert group.2.Based on the index measuring method of text mining technology,using web crawler and text mining technology and by grasping the Internet data and using the semantic analysis method based on bayesian naive classification algorithm,it is converted into the ordinal data based on the collection of evaluation level to solve the data missing in the growth evaluation of small and medium-sized enterprises,or the measurement problem of sensitive evaluation index.Thirdly,build a growth prediction model of small and medium-sized enterprises based on machine learning classification algorithm.In the final evaluation results obtained by the growth evaluation method of small and medium-sized enterprises,there exist problems that the information is insufficient and the effectiveness is difficult to detect.This paper introduces the machine learning technology into small and medium-sized enterprise growth evaluation method,builds a growth prediction model of small and medium-sized enterprises based on two classification algorithms,Random forest and Adaboost,and tests the effectiveness of the evaluation method by increasing the growth classification information of object evaluation and comparing with the conclusion of comprehensive evaluation method.