Research On Evaluation Index System Of Deep Learning For College Students In Blended Learning Environment

Under the background of the knowledge economy,cultivating innovative talents in the21 st century that can meet the challenges of the future work and life is the top priority in China's current education reform.The implementation of deep learning is becoming a powerful“handle”that will promote the development of education in the future and improve student's ability to think in high-order and solve problems.At the same time,under the impetus of information technology,the construction of the blended learning environment is becoming more and more mature.It provides powerful environmental support,technical support,resource support,and interactive support for learners' deep learning.Therefore,practical and effective promotion of learners' deep learning in the blended learning environment has become a new direction for current deep learning research.From the current point of view,the practical study of deep learning has received extensive attention,but research on deep learning evaluation systems is still relatively inadequate.This has led most experimenters to adopt a general evaluation method to evaluate learners' deep learning,which is not conducive to the effective development of empirical research of deep learning.Based on this,this paper constructs a deep learning evaluation index system in the blended learning environment,in order to provide some references for the theoretical and practical research of deep learning in China.The main content of this article is as follows:Chapters 1 and 2,introduction and related research review,on the basis of clarifying the research background,research contents and research significance of this topic,through a large number of literature studies to understand the research status of deep learning and deep learning evaluation at home and abroad,and combing the concepts and characteristics of blended learning and deep learning.This will provide a solid theoretical foundation for the construction of the deep learning evaluation index system in the blended learning environment.Chapters 3,deep learning evaluation index system construction in the blended learning environment.This chapter establishes the principles and process for constructing the indicator system.After detailed analysis of deep learning under both orientations of the process and results,this paper redefines the connotation of deep learning in a mixed learning environment and preliminarily determines the four dimensions of the evaluation index system: deep learning motivation,deep learning input,deep learning strategies and deep learning results,then combined with the existing evaluation index system,specifically designs the evaluation indicators for each dimension.Chapters 4,Development of a deep learning assessment scale in a mixed learningenvironment,and reliability and validity tests.Firstly,after combining with the results of the open questionnaire,after conducting multiple interviews with experts,some evaluation indicators were revised and adjusted.Then,an initial measurement table was prepared according to the principle of compilation of scales,and questionnaires were issued and collected by undergraduate students of Liaoning Normal University.Next,carry out project analysis,exploratory factor analysis,etc.on the initial measurement table data in the SPSS19.0 software,and form an official scale after deleting unrelated questions.Finally,the reliability and validity of formal scales were tested in SPSS19.0 and AMOS17.0 software.Chapters 5,application and effect analysis of deep learning evaluation scale in blended learning environment.This chapter further examines the effectiveness of the deep learning assessment scale in the blended learning environment by comparing the application results of the scale with the interview results.Chapters 6,research conclusions and prospects,this chapter summarizes the research results of the paper,points out the deficiencies in the research,and puts forward the direction of future research.