Research On The Construction Of Profile In Teachers' Online Training Community Driven By Data

With the normal development of teachers' online training,the construction and evaluation of the online training community has become a hot spot for both education and academia.The traditional online training community evaluation method promotes the improvement of the training community function and service,but there are still problems such as weak evaluation decision-oriented function,biased evaluation process limitations,mechanical rigidity of evaluation methods,and single form of evaluation results.The development of big data analysis technology provides methods and guidance for the accurate evaluation of teachers' online training communities,profile as a data-driven evaluation tool,applied to the teacher's online training community can promote community evaluation results to become objective,formative,Integration and intelligence,has the important value of research and application.Based on the above background,this study mainly did the following work: Firstly,proposed the data-driven teacher online training community profile construction process,which mainly includes five Stage of "data collection,data preprocessing,profile model construction,profile model application,precise decision-making and intervention".Secondly,with the guidance of the big data research paradigm of data science,based on the 235 sample community data sets of the Zhejiang Famous Teachers Network,using principal component analysis and noun interpretation to extract the main characteristics of training community basic information,training resources,training activities and training effectiveness,including training community information,community moderator information,interactive generation resources,famous teacher resources,teaching practice activities,teaching reflection activities,member training results and community building results,forming a teacher online training community profile model.Then,based on the teacher training community profile model,150 training communities of Zhejiang Famous Teachers Network were selected as the application objects of the model.Cluster analysis found that there are three characteristic types and operation modes of the training community,including potential growth type,specialty development type and all-round outstanding type,Stepwise regression analysis found the influencing factors of the training effectiveness of the three types of training communities,generated intervention strategies for the training communities,and synthesized the above analysis results to generate a profile of the teacher online training community.Finally,design questionnaires and interview outlines to verify the perceived usefulness,perceived ease of use and attitude of community profile of the online training community,and obtain profile improvement suggestions.The research results show that training community profiles can accurately identify different types of training communities,and provide effective intervention strategies to assist community leaders in their accurate evaluation and implementation of policies,and have been highly recognized by community leaders.With the guidance of the big data research paradigm based on data science,this study completed the construction and application of the teacher's online training community profile model,and changed the community evaluation from a semi-empirical and semi-data evaluation model to a full data-driven evaluation model.To a certain extent,this study has promoted the innovation and development of the evaluation of training communities.