Study On The Construction Of Learners' Model In Teachers' Training Community

With the development of China's education in full swing,the importance of helping teachers enhance their professional and quality and comprehensive competence will be more heightened in the future.It is therefore imperative to build teachers' training systems both catered for the practical needs of teachers and supported by advanced technology and methodology.As the wide use of Internet technology shows its great impact in all aspects of education,technology-assisted teaching will become a common practice and important impetus for the development of education.As a new trend of teacher training,online training community has become the focus of discussion as soon as it emerges in recent years.Up to now it has achieved remarkable results in its implementation in pilot cities,with considerable practical results."A" Teachers'Training Platform is a mainstream platform with great influence among current teachers' online training platforms.It is based on the two pillar functions of distance training and network research,and provides platform support for teachers to conduct collective research,learning exchange,group teaching,research discussion,exchange of teaching experience and exchange of classic cases.Moreover,with many years of operation and users' tests,the platform has stored statistics on each member's own personal information and related learning behaviors,which are included in its massive database.With the help of massive multi-dimensional data,we can make a comprehensive induction and integration,and further sketch a more targeted and practical picture of learners,which can directly display the clear information and characteristics of learners.It is of great significance to optimize the efficiency and better experience of this system,so as to lay a solid foundation for maximizing the satisfaction of scholars and teachers.Therefore,the overall goal of this study is to support scientific decision-making and efficient management of the "A" system,and select as the research theme an important topic that "A" cares about in the near future-the learner portrait of "A" teacher training platform.First,summarize and sort out the platform data,and preprocess it,based on the actual business needs,establish a more comprehensive data aggregation table,so as to further design data assets evaluation report;secondly,the principal component analysis method is used to design a comprehensive evaluation system around the actual situation of the students,and then experts are invited to fine tune the accuracy of the system and the parts to be optimized;thirdly,the author tries to select k-means algorithm,cluster analysis based on the platform students,and carry out targeted analysis on the specific situation of each student dimension according to the study.Finally,the collaborative filtering algorithm is used to further improve the automatic recommendation of platform counterpart teaching resources and teaching companion courses,recommend learning partners with high relevance for users,timely recommend the information of research activities,and verify the recommendation effect according to the final data.The research results are as follows:first,to further provide theoretical guidance for the big data burying and collection of "A" platform,to build a more advanced and scientific data aggregation table,and to establish a more targeted and comprehensive data warehouse;second,to build a three-dimensional and diversified evaluation model that can perfectly fit the logic and steps of learning behavior;third,to help platform teachers and management directors to be more direct in understanding the actual situation of learners,further recommend counterpart learning services and learning partners for learners,truly ensure the rapid progress of the course,capture the hot spots of public demand,and give learners a more targeted and practical course learning system;fourthly,with the help of the research results,it can contribute to the empirical research of the same type of data mining and the optimization mode of similar fields.