Research On Feedback Network Teaching Assistant Strategy Based On Attention

Since entering the era of intelligent education,with the continuous development of educational technology,online education has received more attention in various types and levels of education and teaching fields due to its significant advantages such as the richness of learning resources,the openness of time and space,and the promotion of fairness in education.The development of learning forms has shown a trend of diversification.Among them,the learning form of live video teaching is the closest to that of traditional classrooms among various online education forms,resulting in the highest acceptance of this method by teachers and students.However,there is still a lack of feedback from online live teaching,resulting in the inability of teachers to change teaching strategies in real time by observing the learning status of learners.Learner concentration is one of the important teaching feedback in the process of teaching implementation,which can most intuitively reflect the learners’ immediate learning state.Based on this,further analysis of the phased change trend of learning state and class learning atmosphere can be conducted to obtain teachers’ teaching strategies in this situation,which can better promote the implementation of teaching activities.In order to improve the lack of teaching feedback in online teaching and assist teachers in class management during the live teaching process,this study aims to improve the teaching quality in video teaching,and from the perspective of intelligence and intelligence,studies the feedback based online teaching assistant strategy based on attention.The main work and innovations are as follows:(1)Based on facial expression recognition technology,through improving the VGG model,improve the accuracy of this technology on the experimental set,and combine with the video shooting angle and learner attention during the learning process.From the comprehensive consideration of the above three aspects,construct an EVA learning state evaluation model to quantitatively evaluate learners’ learning focus during the live teaching process.(2)Based on the constructed EVA learning state focus evaluation model,video teaching feedback algorithms are constructed from two dimensions: individual and overall,real-time and phased,to achieve automatic evaluation of teaching feedback,and to support subsequent generation of different teaching decisions for teaching individuals and teaching overall.(3)Based on the different performance of teaching feedback in different scenarios and different decision-making strategies related to them,a feedback based teaching decision support strategy is constructed.An experimental class is set up in a vocational college for trial use,and the experimental class is compared with the ordinary class on multiple different teaching indicators.The experimental results are analyzed to evaluate the effectiveness of the teaching decision support model.The experimental results show that the teaching decision support strategy designed in this article has a certain promotion effect on the teaching quality of the online live streaming teaching form.