Research Of Video Teaching Evaluation Program Based On EEG And Cognitive Load

The traditional teaching evaluation mainly stands in the perspective of teachers’ teaching,and the main evaluation criteria include teaching objectives,teaching contents,teaching achievements,etc.As the main body of learning,students’ classroom learning status should be paid enough attention.Therefore,many experts and scholars try to improve the traditional evaluation standard and advocate the evaluation orientation of learning evaluating teaching.At present,the evaluation of students’ learning state in the teaching process mainly depends on the subjective questionnaire method,but this method is easily affected by subjective factors,and the results are often unconvincing.With the advancement of brain science and educational psychology,this situation is expected to be improved.Over the last few years,brain computer interface technology has been applied in educational scenes more and more widely because of its advantages of directly extracting and analyzing students’ learning state.However,most of the schemes have limited influence for teaching design due to the lack of exploration on individual differences of students and less discussion on the relationship between learning efficiency and cognitive state reflected by EEG,and so on.Therefore,this thesis is committed to further improve the application value of EEG technology in the actual teaching scene.The scheme adopts the EEG equipment with research and practical value to collect students’ EEG data and fully considers the individual differences of students from the perspective of students’ cognitive load in the process of watching teaching video.The thesis also establishes the mapping relationship between cognitive load and efficiency,and finally the effective evaluation of the video teaching effect is realized on the basis of fully analyzing the characteristics of the learning materials.In order to identify the learning state of students in the video teaching scenes,this thesis needs to complete the training of the model in a quantifiable experimental environment,and then applies the model to the actual scene.Therefore,this thesis selects basic calculation problems as the experimental environment for model training,and completes the following tasks:1.A classification and identification method for individual cognitive load levels is proposed.By designing an experimental paradigm for classification and identification of individual cognitive load,subjects complete increasingly difficult computational problems under a certain incentive mechanism.The proposed scheme utilizes Support Vector Machine(SVM)to train samples,and the average classification accuracy of cognitive load level reaches to 86.8%.2.Establishing an accurate mapping from individual cognitive load level to efficiency level,and designing the best cognitive load discrimination experiment by analyzing the relationship between cognitive load and efficiency.After that,the acquired EEG data are processed to obtain the corresponding relationship between the cognitive load level and the efficiency value distribution,and to find the best cognitive load level for each subject,which provides a theoretical basis for subsequent research.3.Designing and implementing a video teaching evaluation scheme based on the best cognitive load judgment method.This scheme first needs to analyze the teaching video materials to locate the time period corresponding to the key knowledge points,and then collects the EEG data in each time period.Then,based on the established mapping relationship between the cognitive load level and the efficiency value distribution,the thesis combines with the subjective evaluation results of the subjects to obtain the actual learning state of the subjects,and finally through the statistical analysis of different learning status of the subjects,the thesis completes the scoring of the teaching video and gives suggestions for improvement.Compared with subjective evaluation,this scheme has a more objective evaluation angle and a larger number of evaluation samples,which can provide a more valuable reference for teaching design.