Research On Learning Anxiety Recognition Andregulation Technology Based On Galvanic Skin Response Signal

In the field of psychology and pedagogy,learning anxiety has been paid more and more attention by scholars and researchers,but there are still many problems.At present,no equipment can accurately and quantitatively reflect the existence of individual learning anxiety,which mostly is characterized qualitatively on learning anxiety by means of questionnaire and physiological examination,in lack of reliable methods and techniques for quantitative analysis.At the same time,the lack of improvement in rea l-time monitoring and effective regulation strategies on learning anxiety,which does not make the regulation accurately grasp the individual fitness and acceptance,have greatly reduced the operability of emotion regulation.Because of the objective reality of physiological signals,the recognition of learning anxiety based on physiological signals has become an important research direction in the field of affective computing.This paper designed an experimental scheme to collect signals skin of subjects,and proposed an improved particle swarm optimization algorithm for feature selection,and used the BP neural network as recognition model.In the aspect of emotion regulation,referring to the emotion regulation model of Gross,this paper put forward the regulation model of learning anxiety in the interactive environment,finally,according to the research results,we designed and developed a learning anxiety recognition and regulation assistant based on Android platform.Specific work is as follows:(1)Experimental data acquisition and feature extraction.According to the characteristics of Shimmer3 and GSR,the paper designed an experiment to collect the GSR signals of learning anxiety,The experiment set two group experimental scene,one is for simulation of the foreign language classroom environment,it is a collection of learning anxiety of 10 min data for subjects;The other is to watch the relaxed environment of the video,it is mainly collecting in the normal state of 10 min data.Then,the data preprocessing,which is based on the response of the subjects during the experiment,is used to intercept the signal of the 20 S subjects under the state of learning anxiety and non-learning anxiety.In the experiment,43 subjects participated in the experiment,after the intercepted and screened,the final test data of the 35 subjects were qualified,that is,the experimental data of the 35 learning anxiety signals and the experimental data of the 35 non-learning anxiety signals.Thus,70 original sample data are produced.Then,the original samples are de-noised by wavelet transform;at last,the statistical characteristics of 30 time domain and frequency domain are extracted for each sample.(2)Feature combination optimization and identification of learning anxiety.this paper adopted the discrete binary particle swarm optimization algorithm(BPSO),and from the aspects of enhancing the diversity of particles,improving the convergence speed and jump ing out of local optimum to improve the discrete binary particle swarm optimization algorithm.In the process of learning anxiety recognition,the BP neural network was used as the identification model,and the fitness function of the optimization was determined.Finally,the results of feature optimization and recognition of learning anxiety were given.The experimental results showed that the best subset of features chosen by the improved particle swarm algorithm in the BP neural network,the convergence was better,and recognition rate was higher.(3)Establishing the regulation model of learning anxiety.This paper established a model of learning anxiety regulation based on Gross emotion regulation model,this model can adjust from environmental control,attention change,user cognitive reappraisal and express ion suppression and other aspect of the comprehensive consideration of learning anxiety regulation.(4)Design and implementation of learning anxiety recognition and regulation assistant based on Android.Using the portable GSR acquisition equipment to acquire GSR data to the user's mobile phone,and according to the theory and regulation and recognition on learning anxiety before,the paper designed and implemented an App for monitoring the learning anxiety of user,If the user has learning anxiety,the APP should adjust user to normal state.