A NOVEL METHOD FOR ENTREPRENEURIAL STUDENT MENTAL HEALTH ASSESSMENT USING EEG SIGNAL ANALYSIS ENHANCED BY GENERATIVE ADVERSARIAL NETWORKS

Objective: This study aims to evaluate the mental health of college students who have failed in entrepreneurship by analyzing their electroencephalogram (EEG) signals. Long-term negative emotions are closely related to mental health problems. This study explores how to evaluate the mental health of college students who have failed in entrepreneurship through EEG signals and the use of generative adversarial networks (GANs) to improve classification accuracy. Methods: We first collected EEG data from college students who have failed in entrepreneurship. These data were used to extract multiple features, including time domain features, frequency domain features, and nonlinear features. To improve the accuracy and robustness of the classification model, we introduced GAN. GAN enhances the diversity of the training data set by generating more samples, thereby improving the performance of the EEG signal classification model. This process includes training the generator to generate samples similar to real data and training the discriminator to distinguish between real and generated data. Through this adversarial process, the classification model is continuously optimized. Experimental results show that the EEG signal analysis method based on GAN can effectively classify the long-term emotional state of college students who have failed in entrepreneurship and accurately identify negative emotions such as depression and anxiety. The additional samples generated by GAN significantly improve the classification model’s ability to identify emotional states, making the assessment of mental health status more accurate. Conclusion: This study successfully improved the accuracy of EEG signal classification by introducing GAN technology, thus providing an effective tool for evaluating the mental health status of college students who failed in entrepreneurship. This method not only enhances the ability to detect negative emotions of college students but also provides a novel and practical solution for mental health assessment and intervention in colleges and universities, which helps to detect and intervene in potential mental health problems of college students who failed in entrepreneurship as early as possible.