کنفرانس مهندسی برق ایران

صفحه اصلی / سی و یکمین کنفرانس بین المللی مهندسی برق

A Novel Model for Student's Mental Health Monitoring Based on Hard and Soft Data Fusion

نویسندگان :

Mohammad Fatahi¹ Masoud Alizadeh² Behzad Moshiri³

1- دانشگاه تهران 2- دانشگاه تهران 3- دانشگاه تهران

کلمات کلیدی :

Speech-Emotion Recognition،Hard and Soft Data Fusion،CNN-LSTM،Dempster-Shafer Theory،Mental Health Monitoring

چکیده :

The importance of monitoring mental health has expanded in recent years. To promote the well-being of society, it is necessary to comprehend human behavior and monitor mental health. Even though existing studies provide a method for detecting mental illness, they are difficult to use for early diagnosis. In addition, the growing number of mental health problems among students has resulted in a range of societal challenges. Therefore, the objective of this study is to present a model for recognizing mental health illnesses in students by combining both soft and hard data sources. In previous studies, both types of data have been treated as separate entities in monitoring health status. In this study, we utilized a CNN-LSTM network for speech emotion recognition in the hard data flow to determine the probability of depression among students. In addition, the Dempster-Shafer theory was employed in the soft data stream to model psychologists' opinions. Finally, the results from both hard and soft data were combined to reach a final diagnosis.