Unifying Perspectives: CNN-LSTM Integration for Anxiety and Depression Prediction Through Textual Analysis

Depression is characterized by persistent negative emotions and can significantly disrupt individuals’ daily functioning and well-being. The global prevalence of long-term depressive symptoms continues to escalate, posing significant challenges to mental health professionals worldwide. Unfortunately, individuals dealing with depression may turn to self-harm or in extreme cases, contemplate suicide. Therefore early detection of depressive symptoms is important to enable timely intervention and improve treatment outcomes. This study proposes a novel methodology that integrates long short-term memory (LSTM) and convolutional neural network (CNN) for identifying and analyzing depression and anxiety symptoms in textual data from social media platforms. Through rigorous experimentation, promising results are achieved, highlighting the efficacy of the approach in accurately detecting and understanding mental health symptoms. This research contributes to advancing computational mental health, providing the potential of digital platforms for mental health disorder detection.