Researchers are exploring the potential of Natural Language Processing (NLP) and Machine Learning (ML) algorithms to make Cognitive Behavioral Therapy (CBT) more accessible and affordable to those dealing with depression and anxiety during the COVID-19 pandemic. CBT aims to recognize and restructure negative thinking patterns known as cognitive distortions, which arise due to errors in reasoning.
The project uses linguistic features and different classification algorithms to detect cognitive distortions using NLP. The researchers found that pre-trained Sentence-BERT embeddings to train an SVM classifier yielded the best results with an F1-score of 0.79. This work provides insights into the types of linguistic features inherent in cognitive distortions and how these techniques can be applied on a small dataset.
Read the research: Detecting Cognitive Distortions from Patient-Therapist Interactions
