DAIC-WOZ: On the Validity of Using the Therapist's prompts in Automatic Depression Detection from Clinical Interviews

arXiv:2404.14463v1 Announce Type: cross
Abstract: Automatic depression detection from conversational data has gained significant interest in recent years. The DAIC-WOZ dataset, interviews conducted by a human-controlled virtual agent, has been widely used for this task. Recent studies have reported enhanced performance when incorporating interviewer's prompts into the model. In this work, we hypothesize that this improvement might be mainly due to a bias present in these prompts, rather than the proposed architectures and methods. Through ablation experiments and qualitative analysis, …

abstract agent arxiv clinical conversational cs.ai cs.cl cs.cy cs.lg data dataset depression detection human interviews performance prompts studies type virtual virtual agent