Question-Level Feature Extraction on DAIC-WOZ Dataset
The DAIC-WOZ dataset contains clinical interviews designed to support the diagnosis of psychological distress conditions such as anxiety, depression, and post-traumatic stress disorder. This repository provides code for extracting question-level features from the DAIC-WOZ dataset, which can be used for multimodal analysis of depression levels.
Detaljeret Introduktion
The DAIC-WOZ dataset is a valuable resource for researchers working on psychological distress conditions. This repository provides tools and code for extracting features at the question level, which can help in understanding the role of multimodal features in diagnosing depression. The related paper provides further insights into the methodology and results.
MereDatasæt
HappyDB: a happiness database of 100,000 happy moments
HappyDB is a crowd-sourced collection of 100,000 happy moments designed to advance the understanding of happiness through text analysis. The database is publicly available and aims to support research in natural language processing (NLP) and positive psychology. It provides insights into the causes of happiness and suggests sustainable actions for improving well-being.
tartuNLP/Reddit Anhedonia Dataset - hf-mirror
tartuNLP/reddit-anhedonia by huggingface-mirror (hf-mirror)
SoulChat2.0: Psychological Counselor's Digital Twin Framework
SoulChat2.0 is a framework for constructing the digital twin of psychological counselors, designed to support the development of AI applications in mental health. It includes a data generation module and a modeling module, enabling the creation of personalized counseling models based on limited real-world counseling cases.