All

What are you looking for?

All
Projects
Results
Organizations

Quick search

  • Projects supported by TA ČR
  • Excellent projects
  • Projects with the highest public support
  • Current projects

Smart search

  • That is how I find a specific +word
  • That is how I leave the -word out of the results
  • “That is how I can find the whole phrase”

AI-enhanced Mental Health Diagnosis: Leveraging Transformers for Early Detection of Depression Tendency in Textual Data

The result's identifiers

  • Result code in IS VaVaI

    <a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F23%3APU149784" target="_blank" >RIV/00216305:26220/23:PU149784 - isvavai.cz</a>

  • Result on the web

    <a href="https://ieeexplore.ieee.org/document/10333301" target="_blank" >https://ieeexplore.ieee.org/document/10333301</a>

  • DOI - Digital Object Identifier

    <a href="http://dx.doi.org/10.1109/ICUMT61075.2023.10333301" target="_blank" >10.1109/ICUMT61075.2023.10333301</a>

Alternative languages

  • Result language

    angličtina

  • Original language name

    AI-enhanced Mental Health Diagnosis: Leveraging Transformers for Early Detection of Depression Tendency in Textual Data

  • Original language description

    Early detection and treatment of depression depend critically on mental health assessment. Artificial intelligence-based methods have shown potential in assessing linguistic and cognitive patterns to spot those who are at risk of depression. This study tries to identify depression tendencies based on linguistic and cognitive characteristics by utilizing a transformer-based language model and self-attention. The goal is to assess how well the trained model performs at correctly identifying people who are predisposed to depression. A variant of the BERT (Bidirectional Encoder Representations from Transformers) model, trained on a larger corpus and for a longer duration, results in improved performance due to its strong capabilities in understanding natural language in the context of detecting depression using text data. The model can classify new text inputs and identify potential signs of depression by learning various linguistic cues and characteristics associated with depression such as sentiment patterns, language usage, emotional expressions, and topics discussed. The study also evaluates the efficacy of other machine learning classification models and long short-term memory networks in detecting depression tendencies. The proposed model achieves an astounding accuracy of 96.86% and is proven as the most effective model for detecting depressive tendencies. It underlines the usefulness of the proposed model in early detection and intervention programs for depression. This study advances mental health assessment and offers important insights into the recognition of mental health concerns by utilizing deep learning and natural language processing models.

  • Czech name

  • Czech description

Classification

  • Type

    D - Article in proceedings

  • CEP classification

  • OECD FORD branch

    20205 - Automation and control systems

Result continuities

  • Project

    <a href="/en/project/FW03010273" target="_blank" >FW03010273: Defectoscopy of painted parts using automatic adaptation of neural networks</a><br>

  • Continuities

    P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Others

  • Publication year

    2023

  • Confidentiality

    S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Data specific for result type

  • Article name in the collection

    2023 15th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)

  • ISBN

    979-8-3503-9328-6

  • ISSN

  • e-ISSN

  • Number of pages

    6

  • Pages from-to

    56-61

  • Publisher name

    IEEE Computer Society

  • Place of publication

    neuveden

  • Event location

    Gent, Belgium

  • Event date

    Oct 30, 2023

  • Type of event by nationality

    WRD - Celosvětová akce

  • UT code for WoS article