Investigating machine learning and natural language processing techniques applied for detecting eating disorders: a...

Area of Science:

• Computational linguistics
• Clinical psychology
• Artificial intelligence

Background:

• Natural Language Processing (NLP) and Machine Learning (ML) have advanced automatic text processing.
• Human language expression is crucial for mental health problem detection.
• Written language offers valuable insights for clinical professionals in diagnosing conditions like eating disorders.

Purpose of the Study:

• To systematically overview the latest research on using NLP and ML for eating disorder diagnostics.
• To analyze metadata, datasets, machine learning techniques, and model evaluations in this domain.

Main Methods:

• Systematic literature review.
• Analysis of metadata from published research papers.
• Examination of dataset characteristics (size, topics) and machine learning techniques applied.
• Evaluation of model performance, limitations, and associated risks.

Main Results:

• Identified key research areas including metadata analysis, dataset specifics, ML applications, and model evaluation.
• Assessed the current state of machine learning techniques for detecting eating disorders from text.
• Highlighted the performance, limitations, and potential risks of existing methodologies.

Conclusions:

• NLP and ML show promise in aiding the diagnosis of eating disorders through text analysis.
• Further research is needed to address model limitations and potential risks for ethical implementation.
• This systematic overview provides a foundation for future research and clinical application.