STAT: A Web-based Semantic Text Annotation Tool to Assist Building Mental Health Knowledge Base

Area of Science:

• Health Informatics
• Mental Health Research
• Human-Computer Interaction

Background:

• Mental health issues are prevalent in American adults, with many seeking information online.
• Existing online mental health resources often lack organization and quality.
• Manual curation of scientific literature for a mental health knowledge base (MHKB) is inefficient.

Purpose of the Study:

• To accelerate the curation of an evidence-based mental health knowledge base (MHKB).
• To integrate crowdsourcing for efficient data annotation.
• To design and prototype a web-based tool (STAT) for managing laypeople annotators.

Main Methods:

• Designed and prototyped STAT (Semantic Text Annotation Tool), a web-based annotation platform.
• Integrated real-time annotation recommendation features within STAT.
• Incorporated annotation quality analysis tools into STAT.
• Utilized crowdsourcing to recruit laypeople for annotation tasks.

Main Results:

• Developed a functional prototype of the STAT tool.
• Demonstrated the feasibility of using crowdsourcing for mental health information annotation.
• Enabled real-time recommendations and quality analysis to support annotators.

Conclusions:

• The STAT tool facilitates efficient management of crowdsourced annotators for building a high-quality mental health knowledge base (MHKB).
• Leveraging collective intelligence through crowdsourcing can significantly improve the curation process for online health information.
• This approach addresses the need for organized, evidence-based online mental health resources.