Health Misinformation Detection in the Social Web: An Overview and a Data Science Approach

Area of Science:

• Digital Health
• Information Science
• Machine Learning

Background:

• The internet's vast content creation capabilities have democratized information but also enabled the spread of health misinformation.
• Low health literacy among laypersons and information overload for experts complicate discerning credible health information online.

Purpose of the Study:

• To investigate features and machine learning techniques for assessing the genuineness of online health information.
• To develop automated solutions to aid both expert and non-expert users in identifying health misinformation.

Main Methods:

• Analysis of distinct feature groups relevant to online content.
• Application and evaluation of various machine learning techniques.
• Utilizing recently generated, publicly available datasets for health misinformation assessment.

Main Results:

• Identification of effective features and machine learning models for discerning genuine from non-genuine health information.
• Demonstrated potential for automated systems to combat health misinformation across web pages and social media.

Conclusions:

• Machine learning offers a promising avenue for developing automated tools to combat the growing problem of online health misinformation.
• Effective feature engineering and model selection are crucial for accurately assessing the credibility of digital health content.