Gauging the internet doctor

Abstract

As more and more content is published and consumed online, it is imperative to know if an information nugget found on the Web is trustworthy or not. This is especially important for online medical information as it affects the most vulnerable group of users looking for medical help online. In this paper, we study the feasibility of automatically assessing the trustworthiness of a medical claim based on community knowledge, and propose techniques to assign a reliability score for an information nugget based on support over a community-generated collection. Specifically, we model the trustworthiness of a medical claim based on experiences shared by users in health forums and mailing lists. The proposed claim scores can be used to rank related claims on their relative trustworthiness. We further extend the notion of trustworthiness to a site (or equivalently, a database of claims from the site) and propose a scheme to rank sites based on aggregating the trust scores of claims from the site. Our experiments show that community knowledge can be exploited to help users distinguish reliable medical claims from unreliable ones. The proposed techniques can be applied to other domains where similar corpora are available.

A Study on Online Health Community Users’ Information Demands Based on the BERT-LDA Model

“…Like these studies, our findings indicate that users prefer to receive supportive and uplifting emotions through online health communities. Moreover, the sentiment tendencies identified in this study are similar to those found by Vydiswaran et al [26]. This highlights the significance of understanding the connection between users' health demands and emotional demands in order to improve the efficiency of information retrieval and explain information behavior.…”

“…On the other hand, some researchers have utilized Facebook data to analyze user emotional expression characteristics, identifying types of users who require emotional support and discussing the positive impact of such support [24,25]. Sentiment tendencies within the text are analyzed to study the relationship between users' health demands and emotional demands, which is crucial for enhancing information retrieval efficiency and explaining information behavior [26].…”

Knowledge Discovery and Hypothesis Generation from Online Patient Forums: A Research Proposal

“…So far, the approaches to automatically assessing credibility of health-related information on social media has been limited to three studies (Viviani and Pasi, 2017a). Firstly, Vydiswaran et al (2011) used textual features to compute trustworthiness based on community support. They evaluated their approach using simulated data with varying amounts of invalid claims, defined as disapproved or non-specific treatments, e.g.…”