Abstract:
Massive spreading of medical misinformation on the Web has a significant impact on individuals and on society as a whole. The majority of existing tools and approaches for detection of false information rely on features describing content characteristics without verifying its truthfulness against knowledge bases. In addition, such approaches lack explanatory power and are prone to mistakes that result from domain shifts. We argue that involvement of human experts is necessary for successful misinformation debunking. To this end, we introduce an end-to-end system that uses a claim-based approach (claims being manually fact-checked by human experts), which utilizes information retrieval (IR) and machine learning (ML) techniques to detect medical misinformation. As a part of a web portal called FireAnt, the results are presented to users with easy to understand explanations, enhanced by an innovative use of chatbot interaction and involvement of experts in a feedback loop.
