How to sort trustworthy health online information? Improvements of the automated detection of HONcode criteria

Abstract

When searching on the Web, laypersons and professionals have difficulty in determining the quality or trustworthiness of health websites. No simple approach can differentiate among trustworthy and unreliable health websites in the results provided by major search engines such as Google, Yahoo or Bing. The European project Kconnect proposes to classify the reliability and readability levels of health-related Web sites and pages, tools according to the detection of HONcode criteria, the Health On the Net Foundation (HON) Code of Conduct using automated tools that examine how technical the health information contained in each document is. This article focuses on enhancement of automated detection of HONcode criteria in real-time settings. Applications of the approach include integration into the HONcode certification process, and embedding it as generic filtering tool into user-centered health domain search engines as well as into major general search engines such as a Web browser extension.