Luis Fernandez-Luque*, NORUT (Northern Research Institute), Tromsø, Norway Carlos-Luis Sanchez-Bocanegra, University of Seville, Málaga, Spain Jose Luis Sevillano Ramos, University of Seville, Seville, Spain Stathis Konstantinidis*, NORUT (Northern Research Institute), Tromsø, Norway Randi Karlsen, University of Tromsø, Tromsø, Norway Track: Research Presentation Topic: Health information on the web: Supply and Demand Presentation Type: Rapid-Fire Presentation Submission Type: Single Presentation Building: Sol Principe Room: C - Almudaina Date: 2014-10-09 02:00 PM – 02:45 PM Last modified: 2014-09-03

If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL).

Background: Social media have large volume of accurate and trustworthy health content available over Internet, giving precise health information can be difficult. Flooding health information is mixing with misleading recommendations. We believe that community wisdom obtain accurate and precise health information.
Objectives: To retrieve precise recommendation in correspondance with trusted video contents.
Methods: We designed a method that estimates the precise of recommended links with trustworthiness health videos. In two different experiments (Diabetes and Blood Pressure), four clinicians evaluated recommender links from 23 health videos (6 of Diabetes recollected from diavideos portal (http://ehealth.norut.no/diavideos/) with more than 3 recommended results and 17 of Blood Pressure from HT most popular videos with more than 3 recommended results), they provided 114 ratings.
Results: Our method may be used for an precise recommender links in relation with video content, Overall, inter-rater reliability gave a substantial agreement (0.65 Kappa Cohen) with Blood Pressure precision@3 = 0.91 and (0.69 Kappa Cohen) with Diabetes precision@3 = 0.67. However, the method need to increase the number of recommender links in some videos (only 1 o 2 recommender links).
Conclusions: The method showed promising results (specially in Blood Pressure).The main difference between HT and DM is the number of recommended videos from the HealthTrust algorithm. Our research indicates that the use of semantic web for recomender links may be precise with trusted video contents.