FairGRecs: Fair Group Recommendations by Exploiting Personal Health Information

FairGRecs aims to offer valuable information to users, in the form of suggestions, via their caregivers, and improve as such the opportunities that users have to inform themselves online about health problems and possible treatments. Specifically, FairGRecs introduces a model for group recommendations, incorporating the notion of fairness. For computing similarities between users, we define a novel measure that is based on the semantic distance between users’ health problems. Our special focus is on providing valuable suggestions to a caregiver who is responsible for a group of users. We interpret valuable suggestions as ones that are both highly related and fair to the users of the group.

[1]  Peter Emerson,et al.  The original Borda count and partial voting , 2013, Soc. Choice Welf..

[2]  Mark Mosley,et al.  Evaluating the quality of online information about concussions , 2014, JAAPA : official journal of the American Academy of Physician Assistants.

[3]  Manolis Tsiknakis,et al.  Digital patient: Personalized and translational data management through the MyHealthAvatar EU project , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[4]  Hans-Peter Kriegel,et al.  Fast Group Recommendations by Applying User Clustering , 2012, ER.

[5]  Uri Kartoun,et al.  A Methodology to Generate Virtual Patient Repositories , 2016, ArXiv.

[6]  Hans-Peter Kriegel,et al.  "Strength Lies in Differences": Diversifying Friends for Recommendations through Subspace Clustering , 2014, CIKM.

[7]  Kostas Stefanidis,et al.  Fairness in Group Recommendations in the Health Domain , 2017, 2017 IEEE 33rd International Conference on Data Engineering (ICDE).

[8]  Martin Wiesner,et al.  Adapting recommender systems to the requirements of personal health record systems , 2010, IHI.

[9]  Nikos Mamoulis,et al.  Recommending Packages to Groups , 2016, 2016 IEEE 16th International Conference on Data Mining (ICDM).