Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Engage Grants Program

Université Laval

$24,926.00

Personalized Recommendations for a Social Network of Photographers

FY 2016-17

Summary
500px is a major photography social network and marketplace aimed at aspiring and professionalphotographers. While the 500px platform provides photo recommendations to its users, these recommendationsare only lightly personalized through the links between the users on the social network and rely mostly on thenovelty and popularity of photos. Our team at Université Laval will take advantage of its expertise in machinelearning and stochastic optimization in order to design and prototype novel and effective recommendationalgorithms adapted to 500px data. Specifically, we will develop highly personalized recommendations byexploiting user characteristics (i.e., profile, social links, activities), photo information (i.e., a photo's metadata,neural network features, tags, subject categories), and activities on the photos (i.e., views, likes, comments).Two approaches for personalized recommendations will be investigated, one based on matrix factorization andthe other based on a contextual bandits technique. Analysis over the 500px data will also be conducted, toensure that the proposed algorithms can properly handle the sparsity of data and that the representation used isappropriate for the task at hand. Scalability of the algorithms over very large datasets is also an importantelement that will be considered. Achieving better personalized recommendations of photos contributes directlyto the business activities of 500px and will help it consolidate its position as a leading social network forphotographers.
Program
Engage Grants Program
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
492395-2015
Name nom
Gagné, Christian
Department département
Génie électrique et génie informatique
Organization id
51
Institution établissement
Université Laval
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2016
Competition year année de concours
2015
Award amount
24926
Program id
EGP
Program name en
Engage Grants Program
Program name fr
Programme de subventions d'engagement partenariat
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1552
Committee name en
Quebec Internal Decision Committee
Committee name fr
Comité de décision interne pour le Québec
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2800
Research subject group en
Artificial intelligence
Research subject group fr
Intelligence artificielle
Research subject en
Artificial intelligence (computer vision, use2603)
Research subject fr
Intelligence artificielle (vision artificielle, utiliser 2603)
Partie
1
Nb partie
1
Application title
Personalized Recommendations for a Social Network of Photographers
Keyword
NSERC
Application summary
500px is a major photography social network and marketplace aimed at aspiring and professionalphotographers. While the 500px platform provides photo recommendations to its users, these recommendationsare only lightly personalized through the links between the users on the social network and rely mostly on thenovelty and popularity of photos. Our team at Université Laval will take advantage of its expertise in machinelearning and stochastic optimization in order to design and prototype novel and effective recommendationalgorithms adapted to 500px data. Specifically, we will develop highly personalized recommendations byexploiting user characteristics (i.e., profile, social links, activities), photo information (i.e., a photo's metadata,neural network features, tags, subject categories), and activities on the photos (i.e., views, likes, comments).Two approaches for personalized recommendations will be investigated, one based on matrix factorization andthe other based on a contextual bandits technique. Analysis over the 500px data will also be conducted, toensure that the proposed algorithms can properly handle the sparsity of data and that the representation used isappropriate for the task at hand. Scalability of the algorithms over very large datasets is also an importantelement that will be considered. Achieving better personalized recommendations of photos contributes directlyto the business activities of 500px and will help it consolidate its position as a leading social network forphotographers.