Vulcanus: A Recommender System for Accessibility based on Trails
Ismael Cardoso$^{1}$, Bruno Mota$^{1}$, Jorge Barbosa$^{1}$, Rodrigo Righi$^{1}$
$^{1}$University of Vale do Rio dos Sinos. São Leopoldo Brazil
email: ismaelcardoso@yahoo.com.br, brunomota55@gmail.com, jbarbosa@unisinos.br, rrrighi@unisinos.br

Schedule:Mon 19th@10:45, Room: C

The use of recommender systems is widespread in several everyday systems. People are now exposed to different offerings, based on their interests, in order to anticipate decisions. However, in a different way recommender systems for accessibility domain require further studies and researches. In order to provide resources for people with special needs, we developed the Vulcanus, a recommender system designed for people with disabilities, offering access to resources according user’s needs. The recommender system approaches concepts from ubiquitous computing, such as user profiles, context awareness, trails management, and similarity analysis. Vulcanus uses two different approaches: resources patterns and categories patterns; where different systems can use it, for example as a Web Service. Vulcanus was evaluated in different scenarios. These scenarios used more than 800 thousands of trails that simulate wheelchair users using resources along six months. The obtained results show that both approaches are able to provide relevant recommendations, providing resources that in fact can support users’ needs.

BibTex

@InProceedings{CLEI-2015:144469,
	author 		= {Ismael Cardoso and Bruno Mota and Jorge Barbosa and Rodrigo Righi},
	title 		= {Vulcanus: A Recommender System for Accessibility based on Trails},
	booktitle 	= {2015 XLI Latin American Computing Conference (CLEI)},
	pages 		= {385--396},
	year 		= {2015},
	editor 		= {Hector Cancela and Alex Cuadros-Vargas and Ernesto Cuadros-Vargas},
	address 	= {Arequipa-Peru},
	month 		= {October},
	organization 	= {CLEI},
	publisher 	= {CLEI},
	url 		= {http://clei.org/clei2015/144469},
	isbn 		= {978-1-4673-9143-6},
	}