8265

Preference Based Filtering and Recommendations for Running Routes

Hassan Issa, Amir Guirguis, Shary Beshara, Stefan Agne, Andreas Dengel

Proceedings of the 12th International Conference on Web Information Systems and Technologies International Conference on Web Information Systems and Technologies (WEBIST-2016), 12th International Conference on Web Information Systems and Technologies, April 23-25, Rome, Italy , Pages: 139-146 , SCITEPRESS, Rome , 2016
With the current trend of fitness and health tracking and quantified self, hundreds of relevant apps and devices are being released to the consumer market. Remarkably, some platforms were created to collect running-route data from these different sources in order to provide a better value for users. Such data could be employed in finding running routes based on the user’s preferences rather than being limited to the proximity to the user's location. In this work, a classification system for running routes is introduced considering performance factors, visual factors and the nature of route. A running-route content-based recommender system is built on top of this classification enabling learning user preferences from their performance history. The system was evaluated using data from active runners and attained a promising recommendation accuracy averaging 84% among all subject users.

Show BibTex:

@inproceedings {
       abstract = {With the current trend of fitness and health tracking and quantified self, hundreds of relevant apps and devices
are being released to the consumer market. Remarkably, some platforms were created to collect running-route
data from these different sources in order to provide a better value for users. Such data could be employed
in finding running routes based on the user’s preferences rather than being limited to the proximity to the
user's location. In this work, a classification system for running routes is introduced considering performance
factors, visual factors and the nature of route. A running-route content-based recommender system is built on
top of this classification enabling learning user preferences from their performance history. The system was
evaluated using data from active runners and attained a promising recommendation accuracy averaging 84%
among all subject users.},
       number = {}, 
       month = {4}, 
       year = {2016}, 
       title = {Preference Based Filtering and Recommendations for Running Routes}, 
       journal = {}, 
       volume = {}, 
       pages = {139-146}, 
       publisher = {SCITEPRESS, Rome}, 
       author = {Hassan Issa, Amir Guirguis, Shary Beshara, Stefan Agne, Andreas Dengel}, 
       keywords = {},
       url = {}
}