A method for location recommendation via skyline query tolerant to noisy geo-referenced data

This work presents a method to perform a location recommendation based on multiple criteria even when there is noise in the coordinates. More specifically, the skyline query is adapted to handle this noise by modeling the errors of geo-referenced points with an appropriate probability distribution and modifying the traditional dominance criterion used by that technique. The method is applied to a scenario in which the coordinates are set by a geocoding process in a sample of schools in a Brazilian city. It enables one to choose the level of confidence in which a point is removed from the skyline solution, i.e., the location recommendation.