A METHOD FOR LOCATION RECOMMENDATION VIA SKYLINE QUERY TOLERANT TO NOISED GEOREFERENCED DATA

Abstract

This paper presents a method to perform a location recommendation based on multiple criteria allowing noised coordinates. More speciï¬ cally, the skyline query is adapted to handle those noises by modeling the errors of georeferenced 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 speciï¬ c Brazilian city. It enables one to choose the level of conï¬ dence in which a point is removed from the skyline solution (the location recommendation).