Consistently handling geographical user data: Merging of coreferent POIs

Abstract

In the context of geographic information systems (GIS), points of interest (POIs) are descriptions that denote geographical locations which might be of interest for some user purposes. Examples are nice views, historical buildings, good restaurants, recreation areas, etc. Because information gathering with respect to POIs is usually a resource consuming task, the user community is often involved in this task. In general, POI data that originates from different sources (or users) is vulnerable to imperfections. Different POIs referring to, or describing the same physical geographical location might exist. Such POIs are said to be coreferent POIs. Coreferent POIs must be avoided as they could introduce uncertainty in the data and blemish the database. In this paper, a novel soft computing technique for the (semi-)automatic detection and merging of coreferent POIs is presented. Hereby the focus is on the aspects of the merging technique. Fuzzy set and possibility theory are used to cope with the uncertainties in the data. An illustrative example is provided.