Citizen Empowered Mapping by Michael Leitner & Jamal Jokar Arsanjani

Author:Michael Leitner & Jamal Jokar Arsanjani
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

7.2.3 Geospatial Entity Resolution

In order to compare the routes computed based on crowdsourced POIs with the routes computed based on POIs obtained through other approaches, a matching method is needed. One such method is “Geospatial entity resolution” which determines a set of “true” locations from multiple data sources (Kang et al. 2007). More specifically, the geospatial entity resolution method determines whether two location references, in different datasets, represent the same real-world location (Sehgal et al. 2006). Several approaches have been applied in a real-world setting. The most common method relies solely on spatial coordinates and matches objects using a one-sided nearest join (Beeri et al. 2004). Other methods include string-matching algorithms and feature-based matching using location features (i.e., attributes) (Sehgal et al. 2006). A newer development is the use of interactive visualization tools to aid in the matching process (Kang et al. 2007).

There are several issues when dealing with entity resolution in the geospatial domain such as data collectors using different terms and language, and data collectors’ use of varying scales and geometries (points vs. areas) to represent the real-world entities (Sehgal et al. 2006). POI attributes such as type (e.g., school, hospital) can be used to increase the confidence of a match among entities (Sehgal et al. 2006). Geospatial databases generally include attributes alongside the POI geometry, thus, the method employed in Sehgal et al. (2006) is used in this work.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.