Home // GEOProcessing 2012, The Fourth International Conference on Advanced Geographic Information Systems, Applications, and Services // View article

Automated Extraction and Geographical Structuring of Flickr Tags

Authors:
Omair Chaudhry
William Mackaness

Keywords: data mining; information retrieval; vernacular geography; granularity modelling

Abstract:
The volume and potential value of user generated content (UGC) is ever growing. One such source is geotagged images on Flickr. Typically, images on Flickr are tagged with location and attribute information variously describing location, events or objects in the image. Though inconsistent and ‘noisy’, the terms can reflect concepts at a range of geographic scales. From a spatial data integration perspective, the information relating to ‘place’ is of primary interest and the challenge is in selecting the most appropriate tag(s) that best describe the geography of the image. This paper presents a methodology for searching among the ‘tag noise’ in order to identify the most appropriate tags across a range of scales, by varying the size of the sampling area within which Flickr imagery falls. This is applied in the context of urban environments. Empirical analysis was then used to assess the correctness of the chosen tags (whether the tag correctly described the geographic region in which the image was taken). Logistic regression was then used to build a model that could assign a probability or confidence value to each selected tag as being a appropriate geographic tag. The high correlation values achieved bodes well for automated environments - environments in which this methodology could be used to automatically select meaningful tags and hierarchically structure UGC in order that it can be semantically integrated with other data sources.

Pages: 134 to 139

Copyright: Copyright (c) IARIA, 2012

Publication date: January 30, 2012

Published in: conference

ISSN: 2308-393X

ISBN: 978-1-61208-178-6

Location: Valencia, Spain

Dates: from January 30, 2012 to February 4, 2012