A methodological toolbox for exploring collections of textually annotated georeferenced photographs

Abstract

This chapter provides a brief overview of some methodologies used to extract meaning from the analysis of geotagged images. Broadly they draw from research in natural language processing and statistical and exploratory techniques. The confidence we attach to outputs from such analysis depends upon the questions we ask, our ability to take account of both the behaviour and motivation of the users contributing to user generated content, and the close relationship between how the data are spatially aggregated and the meanings associated with descriptions of images.