Mapping Space-Related Emotions out of User-Generated Photo Metadata Considering Grammatical Issues

Current location based services mainly provide objective information and collections of facts. Subjective components such as emotions and opinions can provide additional alternative information useful in decision making, e.g. in tourism, business, entertainment and the like. Therefore research on effect analysis was carried out by capturing and analyzing georeferenced emotions from user-generated content. An approach was developed for extracting location-based emotions from the written language in the metadata of georeferenced Flickr and Panoramio photos, i.e. from their titles, descriptions and tags. Within this extraction approach various grammatical issues were considered, like negations of words or amplifications. Procedures were developed for modifying the affected emotions, for example for inverting or intensifying them. The approach was applied to the study area of Dresden, Germany. The obtained emotions were documented in emotional maps of geospace as well as in valence-arousal-space originating from psychology.