Geography Reference
In-Depth Information
Table 1
Methods for semantic enrichment
Method
Achieves
Spatial and spatio-temporal
clusters
Grouping of the photos into clusters using distance metrics and
timestamps
Text analysis
Grouping of the photos into contexts using textual information
(title, tags)
Content-based analysis
Grouping of the photos into contexts using visual similarity
Analysis of events (time
series)
Temporal characteristics of the cluster. Frequency of events
POI database and entity
relations
Retrieving photo topics (nature, landmarks) and hierarchies of
concepts
User profiling
Profiling of users in a cluster using photo semantics
4.2.2 Points of Interest
The Wikipedia database (
http://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_
Georeferenzierung/Wikipedia-World/en#Static_layer
)
can be used as a source of
POI data. This database is an on-going community project aimed at applying
geographic annotation to articles describing interesting sites around the world. The
database that is currently available contains 815,085 entries of various sites such as
cities, landmarks, monuments, buildings, towers annotated with coordinates and
titles.
4.2.3 Geographical Features
GeoNames (
http://www.geonames.org
)
, a geographical database contains over
eight million geographical names and consists of 7 million unique features,
2.6 million populated places and 2.8 million alternate names. All features are
categorized into one out of nine feature classes, and are further subcategorized into
one out of 645 feature codes (mountains, lakes, monuments). The elements of the
dataset are organized into isA (conceptual inheritance) or partOf (spatial inclusion)
relations. This dataset is freely available for download or accessible through web
services.
4.3 Methods for Semantic Enrichment
In this section we describe six main methods for semantic enrichment. Table
1
briefly summarizes the proposed methods.
