In the future these and other services will play a more ­pronounced role for the assurance as well as for the generation of new ­revenue sources.  According to predictions¹ a market ­volume of US$18 billion for location-specific advertising in 2015 can be expected. Moreover, in comparison with “conventional” Internet ­advertising, the personalized network access in combination with site information significantly increases the success ratio of location based advertising. In particular mobile network operators are well positioned for this market because of their detailed knowledge about the customer service usage and the most efficient sales channel to address the customers. In addition, these operators have dense networks and consequently exact information about the location of their customers at any time. The reach and success ratio of location-specific ­advertising can be increased substantially, if information about the ­customer behavior for a targeted placement of advertisements instead of simple billing-related content is used.

Spatial data mining reveals your everyday actions

Spatial data mining allows a systematic analysis of all of the ­above-described data sources with respect to spatial and ­temporal correlations.

Data mining is the semi-automated search for hidden contexts, typically in large data pools. The procedure derives from ­machine learning, statistics, and database theory. Spatial data mining adds a spatial analysis component so that correlation ­analyses of disjunct data sources, ranging from static and ­dynamic data to geo-referenced data, can be conducted with the aid of ­geo-information systems.² The advantages of such procedures are simple automation and good scalability of the analyses.

These procedures have been applied successfully in areas such as public security, transport, epidemiology, or climatology by institutions such as NASA and the United States Department of Transportation (USDOT), but are now being used more and more frequently in the field of geo-marketing. Typical ­examples are the optimization of sales regions, the ideal placement of branch stores of a retail chain or the determination of areas with low credit ratings so that certain payment methods can be excluded for mail-order business. As source serve easily accessible data such as population figures, population density, demographics, and purchasing power, i.e. data with high spatial resolution, provided either by communities and governments or purchased from vendors.