Evaluation of device-independent internet spatial location

Abstract

Device-independent Internet spatial location is needed for many purposes, such as data personalisation and social behaviour analysis. Internet spatial databases provide such locations based the IP address of a device. The free to use databases are natively included into many UNIX and Linux operating systems. These systems are predominantly used for e-shops, social networks, and cloud data storage. Using a constructed ground truth dataset, we comprehensively evaluate these databases for null responses, returned country/region/city, and distance error. The created ground truth dataset differs from others by covering cities with both low and high populations and maintaining only devices that follow the rule of one IP address per ISP (Internet Service Provider) and per city. We define two new performance metrics that show the effect of city population and trustworthiness of the results. We also evaluate the databases against an alternative measurement-based approach. We study the reasons behind the results. The data evaluated comes from Europe. The results may be of use for engineers, developers and researchers that use the knowledge of geographical location for related data processing and analysis, such as marketing.