Article Preview
TopIntroduction
In the last decade, the World Wide Web became a global access point for information providing not only linked documents, but also an extensive pool of mashup services that require raw and structured data. The Linked Data (Heath & Bizer, 2011) paradigm has evolved over the time from the vision of the Semantic Web (Berners-Lee, Hendler, & Lassila, 2001) and took the driving force position in pushing the Web of Data. Taking the Web of Documents and its key standards as a basis, Linked Data describes an approach for publishing, sharing and linking structured data on the Web. Data resources are identified via dereferenceable HTTP Unified Resource Identifiers (URIs) and are related to each other in the form of subject-predicate-object triples using the Resource Description Framework (RDF) (W3C, 2004a). This is done within a dataset as well as with resources from other datasets.
Over the years, many researchers, institutes, libraries, etc. published their data according to those principles covering various content domains, such as Media, Life Sciences, Geographic Data or Government, all directly linked to each other building a huge graph of interlinked structured data, which is also known as the Linking Open Data (LOD) Cloud (see Figure 1). The number of datasets in this cloud has been increasing rapidly and has reached a total of 338 by April 2013. These datasets provide a huge pool of contextual information that can be exploited for new and innovative context-aware services.
Figure 1. Linking open data cloud, by Richard Cyganiak and Anja Jentzsch, http://lod-cloud.net/, © 2011, CC-BY-SA license
In parallel to the developments in the Web, the emergence of smartphones and the roll-out of different wireless network technologies including GSM, UMTS, LTE and WiFi, revolutionized the mobile market and enabled the realization of mobile services that have a major impact on our daily life. In particular, location-based services (being a sub-group of context-aware services) have become very popular. According to Shen et al. (2011), the number of location-based service users was supposed to reach 1.4 billion by 2012.
A location-based service uses the location of a user (through his mobile device) in order to perform any kind of related action, such as recommending restaurants or points of interest in the vicinity. The position of the user can most accurately be determined by utilizing the Global Positioning System (GPS). Cell-ID and WiFi Positioning, on the other hand, are less accurate, but fast and energy-efficient alternatives. The basic working principle of these positioning methods is that the device sends a query to a cellular location provider (e.g., Google) including information about the radio cells and/or WiFi access points nearby. This information usually consists of a unique identifier for each cell, the received signal strength, and additional data. The cellular location provider uses a database containing network topology data to identify the position of the device and returns the location information.
In order to establish such a cellular database, companies like Google or Apple exploit the capabilities of smartphones via a crowdsourcing approach measuring cell data with which the network topology can be approximately derived. Since this data is quite valuable for them due to many reasons (Bareth, 2012), they do not provide them in an open-source environment. In addition, their APIs are quite restrictive and only provide WGS84 coordinates for a position request. In contrast to the aforementioned commercial cellular location providers, open databases, such as OpenCellID (8Motions, 2013) and OpenBMap (OpenBMap, 2013), allow access to the entire available dataset collected by their own, but much smaller crowdsourcing communities. These projects utilize various position estimation methods for deriving the position and coverage area of a base station out of a collection of single measurements.