5.4 Discussion
6.1.3 Reasoning Service
The reasoning service, as shown in Figure 6.21, is a service that takes an RDF document as an input, applies any reasoning rules and generates a new RDF document as an output. The reasoning service can help to generate links between resources which are ‘logically’ related. Taking the example as shown in Figure
6.12, it assumes that a user would like to discover data about the county of Hampshire. Although Hampshire contains Winchester, Eastleigh and Fareham, due to the missing link, the user would not be able to access these data.
RDF Reasoning RDF
Service
input output
Data
Figure 6.11: The reasoning service
os:7000000000017765 The County of Hampshire
parliament:member/10395 parliament:cons-426 dc:coverage http://parliament.psi.enakting.org Winchester contained_in parliament:member/101 parliament:cons-203 Eastleigh dc:coverage . . parliament:member/11884 parliament:cons-228 Fareham dc:coverage
Figure 6.12: Resource irretrievable via geographical gap
between location-based data. The service is designed for querying the UK terri- tory structure. The Ordnance Survey (OS) provides an ontology1 and an RDF
dump about spatial relations between UK regions. The triples from OS have been parsed and only the relation of physical containments have been retained, nor- malised and completed with the inverse relations in a separate knowledge base. The service presented here manages the spatial representation and reasoning by using the region connection calculus (RCC) as it is shown in Figure 6.13. For the sake of simplicity and efficiency2, the service only manages the non-tangential
proper part (NTPP), and the relative inverse, NTPPi relations. The knowledge extracted from the OS data set has been then normalised in terms of an internal ontology that represent qualitative spatial relations.
1http://www.ordnancesurvey.co.uk/oswebsite/ontology/SpatialRelations/v0.2/
SpatialRelations.owl
2In this way the data to manage has been drastically reduced in order to provide a very
Figure 6.13: RCC Eight Jointly Exhaustive and Pariwise Disjoint Relations A simple example of how the normalised triples from the OS ontology are used in coupling with a co-reference service for bridging the navigational gap for different data sets is depicted in Figure 6.14. In the figure it is possible to see that a single statement from OS describing the fact that the County of Hampshire contains Fareham and Winchester1:
os:7...17765 os:contains os:7...25157. os:7...17765 os:contains os:7...25128.
This has been translated into an internal representation containing both relations: part, and part of, as shown below:
os:7...17765 geoservice:part os:7...25157. os:7...25157 geoservice:part_of os:7...17765 . os:7...17765 geoservice:part os:7...25128. os:7...25128 geoservice:part_of os:7...17765 .
The containment relations so normalised (see central part of Figure6.14) are then internally stored in the system and queried for serving users’ requests.
The normalised containment relations are integrated with the information pro- vided by the co-reference system that allows bridging different data sources, both in the input phase (i.e. where the input URI must be translated into its OS equivalent, see top part of Figure 6.14) and the output phase (i.e. when the re- sults must be translated into a target data set provided by the user, see bottom part of Figure 6.14). The co-reference service used is the http://sameas.org
service as mentioned previously. The relevant bundles have been retrieved from the service and cached for performance. It is important to note that, in order to choose the required quality of service, one could opt for using one co-reference service instead of another. The functionality provided is transparent from the provenance of the co-reference bundles.
Exploiting co-reference services and OS ontology, it is therefore possible to infer containment relations between resources from different data sets. For exam- ple:
dbpedia:Hampshire owl:sameAs os:7...17765 AND
os:7...17765 geoservice:part os:7...25128 AND
os:7...25128 owl:sameAs dbpedia:Winchester =⇒
os:7000000000017765 os:7000000000025157 Hampshire county Fareham os:7000000000025128 Winchester ... dbpedia:Fareham (UK Parliament constituency)
dbpedia:Winchester (UK Parliament constituency)
... dbpedia:Hampshire http://dbpedia.org http://dbpedia.org http://data.ordnancesurvey.co.uk crime:Hampshire http://crime.psi.enakting.org parliament:cons-637 http://parliament.psi.enakting.org parliament:cons-228 ...
owl:sameAs part part_of
Figure 6.14: The architecture of reasoning service and interaction with co- reference system
For a location URI, the services will provide a list of geographical regions ei- ther containing the location or contained by the location. For example, when querying a URI about the city Winchester, the service will return a list of lo- cation URIs such as its county (Hampshire), or a list of constituent locations contained in Winchester. This service will enable the user to group or slice location-based data at different ways, thus enabling navigation at different gran- ularities. The service can be accessed via http://geoservice.psi.enAKTing.
org/{command}/{dictionary}/{format}/{URI}. The command option is either
‘contains’ or ‘container’ of the location, where the dictionary option is essentially a filter that enables the user to select only results from certain datasets. The
service supports reasoning from DBpedia, Ordnance Survey, UK National Statis- tics, Geonames, PSI, OpenCYC, and the Openly Local. More information about the reasoning service can be found in Appendix 7.2.
The reasoning service can be implemented on precomputed data, or it can be processed on the fly. In order to ensure its extensibility, the reasoning component can be implemented by using the Semantic Web Rule Languages, so that the reasoning components can be reused and enhanced easily.