Quality and Provenance Ontology Clause Samples

Quality and Provenance Ontology. Figure 7 shows the quality ontology used for the Reliable Information Processing. It contains the parameters described in previous Section. The quality ontology uses the PROV-O ontology to describe the provenance and reputation of data sources. An alternative to the PROV-O could be the Open Provenance Model14 (OPM) ontology as it allows a similar description of provenance. Since the PROV-O as it is used in the SOA ontology (Section 4.2) to describe the relation between StreamData and StreamEvent, it is also reused for the provenance annotation. The quality ontology uses the same concept with StreamData as the Stream Annotation Ontology described in Section 4.2. To specify the provenance of the StreamData a property “hasProvenance” is defined. Destination of this relation is an “Agent” of the PROV-O ontology who could be an organization, software agent or a person. Every Agent has a Reputation via the “hasReputation” property to describe the trustworthiness. As a result it is possible to get a reputation for every single StreamData. Main aspect of the quality ontology is the description of data streams with quality attributes. This quality can be used in the further processing of the streams, for example to find a stream with a high precision that is enough for a specific use case. The Quality class has the five subclasses Cost, Security, Accuracy, Communication and Timeliness. The quality ontology is an important part of the Reliable Information Processing component. Via the ontology it is possible to annotate data with additional information regarding its quality. In addition data sources can be annotated with reputation to determine their trustworthiness. Both, the QoI and the reputation, ensure that applications for the CityPulse framework can operate accurately with the best available information. An example for handling information with QoI and reputation could be a simple bus planning application. The planning application needs a data stream with the times of bus departures for a bus stop. There are different data streams available. On the one hand an official data stream of the municipality with the departure times. As this is the official timetable it has a high reputation with a good QoI. On the other hand there is an aggregated data stream of two persons who are reporting that the bus is delayed. As the aggregated stream only contains two reporting persons, who are unknown to the CityPulse framework, this data stream has a lower reputation as the...