From ESIWiki

Name

Information Quality of Linked Sensor Data

Owner

Chris Baillie (c.baillie@abdn.ac.uk)

Linked Open Data Source

dot.rural Sensor Data repository http://dtp-126.sncs.abdn.ac.uk:8081/

Background

The Web has evolved from a collection of hyperlinked documents to a vast ecosystem containing data describing documents, services, devices and even people. However, the inherent open nature of the Web allows any ‘thing’ to publish this data and, as such, the quality of information published on the Web is a major issue.

As part of my PhD, I am exploring how semantic web technologies can be used to facilitate Information Quality (IQ) assessment in the Web of Linked Sensor Data. Sensor observations can be annotated with metadata describing the characteristics of the sensor and we can also capture observation provenance: when was it created it, what it was created by, who owns the sensor and even why it was created.

IQ is generally seen as a multi-dimensional construct, composed of certain quality indicators such as accuracy, timeliness, consistency, etc… Sensor observation metadata and provenance can be used to quantify these quality indicators in order to drive reasoning about quality using appropriate metrics.

Use case Scenario

We have developed a sensor network test-bed that publishes sensor observations as Linked Data. However, the quality of the sensor’s observations is currently unknown and therefore we require some means to evaluate this. Examining the provenance records of the observations and the sensor may provide some insight into how the sensor generally performs but, critically, we aim to append the outcomes of quality assessment to the provenance graph and investigate how this may facilitate future quality assessment.

Problems and Limitations

There is no standardised way to utilise provenance information in quality assessment; no fixed way to map provenance elements to quality indicators.
Incorrect provenance information can lead to inaccurate quality assessment.

Requirements for Provenance

Describing how observations have been created – created by what?
Detailing the owners of the sensor.
Detailing who has previously used the sensor data and why.
Capturing the outcomes of quality assessment.

Related Work

W3C SSN Ontology http://www.w3.org/2005/Incubator/ssn/

Assessing Quality in the Web of Linked Sensor Data http://www.csd.abdn.ac.uk/~cbaillie/papers/WebIQ.pdf

Provenance in the Web of Linked Sensor Data https://www.horizon.ac.uk/images/stories/p39-Baillie.pdf