We approach Data Quality from a Services perspective. That is, our main aims are to address 1) upholding properties of data representation according to some agreed set of service rules or business objectives and 2) the correctness of service transaction data for a particular service operation. We adopt the the principles of the Semiotic Information Quality Framework [1] which specifies three important aspects of data quality for the application of semiotics to information systems. Firstly on a syntactic level, does the data conform to integrity rules (e.g. a booking number is never 0)? Secondly, on a semantic level, does the data represent the actual real-world phenomena (e.g. ship booking actually represents the real-world ship booking entity)? Thirdly, at the pragmatic level, how useful is the data for a particular task? The table below lists each of these levels along with description, examples and validation that can be performed to analyse these levels.
| Level | Property | Example | Validation |
|---|---|---|---|
| Syntactic | Integrity Rules | ship.bookings < 50 | Through Meta-data Integrity Rules |
| Semantic | Ontological Semantics | Prior to the ship departing, cabins are available for booking. | Data is complete and meaningful in a particular context or scenario |
| Pragmatic | Usefulness for Purpose | Customer details are provided as a part of a booking transaction. | Accessible and secure data regulations |
These Semiotic Information Quality assets are integrated in to an overall approach known as SEMDR.
Semantic Meta-Data Repositories (SEMDR)
For further details of SEMDR, please see the Verification and Validation overview.
[1] Price, R. and Shanks, G., A semiotic information quality framework: development and comparative analysis. Journal of Information Technology 20(2), June 2005, 88{102).