Skip to main content

Data quality 'differs for addresses and product information'

Rachel Wheeler Archive
A number of challenges exist when extending name and address data quality procedures to product information, an expert asserts.

Chief executive of The Information Difference Andy Hayler tells that the latter issue is one many companies face, although a number of algorithms enable address standardization.

These are designed to fix data records which contain partly misspelt or incomplete addresses but unstructured information is harder to deal with, he states.

Ensuring data quality among product information is more difficult due to the range of attributes and the need to draw on a number of systems, all of which can contain overlapping information, he tells the news provider.

Mr Hayler says the quality of product data is assured by building up semantic rules that allow the software to recognise the elements of product information even from a partially constructed record.

William Pulleyblank, chief of IBM's center for business optimization, recently told that data quality is a crucial part of his firm's attempt to automate business decision making.