Inducing intentional errors in concatenated databases
School of Computer and Information Science, Edith Cowan University
Place of Publication
Perth, Western Australia
School of Computer and Information Science
Raw information from surveys is processed to produce useable information for decision support, characterise population segments and construct databases. Concatenation of databases reuses data from multiple sources and requires that duplicate entries are detected, classffied and eliminated. Errors that pass through the 'data noise' removal process occur and may be deliberate on the part of the respondent. Deception on an individual basis is unlikely to be effective though a concerted, decentralised campaign may have substantial effect. Deceptive information provided to a number of surveys may not be detected until concatenation and widespread deception may attack the validity o,f the survey and the reputation of the surveyor or data user.