Identity attributes quantitative analysis and the development of a metrics model using text mining techniques and information theory

Term weighting has been applied to quantify and rank text data in information retrieval. Shannon´s information theory called entropy is another area that is used to quantify information. In this paper, term weighting and entropy are used to compose an identity attribute metric model. A set of application forms are used to form a sample space of identity attributes and three corpora are used to generate the required statistics used to compose an identity attribute metric model. The composed metric model has application in point based authentication systems, such as banking, immigration and implementing intelligent authentication systems.