Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.393205
Title: Mining credit card data
Author: Blunt, Gordon
ISNI:       0000 0001 3468 5296
Awarding Body: Open University
Current Institution: Open University
Date of Award: 2002
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Data mining is the process of finding interesting or valuable structures in large data sets. It is a modern discipline, and takes ideas and methods from statistics, machine learning, data management and other areas. In many ways, it is similar to exploratory data analysis, although the size of current data sets distinguishes between data mining and standard exploratory data analysis. Data mining can pose novel challenges because of the amount of data to be analysed. This thesis is concerned with modelling different aspects of credit card holders' behaviour and detecting patterns in the ways customers use their card accounts. A review of the literature on consumer purchasing behaviour and data mining is given, and for the latter the differing viewpoints of the statistical and the computer science communities are discussed. Two types of models are examined: descriptive models, which describe how customers have been observed to behave; and predictive models, which predict how customers are likely to behave in the future. Both types of model are applied to the main types of credit card use: repayment behaviour and transactional behaviour. Relationships between the two different sorts of behaviour are described, and models are examined which can link the two. Some valuable insights and discoveries of potential commercial value are described. Simple graphical tools are used to illustrate the unearthing of unexpected patterns and relationships, and it is shown how more sophisticated modelling can build on such discoveries.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.393205  DOI: Not available
Share: