Social security data mining : an Australian case study

Publication Type:
Thesis
Issue Date:
2013
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail01front.pdf270.06 kB
Adobe PDF
Thumbnail02whole.pdf3.83 MB
Adobe PDF
Data mining in business applications has become an increasingly recognized and accepted area of enterprise data mining in recent years. In general, while the general principle and methodologies of data mining and machine learning are applicable for any business applications, it is often essential to develop specific theories, tools and systems for mining data in a particular domain such as social security and social welfare business. This necessity has led to the concept of social security and social welfare data mining, the focus of this thesis work. Social security and social welfare business involves almost every citizen’s life at different life periods. It provides fundamental and crucial government services and support to varied populations of specific need. A typical scenario in Australia is that it not only connects one third of our populations, but also associates with many relevant stakeholders, including banking business, taxation and Medicare. Such business engages complicated infrastructure, networks, mechanisms, policies, activities, and transactions. Data mining of such business is a brand new application area in the data mining community. Mining such social welfare business and data is challenging. The challenges come from the unavailable benchmark and experience in the data mining for this particular domain, the complexities of social welfare business and data, the exploration of possible doable tasks, and the implementation of data mining techniques in relation to the business objectives. In this thesis, which adopts a practice-based innovative attitude and focusses on the marriage of social welfare business with data mining, we believe we have realised our objective of providing a systematic and comprehensive overview of the social security and social welfare data mining. The main contributions consist of the following aspects: • As the first work of its kind, to the best of our knowledge, we present an overall picture of social security and social welfare data mining, as a new domain driven data mining application. • We explore the business nature of social security and social welfare, and the characteristics of social security data. • We propose a concept map of social security data mining, catering for main complexities of social welfare business and data, as well as providing opportunities for exploring new research issues in the community. • Several case studies are discussed, which demonstrate the technical development of social security data mining, and the innovative applications of existing data mining techniques. The nature of social welfare is spreading widely across the world in both developed and developing countries. This thesis work therefore is timely and could be of important business and government value for better understanding our people, our policies, our objectives, and for better services of those people of genuine needs.
Please use this identifier to cite or link to this item: