Jrpit 39.1.qxp
نویسندگان
چکیده
Despite significant efforts by merchants, card issuers and law enforcement to curb fraud, online fraud continues to plague electronic commerce web sites. More advanced solutions are desired to protect merchants from the constantly evolving problem caused by fraud. The supervised machine learning technique for the most well known fraud detection algorithms makes them inadequate for an online system, which usually contains a mammoth size of non-stationery data. This paper describes a method to dynamically create user profile for the purpose of fraud detection. We use a data mining algorithm to adaptively profile legitimate customer behaviour in a form of association rule set from a transaction database. Then the incoming transactions are compared against the user profile to indicate the anomalies. A novel pattern match approach is proposed to evaluate how unusual the new transactions are. An empirical evaluation shows that we can accurately differentiate the anomaly behaviour from profiled user behaviour.
منابع مشابه
Jrpit 39.4.qxp
This paper examines factors which lead to the failure of electronic payment systems (EPS). In order to understand these factors a single case study of a failed Australian EPS was conducted. The findings confirm the influence of EPS adoption factors identified from the literature, which include cooperation with established entities, simplicity, trust, security and mutuality of stakeholder benefi...
متن کاملJrpit 39.3.qxp
With the development of e-commerce in China, some obstacles, such as poor Internet infrastructure and logistics problems have been gradually removed. Consumers’ distrust of online stores becomes the main problem impeding the acceptance of online shopping. Initial trust is the focus of this study because most Internet users in China have no experience of online shopping. Although the technology ...
متن کاملA Six Sigma-Based Process to Improve COTS Component Filtering
specification of a component C, a set of components K relevant to C and called candidate solution SO, and a set of pre-selected components from SO, called solution SN. Sc(i) represents the map associated to the input value i defined in the domain of Sc. This map should provide a valid value on the output domain of Sc, i.e. there is no empty maps or invalid associations. A similar assumption is ...
متن کاملEditorial: Organic carbon pools in permafrost regions on the Qinghai–Xizang (Tibetan) Plateau
The current Northern Circumpolar Soil Carbon Database did not include organic carbon storage in permafrost regions on the Qinghai–Xizang (Tibetan) Plateau (QXP). In this study, we reported a new estimation of soil organic carbon (SOC) pools in the permafrost regions on the QXP up to 25 m depth using a total of 190 soil profiles. The SOC pools were estimated to be 17.3± 5.3 Pg for the 0–1 m dept...
متن کامل