Search

Credit card fraud is a serious and ever-growing problem with billions of dollars lost every year due to fraudulent transactions. Fraud has always been present and will always be. It is also ever changing, as the technology and usage patterns change over time, which makes CCFD (credit card fraud detection) a particularly hard problem. Traditionally, fraud detection relied solely on domain experts’ detection rules, but in the past decade or two, such solutions are being augmented with data mining models for fraud detection. The progress in this area is impeded both by the sensitive nature of the data and great commercial potential – the industrial solutions are understandably kept secret and authentic datasets are rare and few. In this paper we study the CCFD problem with its typical problems and state of the art solution. We survey the recent literature and bring a structured overview of relevant fraud detection features and data mining approaches to this problem.

Currently, the World Wide Web is mostly composed of isolated and loosely connected "data islands". Connecting them together and retrieving only the information that is of interest to the user is the common Web usage process. Creating infrastructure that would support automation of that process by aggregating and integrating Web data in accordance to user's personal preferences would greatly improve today's Web usage. A significant part of Web data is available only through the login and password protected applications. As that data is very important for the usefulness of described process, proposed infrastructure needs to support authorized access to user's personal data. In this paper we propose a semantically enhanced Web portal that presents unique personalized user's entry to the domain-specific Web information. We also propose an identity management system that supports authorized access to the protected Web data. To verify the proposed solution, we have built Sweb - a semantically enhanced Web portal that uses proposed identity management system.

Data warehouse is a copy of transaction data specifically structured for query and analysis [1]. Within data warehouse, data is usually stored in a dimensional model. On-Line Analytical Processing (OLAP) is an approach to analysis and reporting that enables a user to easily and selectively extract and view data from different points of view based on a multidimensional data structure called a cube [2]. There are numerous data warehousing platforms available on the market. Exhaustive data warehousing platform comprises of data warehouse server, OLAP server and some data mining functionality. There are no present standards to support interoperability between different OLAP systems. In this article, two major initiatives towards standardization are examined: XMLA and JOLAP. Web service is a piece of business logic, located somewhere on the Internet, that is accessible through standard-based Internet protocols such as HTTP or SMTP [3]. Web services are neither language nor platformspecific. Using Java programming language, general purpose client tool prototype for standardized OLAP data access is developed. Client communicates with OLAP data source through web service and is based on XMLA 1.1. specification. It is concluded that the client encompasses satisfactory set of functionalities.