Event Mining for Knowledge Delivering in Real-Time

摘要

In this paper we will present the event mining. It is promising area of research and applications. Moreover, event mining is one of the tools for knowledge delivering in real-time. Data mining have many industrial and scientific applications (Mattison, 1997; Amat, 2003; Shawa, Subramaniama, Tana and Welgeb, 2001). Data for these type of models can be loaded from various sources. We can distinguish two main types of input data: databases that stores information about object attributes and databases that stores information about transactions. Transactions databases reflects the events in the real word. The algorithms for finding associations in large databases have been introduced by Agrawal, Imielinski and Swami (1993). They have considered database of customer transactions. Each transaction has consisted of items purchased by a customer in a visit. The existing algorithms consider limited information about the events. Recently, it can be observed increased importance of events in modeling and understanding complex systems. Luckham (2002) provides us with a framework for thinking about complex events and for designing systems that use such events (see also (Perrochon, Mann, Kasriel and Luckham, 1999)). Events are especially challenging for real-time analysis. Gartner Inc., a technology research and advisory firm, defines real-time as the complete compression of lag between the detection of an event, the reporting of that event, the decision-making, and the response. They further observe that the RTE (Real-Time Enterprise) is an enterprise that competes by using up-to-date information to progressively remove delays to the management and execution of its critical business processes. Therefore, real-time computing might be the focal point of IT departments because it allows companies to provide on-line information for effective decision making.