Year: 2010
Type: conference-paper
Abstract: The dramatic drop in the prices of data collection and storage devices has not only enabled organisations to store almost every activity of their business processes, they can also retain every state of these activities as well. Availability of these masses of data also means that by implementing different data mining techniques we can yield more accurate and useful information to be used for important decision making. One of the key mining techniques on such data is to discover sequential patterns. Most of the existing sequential pattern mining approaches mainly deal with finding the positive behaviour of a sequential pattern that can help in predicting the next event after a sequence of events. In this paper we propose the concept of Negative Behaviour Against the Sequential Pattern (NBASP) that is to discover the events/event-sets which are unlikely to follow the given sequential pattern and discuss its applications in a variety of domains. A comprehensive problem definition and efficient algorithm to discover NBASP is presented.
Institutions: University of Manchester, University of Westminster
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2140270488