EHMIN: Efficient approach of list based high-utility pattern mining with negative unit profits

Abstract

High-utility pattern mining is an important sub-literature in the data mining literature. This literature discusses the discovery of useful pattern information from large databases by considering not only supports of patterns but also profits and quantities of items. This literature has the potential to be applied to various problems in the real world, so many methods for the improvement of the algorithm performance have been studied. Moreover, there have also been attempts to extend the flexibility of this literature. The traditional approaches in this literature considered the positive unit profits of items in a given database only. However, this literature can take extended flexibility into account by considering negative as well as positive unit profits of the items. In this paper, we suggest an efficient approach for mining high-utility patterns with negative unit profits. Moreover, the experimental performance tests, which are performed on various real and synthetic datasets in this paper, show that the proposed algorithm has a better performance than the state-of-the-art methods in this literature in terms of the runtime, memory usage, and scalability.