OCCUPANCY BASED PATTERN MINING: CURRENT STATUS AND FUTURE DIRECTIONS
Jhimli Adhikari · INTERNATIONAL JOURNAL OF NEXT-GENERATION COMPUTING · 2020
The main purpose of data mining and analytics is to find novel, potentially useful patterns that can be utilized in real-world applications to derive beneficial knowledge. In recent years, a new measure of pattern interestingness called occupancy of a pattern was introduced to ensure that each pattern found represents a large part of transactions where it appears. Main objective of this measure is to enhance the quality of a pattern. This article surveys recent studies on pattern mining and its applications based on occupancy. The goal of the paper is to provide both an introduction to occupancy based pattern mining (OPM), and a survey of recent advances and research opportunities. Moreover, main approaches and strategies to solve occupancy based pattern mining problems are also presented. The paper also presents challenges and research opportunities of using occupancy measure in other popular pattern mining problems.