C-Idea

Abstract

Discovering trend reversals between two data cubes provides users with novel and interesting knowledge when the real-world context fluctuates: What is new? Which trends appear or emerge? With the concept of emerging cube, the authors capture such trend reversals by enforcing an emergence constraint. In a big data context, trend reversal predictions promote a just-in-time reaction to these strategic phenomena. In addition to prediction, a business intelligence approach aids to understand observed phenomena origins. In order to exhibit them, the proposal must be as fast as possible, without redundancy but with ideally an incremental computation. Moreover, the authors propose an algorithm called C-Idea to compute reduced and lossless representations of the emerging cube by using the concept of cube closure. This approach aims to improve efficiency and scalability while preserving integration capability. The C-Idea algorithm works à la Buc and takes the specific features of emerging cubes into account. The proposals are validated by various experiments for which we measure the size of representations.