Association Rules over Time

Abstract

after obtaining the results of specific ML methods, in order to discover any additional knowledge hidden in data, and, thus, help the user to understand the results properly.Association Rule Mining (ARM) is a well-known ML method that normally generates a huge amount of association rules, from which it is hard to make the proper decisions easily. An additional problem is presented by the complex form of the results, which are represented as an implication X ⇒ Y , meaning "'if antecedent X is true then its consequent Y is also true". As a result, there we are confronted with two problems: (1) How to discover the relevant association rules in a huge archive, and (2) How to express the relations between attributes in those implication rules visually in a form that is understandable and interpretable by users.Interestingly, the visualization of association rules has rarely been treated in the literature. Indeed, the papers which referred to these topics can be summarized in the following review: The authors in [27] presented a design that is able to handle hundreds of multiple antecedent association rules in a three-dimensional display with minimum human interaction, low occlusion percentage, and no screen swapping. The authors in [16] show that the use of Mosaic plots and their variant, called Double Decker plots, can be applied for visualizing association rules. Ong et al. [23] prototyped the two visualizations, called grid view and tree view, for visualizing the association rules in their application called CrystalClear. Appice and Buono [2] presented a graph-based visualization that supports data miners in the analysis of multi-level spatial association rules, while Herawan et al. [15] proposed an approach for visualizing soft maximal association rules. A very interesting interactive visualization technique, which lets the user navigate through a hierarchy of association rule groups is presented in paper [14]. The authors in paper [18] explored the Hasse diagrams for the visualization of Boolean association rules. Fister et al. [7] proposed a method for identifying dependencies among mined association rules based on population-based metaheuristics [8,12,13] and complex networks, while the paper of Fister Jr. et al. [19] looked for a visualization method capable of telling stories based on mined association rules. However, there are also generic tools, like the CloseViz [4] and the SPMF open-source data mining library Version 2 [9], specialized primarily in pattern mining, offering visual implementation of discovered data mined by ML algorithms that could also be used for visualization of ARM.This chapter proposes monitoring the performance of an athlete involved in sport training sessions during the past seasons (i.e., years). Thus, these seasons are divided into four time periods, where each of the periods captures a quarter of the data. The performances of the athlete are measured using real data obtained by mobile devices worn by the athlete during the training sessions. The performance data (e.g., duration, a...

Efficient Mining Support-Confidence Based Framework Generalized Association Rules

“…Discovering association rules is a fundamental and essential subject in data mining and has been extensively investigated since its inception in [1,2]. Over the past few years, the use of association rule mining in varied application scenarios [3][4][5][6][7] have been intensely discussed [8,9]. The idea consists of discovering causal relationships, where the presence of some items suggests that other items follow from them.…”