Measurement, selection, and visualization of association rules: A compositional data perspective

Association rule mining is a powerful data analytic technique used for extracting information from transaction databases with collection of itemsets. The aim to indicate what item goes (ie, an association rule) in set collected transactions. It extensively text analytics records or social media. Here we use Compositional Data analysis (CoDa) techniques generate new visualizations and insights mining. These CoDa methods show the relationship between itemsets, their strength, direction dependency. Moreover, after expressing each as contingency table, discuss two statistical tests guide identification relevant rules by analyzing relative importance elements table. As example, these investigating negative mood emotions various types headache/migraine events. those comes N1-HeadacheTM, digital platform where individual users record attacks symptoms well daily exposure list potential factors.