Visualization of Numerical Association Rules by Hill Slopes

Abstract

Association Rule Mining belongs to one of the more prominent methods in Data Mining, where relations are looked for among features in a transaction database. Normally, algorithms for Association Rule Mining mine a lot of association rules, from which it is hard to extract knowledge. This paper proposes a new visualization method capable of extracting information hidden in a collection of association rules using numerical attributes, and presenting them in the form inspired by prominent cycling races (i.e., the Tour de France). Similar as in the Tour de France cycling race, where the hill climbers have more chances to win the race when the race contains more hills to overcome, the virtual hill slopes, reflecting a probability of one attribute to be more interesting than the other, help a user to understand the relationships among attributes in a selected association rule. The visualization method was tested on data obtained during the sports training sessions of a professional athlete that were processed by the algorithms for Association Rule Mining using numerical attributes.