Association Rules and Statistics

Martine Cadot (1) , Jean-Baptiste Maj (1) , Tarek Ziadé

1 PAROLE - Analysis, perception and recognition of speech

In this paper, the two kinds of knowledge extracted by Statistics and by Association Rules, both working on the same data structured in tables [Objects *x* Properties], are compared. These two methods establish the links between properties of objects. When the amount of data properties increases, Statistics may be replaced by Association rules for better performance. Pros and cons of using Association rules instead of Statistics are described in this paper.

Mots clés: association rules; statistics

Domaines: Statistiques [math.ST]; Autre; Apprentissage [cs.LG]; Intelligence artificielle [cs.AI]; Recherche d'information [cs.IR]; Algorithme et structure de données [cs.DS]

Martine Cadot, Jean-Baptiste Maj, Tarek Ziadé. Association Rules and Statistics. John Wang, Montclair State University, USA. Encyclopedia of Data Warehousing and Mining, Idea Group Inc., pp.74--77, 2005, Vol. 1. ⟨hal-00578418v2⟩