Ant colony optimization for mining gradual patterns

Gradual pattern extraction is a field in (KDD) Knowledge Discovery Databases that maps correlations between attributes of data set as gradual dependencies. A dependency may take form "the more Attribute K , the less L". In this paper, we propose an ant colony optimization technique uses probabilistic approach to learn and extract frequent patterns. Through computational experiments on real-world sets, compared performance our ant-based algorithm existing item found out outperforms later especially when dealing with large sets.