Frequent itemset mining has been first proposed in, in the context of market basket analysis, as the first step of the association rule extraction process. The problem of discovering relevant changes in the history of itemsets and association rules has been already addressed by a number of research papers ,Active data mining was the first attempt to represent and query the history pattern of the discovered association rule quality indexes. It first mines rules, from datasets collected in different time periods, by adding rules and their related quality indexes (e.g., support and confidence to a common rule base. The application of frequent itemset mining and association rule extraction algorithms to discover valuable correlations among data has been thoroughly investigated in a number of different application contexts (e.g., market basket analysis , medical image processing . The steady growth of business-oriented applications tailored to the extracted knowledge prompted the need of analyzing the evolution of the discovered patterns. Since, in many business environments, companies are expected to reactively suit product and service provision to customer needs, the investigation of the most notable changes between the set of frequent itemsets or association rules mined from different time periods has become an appealing research topic This paper focuses on change mining in the context of frequent itemsets by exploiting generalized itemsets to represent patterns that become rare with respect to the support threshold, and thus are no longer extracted, at a certain point. Previous approaches allow both keeping track of the evolution of the most significant pattern quality indexes and discovering their most fundamental changes. the discovery of higher level correlations, in the form of generalized itemsets, issues new challenges in the investigation of pattern temporal trends. Item set generalization may allow preventing infrequent knowledge discarding. a novel kind of dynamic pattern, namely theHIstory GENeralized Pattern (HIGENs) .AHIGEN compactly represents the minimum sequence of generalizations needed to keep knowledge provided by a not generalized itemset frequent, with respect to the minimum support threshold, in each time period. If an itemset is frequent in each time period, the corresponding HIGEN just reports its support variations.