Mining Big Sources using Efficient Data Mining Algorithms
Data mining algorithms are widely used in the real world application in order to discover knowledge from large data sources. These algorithms work on historical data to analyze data in order to bring about trends or patterns. Association rule mining or frequent item set mining is very useful in applications like inductive databases, query expansion and others. A frequent itemset is the itemset when a set of records are repeated for specified number of times in a given dataset. When such frequent itemset is no present in other frequent itemset, it is named as maximal itemset. When it is not as part of other itemset, them it is called closed itemset. These itemsets are used to extract patterns or trends in the real world applications that support in decision making. Recently Uno et al. proposed data mining algorithms to discover maximal itemsets, closed itemsets and frequent itemsets. In this paper we practically explore those algorithms. We implement them in a prototype application and the empirical results reveal that they are very useful for many data mining solutions.
Shoban Babu Sriramoju