Frequent Pattern Mining Algorithms: A Review
Sukhbir Singh1, Dharmender Kumar2
1Sukhbir Singh, Master of Technology, Department of Computer Science and Engineering, GJUS & T, Hisar, Haryana, India.
2Dharmender Kumar, Assoc. Prof., Department of Computer Science and Engineering, GJUS & T, Hisar, Haryana, India.
Manuscript received on September 10, 2014. | Revised Manuscript Received on September 12, 2014. | Manuscript published on September 18, 2014. | PP: 01-07 | Volume-1 Issue-10, September 2014.
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Mining frequent patterns is one of the most important concepts of data mining. Frequent pattern mining has been a highly concerned field of data mining for researcher for over two decades. Several algorithms have been developed for finding frequent itemsets from the databases. The efficiency of these algorithms is a major issue since a long time and has captured the interest of a large community of researchers. In Literature review it is found that great effort has been made in this area so far to development of efficient and scalable algorithms for frequent itemset mining in various types of databases due to their importance in various fields. In 1993, R. Agrawal and R. Srikant first proposed the most classical association rule mining algorithm named as Apriori algorithm. But Apriori has two major drawbacks: large number of candidate itemsets generation and large no of database scan. Like most of the association rule algorithms, first it discover minimal frequent itemsets, then it discover the maximal frequent itemsets by using these minimal frequent itemsets, so all approach of this type take large time to find maximal frequent itemsets and needed large number of database scan, also not suitable for the continuous changing database. To overcome these problems, extensive work have done by many researchers, by enhancement and modification on basic algorithms like Apriori algorithm, FP growth algorithm, Eclat algorithm, and MFI algorithm etc. Maximal frequent itemset (MFI) was proposed by Bayard in the year 1998. (MFI) used to find maximal frequent item. After that lots of improved approaches have been proposed to efficiently mining the maximal frequent pattern such as Mafia, GenMax Smart-Miner etc. The present paper provides an overview of various frequent pattern mining algorithms with the expectation that it would serve as a reference material for researchers in this field.
Keywords: Apriori Algorithm, Association Rules, Boolean matrix, Data Mining, Frequent Itemset, Maximal Frequent Itemset (MFI), Maximal frequent itemset first (MFIF).