Reduction of Frequent Itemsets Mining in Big Data with the Help of FP Algorithm and Msegt-Tree
Srinivasa Rao Divvela1, V Sucharita2
1Srinivasa Rao Divvela, Assistant Professor, Department of CSE, Lakireddy Bali Reddy College of Engineering, Mylavaram.
2Dr V Sucharita, Professor, Department of CSE, Narayana Engineering College Gudur, India.
Manuscript received on January 12, 2020. | Revised Manuscript received on January 22, 2020. | Manuscript published on February 10, 2020. | PP: 2169-2172 | Volume-9 Issue-4, February 2020. | Retrieval Number: D1666029420/2020©BEIESP | DOI: 10.35940/ijitee.D1666.029420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. 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: Frequent itemset mining is very crucial to minimize the cost and time of executions but when considering multiple distributed data streams in big data the frequent itemset mining has been a little cost consuming and taking more space and time complexity. In this paper we reduce the load and minimize the cost while minimizing the space and time complexities of the process by using reduction mechanism and indexing structures for preserving complexities. A 2-level architecture modal which will be helpful in handling the distributed data streams where the root node will be in level-0 and local nodes at level-1 is proposed. Each local node will evaluate the patterns in their specific data stream using the algorithm ‘FP’ which will help in lessening the burden on the root node and will be sent to root. With help of the patterns received from local nodes the root will generate a global pattern set.
Keywords: Frequent Itemset Mining, Distributed Data Streams, Indexing Structures, Space and Time Complexity.
Scope of the Article: Data Mining