Frequent Itemset Matching for Real Time Applications using Reconfigurable Hardware Architecture
J. Samson Immanuel1, G. Manoj2, A. Amir Anton Jone3, P. Esther Jebarani4
1J. Samson Immanuel, Assistant Professor, Department of Electronics and Communication Engineering, Karunya Institute of Technology and Sciences, Coimbatore (TamilNadu), India.
2G. Manoj, Assistant Professor, Department of Electronics and Communication Engineering, Karunya Institute of Technology and Sciences, Coimbatore (TamilNadu), India.
3A. Amir Anton Jone, Assistant Professor, Department of Electronics and Communication Engineering, Karunya Institute of Technology and Sciences, Coimbatore (TamilNadu), India.
4P. Esther Jebarani, Assistant Professor, Kovai Kalaimagal College of Arts and Science, Coimbatore (TamilNadu), India.
Manuscript received on 05 March 2019 | Revised Manuscript received on 17 March 2019 | Manuscript Published on 22 March 2019 | PP: 440-444 | Volume-8 Issue-5S April 2019 | Retrieval Number: ES3461018319/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© 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: Data mining methods remain a quickly developing class of claims that are popular basic usage in numerous fields. An accumulative quantity of data increases the claim for calculating power. Usually human being utilizes enormous size of data and understands that the data and information are widely spread at particular pointer. The algorithms and techniques are known as data mining, remain developed to channel the breach. To utilize the demand for microprocessor systems and use of graphics processing units (GPU) there are numerous methods can be obtained. The added method operates on the hardware accelerators termed as Field programmable gate array (FPGA). Three data mining algorithms nominated for this review: In this apriori algorithm is best to mine the frequent itemsets from the extensive database, and the frequent itemsets are very useful to get the association rule for the discovery of knowledge. In this paper apriori algorithm is modified which reduces the large frequent itemsets and it has implemented in Xilinx Virtex-5 FPGA platform provides up to 5.58 × performance improvement over an equivalent software implementation. Evaluation and investigation are performed for the three selected algorithms using FPGA implementations. To precede with conclusion the investigations executed on common complications, restrictions and resources of various algorithms.
Keywords: The Algorithms and Techniques are Known as data Mining.
Scope of the Article: Communication