Volume 18, No. 6, 2021
Generating Frequent Pattern Mining From Big Datasets Using Hybrid Apriori And Genetic Algorithm
B.Bazeer Ahamed , D.Yuvaraj
Abstract
A major research field in data mining is the frequent pattern of mining. It has attracted the attention of many researchers since its launch. The generation and collection of data diagonally improve all areas exponentially in scale. Data mining is the use of advanced analytics to determine new information in the method of patterns, trends, and correlations in vast volumes of data. Information retrieval and decision-making involve a flexible and effective way of processing and extracting from Big Data relevant information. One of the most widely used methods for extracting useful knowledge from data is Regular Itemset Mining. However, when this approach is applied to Big Data, the combined outburst of the candidate object sets has become a problem. Recent advances in parallel programming have created excellent tools to address this problem. However, these tools have their technical disadvantages, such as unbiased sharing of data and intercom costs. In our work, we examine the applicability of Frequent Mining Items in the Map-Reduce framework. This method is optimized for use in extremely large datasets. Our approach is similar to FP-growing but uses a different data structure based on algebraic topology.
Pages: 232-242
Keywords: Frequent Pattern Mining, Frequent Item Sets, Hybrid Apriori,Big data