European Journal of Molecular & Clinical Medicine

Abstract

Frequent pattern mining is an important field of research in data mining. It has piqued the interest of many researchers since its inception. Data generation and collecting increase in size exponentially diagonally. Knowledge discovery and decision making necessitate the ability to process and extract relevant information from "Big" Data in a scalable and efficient manner. The use of refined analysis to vast volumes of data in order to discover new knowledge in the form of patterns, trends, and associations is known as data mining. Decision-making and information retrieval necessitate a scalable and effective approach for processing and extracting important information from Big Data. Data mining is the sophisticated study of huge amounts of data to discover new information in the form of patterns, trends, and relationships. With the spread of the World Wide Web, the amount of data stored and made available electronically has increased dramatically, and methods for retrieving information from such large amounts of data have grown in importance for both the business and scientific research communities. Frequent Item Set Mining is one of the most widely used methods for extracting relevant information from data. Recent advances in parallel programming have provided excellent methods for overcoming this challenge. Nonetheless, these tools have technical limitations, such as unbiased data sharing and inter-communication costs. In this paper, we investigate the use of Frequent Item Set Mining in the Map Reduce architecture. Big-Frequent-Item set Mining is a new method for extracting big datasets. This approach is designed to work with exceedingly large datasets. Our method is similar to FP-growth, but it employs a distinct data structure based on algebraic topology. We also focused on hybrid Apriori to generate frequent patterns, and the Apriori algorithm's association rule is then optimised using a genetic algorithm. To generate strong association rules, Apriori algorithm association rules were subjected to Genetic Algorithm operators such as selection, crossover, and mutation. To mine recurrent designs with a user-specified lowest provision, a parallel algorithm has been proposed. To compute frequent item sets, the work is distributed among n processors. As a result, the processors will communicate. When compared to other algorithms, the time obligatory to comprehensive the task is very short