Search

Frequent Patterns generation takes much time to execute, compute and the efficiency is less with massive datasets. In this paper, the enhanced FP-Growth (EFP-Growth) based on MapReduce framework using Hadoop approach has been proposed. The new method has high achieving performance with different minimum support. The EFP-growth can works with the large amount of datasets to discover a frequent patterns in a transaction database. With EFP-Growth, many challenges has been solved, to mine frequent patterns such as large memory consumption, high I/O cost, high time and space complexity. In the proposed method, the execution time under different minimum supports and large datasets is decreased.

The Particle Swarm Optimization (PSO) Algorithm is swarm intelligence optimization algorithm. Population's values of PSO algorithm are random which result to a random distribution of search quality and velocity. This paper presents A Modified Particle Swarm Optimization Algorithm using Uniform Design (MPSOUD). Uniform Design (UD) is widely used in various applications and introduced to generate an initial population, in which the population members are scattered uniformly over the search space. Based on abovementioned this technology A Modified Particle Swarm Optimization Algorithm using Uniform Design (MPSOUD) is proposed. Where used UD to resolve the randomness problem of PSO and the time problem. Finally, the performance of MPSOUD algorithm is tested and compared. Tests show that the MPSOUD algorithm faster than the standard PSO algorithm with random populations.