Optimization of Association Rule Mining using FP_Growth Algorithm with GA
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2017, Vol 5, Issue 4
Abstract
Frequent pattern mining is one of the active research themes in data mining which covers a broad spectrum of data mining tasks viz. Association rules, correlations, causality, ratio rules, emerging patterns etc. In this paper, we expand the horizon of frequent pattern mining by introducing an efficient algorithm for mining multi-level and multi-dimensional frequent patterns with flexible support constraints. We analyze the scalability of our algorithm and study its performance on different data sets. We have tested our two hybrid algorithms viz. Apriori+fpga and firefly+fpga with traditional apriori and ga. The objective of this paper is to compare the performance of the genetic algorithm for association rule mining by combining it with other algorithms. The algorithms when tested on abalone dataset that indicates that the accuracy depends mainly on the fitness function which is the key parameter. The crossover probability brings changes in convergence rate with minimal changes in accuracy. The size of the dataset and relationship between its attributes also plays a role in achieving the optimum accuracy. Theoretical analysis and experimental results show that the performance of firefly+fpga is better than other algorithms, however the performance of apriori+fpga was found better than traditional apriori and ga.
Authors and Affiliations
Abhishek Kumar Singh, Deepak Sinwar
Keywords
Related Articles
- EP ID EP23861
- DOI http://doi.org/10.22214/ijraset.2017.4155
- Views 308
- Downloads 9
Enhanced Clustering Technique to Find Replica Nodes in WSN
Sensor networks are highly distributed networks of small, lightweight wireless nodes, deployed in hundreds of thousands to monitor the environment or system by the measurement of physical parameters such as temperature,...
Design Of Pulse Triggered Flip-Flop And Analysis Of Average Power
In this brief, Pulse-triggered FF (P-FF) is a single-latch structure which is more popular than the conventional transmission gate (TG) and master–slave based FFs in high-speed applications. Low power design has become...
Secret Image Sharing Using Quick-Response Code Generation Technique
Patent safety and verification have become increasingly more important in regular life. The digital watermark is one of the methods invented to handle this problem. In this paper, a digitally invisible watermark is inse...
Object Collision Avoidance with Train Using Android System
Railways are providing Eco-Friendly transport system for the mankind. Train accidents can happen very often due to safety violations which results from human errors or limitations in the operation of the existing system...
Study On SQL Injection Attacks: Detection And Prevention
SQL injection also referred to as SQLI in short. It is the most dangerous way in which our data can be hacked. Hackers can get to our databases prudently through Web Interface and can delete, modify the important conten...