Performance Comparison of Apriori, ECLAT and FP-Growth Algorithm for No Biological Data Genes for Association Rule Learning

Abstract

This project is carried out to study the performance comparison of Apriori Algorithm, ECLAT Algorithm and FP-Growth Algorithm for no biological data genes. There are many genes with no biological data, but for this project we have chosen 4 types of no biological data genes. No biological data genes are genes that have no specific data about themselves such as location, behaviour and function of the genes. Association Rule Learning is a technique implementing big data in finding frequent item-sets. Frequent item-sets are items that occur frequently in the database. The performance of these three algorithms is compared through time efficiency and the ability to process small and large datasets. After the comparison, we can conclude that FP-Growth algorithm is the fastest algorithm for small data-set and Apriori algorithm and ECLAT algorithm takes lesser time to generate the frequent item-sets compared to FP-Growth algorithm.