A Study on the Performance Of Ct-Apriori And Ct-Pro Algorithms Using Compressed Structures For Pattern Mining
No Thumbnail Available
Date
2010-09
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Automatic discovery and identification of frequently occurring patterns
from very large database is the most desired technique in businesses and
industries. Data mining and knowledge discovery provides several powerful
algorithms for this purpose. Frequent pattern mining is a technique that is used
to discover patterns from large transactional database.
The frequent pattern algorithms perform the mining process in two
phases. In the first phase, all frequent itemsets that satisfy the user specified
minimum support are generated and in the second phase uses these frequent
itemsets in order to discover all the association rules that meet a confidence
threshold. This research analyzes algorithms that produce compact databases
for knowledge discovery from large transaction databases like market basket
database and web log databases. From these compact representations,
association rule mining is applied to mine frequent patterns. In this research,
two variants of Apriori and FP-Growth algorithms, namely, CT-Apriori and
CT-PRO are compared and their performances are analyzed.