Frequent item set mining and association rule mining is the key tasks in knowledge discovery process. Various customized algorithms are being implemented in Association Rule Mining process to find the set of frequent patterns. Though we have many algorithms apriori is one of the standard algorithm for finding frequent itemsets, but this algorithm is inefficient because of several scans of database and more number of candidates to be generated. To overcome these limitations, in this paper a new algorithm called Coalesce based Binary Table is introduced. Through this algorithm the given database is scanned only once to generate Binary Table by which frequent-1 itemsets are found. To progress the process, infrequent-1 itemsets are identified and removed from the Binary Table to rearrange the items in support ascending order. To each frequent-1 itemset find Coalesce matrix and Index List to generate all frequent itemsets having the same support count as representative items and the remaining frequent itemsets are obtained in depth first manner. The significant benefits with the proposed method are the whole database is scanned only once, no need to generate and check each candidate to find the set of frequent items. On the other hand frequent items having the same support counts as representative items can be identified directly by joining the representative item with all the combinations of Coalesce matrix. So, it is proven that coalesce based Binary Table is panacea to cut short the time in identifying the frequent itemsets hence the efficiency is improved.
The complexity of handling the scalability problem of huge data can be reduced with parallel processing. The efficiency of parallel processing changes as the number of processors or number of threads change. Parallel processing is more appropriate for the field like data mining as it is the technique of analyzing large quantities of data to extract useful knowledge. Data mining is very much essential to the modern society as more and more data is being collected from various fields. Experiments are conducted to test the run time efficiency of the apriori algorithm on dual core processor by changing the number of threads for different databases at different support counts. This paper also present the comparison of real time, user time and system time with multiple threads on dual core compared to sequential implementation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.