22-03-2014, 10:00 AM
Analysis and improvement of Apriori algorithm
Apriori algorithm.pptx (Size: 708.34 KB / Downloads: 21)
Description
The data mining of association rules is an essential research aspect in the data mining fields.
Association rules reflect the inner relationship of data. The association rule mining is fundamentally important task in the process of knowledge discovery in large database.
We analyzing the classic method of association rules-Apriori algorithm and improved the new algorithm AprioriMend algorithm.
Proposed System:
The proposed system of Apriori algorithm is mainly aimed at identifying the frequent items from the transaction set. To improve the efficiency of Apriori algorithm then introduce the newly proposed algorithm AprioriMend algorithm. The execution times of these algorithms are compared and best of them is identified.
Advantages:
Reduce the number transaction database scans
Shrink number of candidates
Facilitate support counting of candidates.
Apriori algorithm:
User enters into the system and provide data connection, database and transactions table. given the support count and confidence values Apriori algorithm scans the transactions table and generating frequent item sets. Find the all frequent item sets and execution time of the apriori algorithm.
Apriori mend algorithm: enters into the system and provides data connection, transactions table.
Non-Functional Requirements
Usability:
System provides a help document for users.
System provides a convenient type of GUI for handling all operations.
Reliability: System is reliable to support any number of transactions.
Performance: The system responds very quickly in storing and retrieving the data from the database and gives accurate results.
Supportability: The system can be supported for changes even after deployment.
Implementation: The system is implemented using c# .NET
[b]Conclusion:[/b]
This algorithm bases on the structure of Apriori algorithm. It uses the default minimum support to prune the database project, and then grouped the pruned database according to the transaction length, establishing a sub-groups tables to meet the group table quickly find all the characteristics of the frequent item sets. After simulation analysis, we found that AprioriMend algorithm is more excellent than the traditional method Apriori algorithm in the efficiency of performance.