期刊名称：International Journal of Computer Science & Technology
印刷版ISSN：2229-4333
电子版ISSN：0976-8491
出版年度：2012
卷号：3
期号：4
页码：869-872
语种：English
出版社：Ayushmaan Technologies
摘要：In recent years the sizes of databases has increased rapidly. This has led toa growing interest in the development of tools capable in the automatic extractionof knowledge from data. The term Data Mining, or Knowledge Discovery inDatabases, has been adopted for a field of research dealing with the automaticdiscovery of implicit information or knowledge within databases.Several efficient algorithms have been proposed for finding frequentitemsets and the association rules are derived from the frequent itemsets, such as theApriori algorithm. These Apriori-like algorithms suffer from the coststo handle a huge number of candidate sets and scan the database repeatedly. A frequent pattern tree (FP-tree) structure for storing compressed and criticalinformation about frequent patterns is developed for finding the complete set of frequent itemsets. But this approachavoids the costly generation of a large number of candidate sets and repeated databasescans,which is regarded as the most efficient strategy for mining frequent itemsets.Finding of infrequent items gives the positive feed back to the ProductionManager. In this paper, we are finding frequent and infrequent itemsets by taking opinions of different customers by using Dissimilarity Matrix between frequent and infrequent items and also by using Binary Variable technique. We also exclusively use AND Gate Logic function for finding opinions of frequent and infrequent items. After finding frequent and infrequent items the apply Classification Based on Associations (CBA) on them to have better classification.
