کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
385202 660863 2012 10 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
A new mining approach for uncertain databases using CUFP trees
موضوعات مرتبط
مهندسی و علوم پایه مهندسی کامپیوتر هوش مصنوعی
پیش نمایش صفحه اول مقاله
A new mining approach for uncertain databases using CUFP trees
چکیده انگلیسی

In the past, many algorithms have been proposed to mine frequent itemsets from transactional databases, in which the presence or absence of items in transactions was certainly known. In some applications, items may also be uncertain in transactions with their existential probabilities ranging from 0 to 1 in the uncertain dataset. Apparently, the processing in uncertain datasets is quite different from those in certain datasets. The UF-tree algorithm was proposed to construct the UF-tree structure from an uncertain dataset and mine frequent itemsets from the tree. In the UF-tree construction process, however, only the same items with the same existential probabilities in transactions were merged together in the tree, thus causing many redundant nodes in the tree. In this paper, a new tree structure called the compressed uncertain frequent-pattern tree (CUFP tree) is designed to efficiently keep the related information in the mining process. In the CUFP tree, the same items will be merged in a branch of the tree even when the existential probabilities in transactions are not the same. A mining algorithm called the CUFP-mine algorithm is then proposed based on the tree structure to find uncertain frequent patterns. Experimental results show that the proposed approach has a better performance than UF-tree algorithm both in the execution time and in the number of tree nodes.

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Expert Systems with Applications - Volume 39, Issue 4, March 2012, Pages 4084–4093
نویسندگان
, ,