Call For Paper Volume:4 Issue:10 Oct'2017 |

Efficient Algorithms for Mining Compact Representation High Utility Itemsets

Publication Date : 05/10/2015



Author(s) :

MUHASINA C M , SHAHINA K K.


Volume/Issue :
Volume 2
,
Issue 9
(10 - 2015)



Abstract :

High utility itemsets mining from databases is an important data mining task, which refers to the discovery of itemset with high utilities (e.g. high profits). Many studies have been proposed for mining HUIs, including Two-Phase , IHUP , TWU-Mining , IIDS  and HUP-Tree UP-Growth . The former three algorithms use TWDC property to find HUIs. Mainly these algorithms consist of two phases. In first Phase , they find all HTWUIs from the database. In second Phase, HUIs are identified from the set of HTWUIs by calculating the exact utilities of HTWUIs. Although these methods capture the complete set of HUIs but  they may generate too many candidates in Phase I, i.e., HTWUIs, which degrades the performance of Phase II and the overall performance. To reduce the number of candidates in Phase I, various methods have been proposed Recently, Tseng et al. proposed UPGrowth Though the above methods perform well in some cases, their performance degrades quickly when there are many HUIs in the databases. A large number of HUIs and candidates cause these methods to suffer from long execution time and huge memory consumption. When the system are limited (e.g. the memory space and processing power), it is often impractical to generate the entire set of HUIs. Besides, a large amount of HUIs is hard to be comprehended or analyzed by users. In this propose a novel framework in this paper for mining closed high utility itemsets (CHUIs), which serves as a compact and lossless representation..


No. of Downloads :

0


Indexing

Web Design MymensinghPremium WordPress ThemesWeb Development

Efficient Algorithms for Mining Compact Representation High Utility Itemsets

September 30, 2015