|
Title:
|
PROJECTION BASED SAMPLING FOR MORE EFFICIENT HIGH UTILITY ITEMSET MINING |
|
Author(s):
|
Alva Erwin, Raj P. Gopalan, N.R. Achuthan |
|
ISBN:
|
978-972-8939-23-6 |
|
Editors:
|
António Palma dos Reis and Ajith P. Abraham |
|
Year:
|
2010 |
|
Edition:
|
Single |
|
Keywords:
|
High Utility Itemset Mining, Sampling, Frequent Itemset Mining |
|
Type:
|
Full Paper |
|
First Page:
|
77 |
|
Last Page:
|
84 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
High Utility Itemset Mining is a generalization of Frequent Itemset Mining, where not only the absence or the presence of items, but also the utility of items in the form of quantity and profit are significant. Mining High Utility patterns is a difficult problem especially from a large database due to the combinatorial explosion of patterns to be considered and the inapplicability of the downward closure property for pruning. Sampling can reduce the size of the dataset to be mined, but its usefulness depends on the accuracy of the result and the level of accuracy required for a given purpose. In this paper we propose a projection based sampling algorithm to mine High Utility Itemsets that improves the accuracy of mining compared to simple random sampling. Experiments have been performed on real and synthetic datasets to show the effectiveness of the proposed algorithm. |
|
|
|
|
|
|
