DocumentCode
3165788
Title
Maximum Entropy Based Significance of Itemsets
Author
Tatti, Nikolaj
Author_Institution
Helsinki Univ. of Technol., Helsinki
fYear
2007
fDate
28-31 Oct. 2007
Firstpage
312
Lastpage
321
Abstract
We consider the problem of defining the significance of an itemset. We say that the itemset is significant if we are surprised by its frequency when compared to the frequencies of its sub-itemsets. In other words, we estimate the frequency of the itemset from the frequencies of its sub-itemsets and compute the deviation between the real value and the estimate. For the estimation we use Maximum Entropy and for measuring the deviation we use Kullback-Leibler divergence. A major advantage compared to the previous methods is that we are able to use richer models whereas the previous approaches only measure the deviation from the independence model. We show that our measure of significance goes to zero for derivable itemsets and that we can use the rank as a statistical test. Our empirical results demonstrate that for our real datasets the independence assumption is too strong but applying more flexible models leads to good results.
Keywords
data mining; maximum entropy methods; Kullback-Leibler divergence; flexible model; independence assumption; independence model; itemsets significance; maximum entropy; Computer science; Data mining; Electronic mail; Entropy; Frequency estimation; Frequency measurement; Itemsets; Predictive models; Proposals; Testing;
fLanguage
English
Publisher
ieee
Conference_Titel
Data Mining, 2007. ICDM 2007. Seventh IEEE International Conference on
Conference_Location
Omaha, NE
ISSN
1550-4786
Print_ISBN
978-0-7695-3018-5
Type
conf
DOI
10.1109/ICDM.2007.43
Filename
4470255
Link To Document