how to determine optimal minimum utility in EFIM algorithm of high utility itemset mining?

Question

I'm supposed to set minimum utility for EFIM algorithm to find high-utility itemset mining

Characterstics of datasets are :

• Number of transactions : 3300K
• Average size of itemsets : 4
• Maximum size of itemsets : 85K
• Average Utility per transaction : 2.3K
• Maximum Utility of transaction : 36000K

I understand that "minimum utility" may vary depending on different circumstances, but I'd like to know if there is a common way to set this value.

Just as if filling mean value is one of the most common way of handling missing value.

I'm assuming the solution for this problem could be either simply using "mean", or half of maximum value. I'm not so sure.

What should I do? Please help

question from:https://stackoverflow.com/questions/65908469/how-to-determine-optimal-minimum-utility-in-efim-algorithm-of-high-utility-items

Answer 1

Waitting for answers