Hi. I was coding the hui-miner ( Mining High Utility Itemsets without Candidate Generation ) in C#, which I encountered a problem.
There is no problem in 1-itemset and 2-itemset items. but in k-itemset, for example :

To create "dfgb" :

u({dfgb}, Each transaction) = u({dfg}, Each transaction) + u({fgb}, Each transaction) - u({fg}, Each transaction)

It has no problems programming, but we do not get the correct answer.

var items_split = number_split.Split(',');
var part_1 = string.Join(",", items_split.Take(items_split.Count() - 1));
var part_2 = string.Join(",", items_split.Skip(1));
var common = string.Join(",", (items_split.Take(items_split.Count() - 1).Skip(1)));

According to the above code:

items = d,f,g,b
items_split = d f g b
part_1 = d,f,g
part_2 = f,g,b
common = f,g

Hi,

In HUI-Miner, you should only combine two itemsets if they are identical except for one item. Thus, you should not combine the itemset

d,f,g

with the itemset

f,g,b

because they have two different items (d and b).

This is one part of the problem.

Best, regads

Thanks a lot

So which items can I combine together?

for dfgb

d,f,g | f,g,b (two different , d and b)
d,f,g | d,g,b (two different , f and b)
d,f,g | d,f,b (two different , b and g)
d,f,b | f,g,b (two different , d and g)

Hello,

In algorithms like HUI-Miner, the items are sorted according to a total order.

What is a total order? It means that there is some order between the items. For example, it could be the alphabetical order. According to the alphabetical order, an item "a" must be processed before an item "b", and "b" must be processed before an item "c".

The algorithms like HUI-MINER will follow this order to generate the itemsets.

Consider that you have the items A, B, C, D.

By following the alphabetical order, you can combine:

AB with AC to generate: ABC

but you cannot combine:

BC with AC to generate ABC because it does not follow the order.


To explain this more clearly, you can only combine two itemsets if they have a single item that is different, and those are the largest items according to the order.

Another example:

You can combine ABC with ABD to generate ABCD

But you cannot combine BCD with ABD to generate ABCD because the order is not respected.

If you apply this simple principle, it will solve your problem.

By the way, in my example, I use the alphabetical order but actually in HUI-Miner they do not use the alphabetical order. If I remember well they use the ascending order of TWU values to sort the items instead.

thanks for your response , best regards