Data Mining: Hash tree based support counting

Data Mining: Hash Tree based support counting

Hash tree is a very quick way to search an item. When there are many itemsets, hash tree could be used to find out if a given itemset has got required support count. But, how do we construct hash tree? The links I came across were very abstarct to define the hash tree implementation.

Suppose we want to insert (i.e. hash)
the following 3-itemsets into the tree

We have taken hashing function h(x) = N mod 3. This has three possible values for h(x) = {0, 1, 2}. Each value is a branch of a Hash tree node. So, each node will have three branches.

Now, lets insert Itemsets = {8, 7, 5} {9, 3, 6}

Start with I1 = {8, 7, 5}


Remember, that we are considering 8 because it is lavel 1 of the tree. Next level, we will consider 7 and the next with 5.

The first item = 8
h(8) = 8 mod 3 = 2


I2 = {9, 3, 6}
The first item = 9
h(9) = 9 mod 3 = 0


Level 1
Now, the secons item of set I1 = 7
h(7) = 7 mod 3 = 1


Now, the secons item of set I2 = 3
h(7) = 3 mod 3 = 0


  One thought on “Data Mining: Hash tree based support counting

  1. Jarek
    November 14, 2012 at 7:57 PM

    Lots of errors in your document, changing layout of the “slides” as well as constantly changing itemsets .. (8,7,1) then (8,7,5) and then even (8,7,6). The article needs to be properly restructured. It appears to me you didnt really know yourself how to demonstrate it for the audience..


    • noName
      August 1, 2013 at 3:53 PM

      +1, this article is very confusing and should get a makeover


  2. shruti
    October 8, 2015 at 8:56 AM

    its irrelevant and not useful…the example demonstration is wrong


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: