Ye, thanks for your reply, Ted. As you know, after construct the FP-tree, we will use the growth fuction to produce the frequent patterns.
As the paper (Mining Frequent Patterns without Candidate Generation) described, we should get the patterns from the bottom of the HeaderTableAttributes (which always are leaf nodes in the pt-tree). So, maybe we just need to growthBottomUp() function. But, i also see growthTopDown() function in the FPGrowth.java. I just confused with this two functions, what are the differeces between them? Wish i describe my question clearly now. thanks On Wed, Jul 28, 2010 at 2:29 PM, Ted Dunning <[email protected]> wrote: > What is it that you don't understand about it? > > If you can give specific questions, you are much more likely to get a good > answer. The paper that you already have should give you most of the > general > hints that you need. From there, you should provide specific questions. > > On Tue, Jul 27, 2010 at 10:13 PM, Andrew Wang <[email protected] > >wrote: > > > Now, only the growth(FPTree tree, MutableLong minSupportMutable, > int > > k, FPTreeDepthCache treeCache, int level, int currentAttribute, > > StatusUpdater updater) function in the FPGrowth.jave cannot be understood > > by > > myself. > > > > Would you please tell how it works? or give me some papers to help > > me to understand it? thanks！ > > > -- http://anqiang1900.blog.163.com/
