Could you please give a detail running example for the EFIM-Closed algorithm. I couldn't found one in the paper and I'm stuck at the forward/backward extension checking.

Thank you

Hi,

For the EFIM algorithm, there is some detailed example in the journal paper:

Zida, S., Fournier-Viger, P., Lin, J. C.-W., Wu, C.-W., Tseng, V.-S. (2017). EFIM: A Fast and Memory Efficient Algorithm for High-Utility Itemset Mining . Knowledge and Information Systems (KAIS), Springer, 51(2), 595-625
http://philippe-fournier-viger.com/EFIM_JOURNAL_VERSION%20KAIS%202016.pdf

For the EFIM-closed, there is the conference paper that explains the main idea. Moreover, I made a Powerpoint presentation about EFIM and EFIM-Closed that gives some additional details:
http://www.philippe-fournier-viger.com/EFIM_and_EFIM-Closed_high_utility_mining.pdf

But for EFIM-Closed, there is no journal paper. So I did not write some detailed example for that algorithm.

However, I released the Java source code of EFIM-Closed in SPMF:
http://www.philippe-fournier-viger.com/spmf/

So for more details, you could have a look at the code. This would show you more details about how it works. Actually, there are some implementation details in the code that are not described in the paper for EFIM and EFIM-Closed that improves their performance such as using buffers, etc.

Also since you have the code, you can run it using the debugger to see steps by steps how it works on a test database, and check what is the data stored in the variables etc. Using the debugger, you can get a detailed example.

By the way, if you want some opinion or ideas about EFIM, we can also discuss. Or if you have some questions, I can also try to answer you. But I do not plan to write a detailed example at this point as it would require some times and I am quite busy.

EFIM is a very memory efficient algorithm. I think it has a lot of potential for being extended to other data mining tasks.

Best,

Philippe



Edited 1 time(s). Last edit at 10/03/2017 09:55PM by webmasterphilfv.

Thanks you for your quick replies and advises. I'm working on the discovering of CHUI problem and found that EFIM-Closed excels in both speed and memory usage compares to CHUI-Miner. For a detail example, I've read thoroughly through the EFIM journal paper, but it doesn't have the forward/backward extension checking since the purpose of EFIM is for discovering HUIs.

For the EFIM-Closed slides, page #14 is quite easier for me to understand than the definitions presented in the conference paper but it's only for determining is a HUI a CHUI or not. I'll do some debugging run on the DB_utility.txt later to make it clearer and report back.

Thank you again for your time

You are welcome. Glad that it is faster than CHUI-Miner. By the way, there is also an algorithm called CLS-Miner which is an improvement of CHUI-Miner. It is not in SPMF but if you contact the main author maybe he can give you the code.

Best,

Philippe