To prevent spam users, you can only post on this forum after registration, which is by invitation. If you want to post on the forum, please send me a mail (h DOT m DOT w DOT verbeek AT tue DOT nl) and I'll send you an invitation in return for an account.
The behavior of Flexible Heuristics Miner in ProM 6.5.1
Dear Sirs,
Recently, I fould a little problem about Flexible Heuristics Miner(FHM) in ProM 6.5.1. I think the FHM does not perform well in some situations. But according to the paper about the Flexible Heuristics Miner, such problem should not appear. So I think maybe the program for HM has a problem.
For the example log which contains 3 types of traces :<A,C,D,F>,<A,D,C,F>,<A,C,E,D,F>, pretend each trace appears 100 times in the log. According to the paper of the FHM, activity A should have both {C,D} and {C} as outputs (I mean after activity A there should be a OR-Split). But the result from the FHM software from ProM 6 seems to be wrong? Im I correct? I have shown the mining result for the example log in the picture bellow.
![Image: https://mail.google.com/mail/u/0/?ui=2&ik=9ab8883c2a&view=fimg&th=1586e82de0c32ab7&attid=0.1&disp=emb&realattid=ii_1586e7f639248cf4&attbid=ANGjdJ9yvFxDqPl10Hsvj4NT8bks0bBcL74DmPXipj2r8T1wzqwiUH2Ehj8lERnxhv55EP_Wc_3DPw4CWGAM2AtiSQjcgLYVywWx9_cC9H-EEUTce0YI_EJAhixW8Jg&sz=w824-h352&ats=1479385103992&rm=1586e82de0c32ab7&zw&atsh=1 1](https://mail.google.com/mail/u/0/?ui=2&ik=9ab8883c2a&view=fimg&th=1586e82de0c32ab7&attid=0.1&disp=emb&realattid=ii_1586e7f639248cf4&attbid=ANGjdJ9yvFxDqPl10Hsvj4NT8bks0bBcL74DmPXipj2r8T1wzqwiUH2Ehj8lERnxhv55EP_Wc_3DPw4CWGAM2AtiSQjcgLYVywWx9_cC9H-EEUTce0YI_EJAhixW8Jg&sz=w824-h352&ats=1479385103992&rm=1586e82de0c32ab7&zw&atsh=1)
Recently, I fould a little problem about Flexible Heuristics Miner(FHM) in ProM 6.5.1. I think the FHM does not perform well in some situations. But according to the paper about the Flexible Heuristics Miner, such problem should not appear. So I think maybe the program for HM has a problem.
For the example log which contains 3 types of traces :<A,C,D,F>,<A,D,C,F>,<A,C,E,D,F>, pretend each trace appears 100 times in the log. According to the paper of the FHM, activity A should have both {C,D} and {C} as outputs (I mean after activity A there should be a OR-Split). But the result from the FHM software from ProM 6 seems to be wrong? Im I correct? I have shown the mining result for the example log in the picture bellow.
Answers
-
Hi Yaguang,
Thank you for your question.
There are indeed some issues with the heuristics miner, which are investigated by Felix Mannhardt. See f.i. http://www.win.tue.nl/promforum/discussion/comment/1821/#Comment_1821
At some point in time we might release a new version of the heuristics miner.
Joos Buijs
Senior Data Scientist and process mining expert at APG (Dutch pension fund executor).
Previously Assistant Professor in Process Mining at Eindhoven University of Technology -
Thank you:). Looking forward to the new edition.
-
Dear Yaguang,
I finished my work on the Heuristic Miner. It is not released in the ProM package DataAwareCNetMiner (sorry for the cryptic name). You can find it in the current nightly build.
It comes with an interactive plug-in: Interactive Heuristic Miner.
Could you try this tool on the event log and see whether it gives the desired output?
-
Howdy, Stranger!
Categories
- 1.6K All Categories
- 45 Announcements / News
- 224 Process Mining
- 6 - BPI Challenge 2020
- 9 - BPI Challenge 2019
- 24 - BPI Challenge 2018
- 27 - BPI Challenge 2017
- 8 - BPI Challenge 2016
- 68 Research
- 1K ProM 6
- 393 - Usage
- 287 - Development
- 9 RapidProM
- 1 - Usage
- 7 - Development
- 54 ProM5
- 19 - Usage
- 187 Event Logs
- 32 - ProMimport
- 75 - XESame