I want to do sequence learning for that I want to find frequent sequential rules ,this rule consider the order of occurrence
$\begingroup$
$\endgroup$
2
-
$\begingroup$ I don't understand why this question is downvoted. $\endgroup$– user122358Commented Sep 9, 2017 at 20:44
-
$\begingroup$ a simple method would be just to count them. If that does not satisfy you, you can use hidden markov model, conditional random fields, or even lstms. $\endgroup$– Gary OngCommented Feb 18, 2022 at 6:52
Add a comment
|
2 Answers
$\begingroup$
$\endgroup$
1
Probably you could try Conditional Random Fields (https://github.com/scrapinghub/python-crfsuite) or a Long Short term Memory RNN (Keras+Tensorflow backend)
-
$\begingroup$ Thanks for reply .does it has any sequence rule mining algorithm implemented .if no can you help me in understanding the library $\endgroup$ Commented Mar 28, 2017 at 4:57
$\begingroup$
$\endgroup$
I know this is possible with the SPMF Java library.
I created a Python wrapper for this recently. Handling sequential rules mining might require some adaption in the wrapper, since I use it mainly for sequential patterns mining, but it's certainly possible.
