2
$\begingroup$

I have multiple sequential data that lead up to an event. All I want to do if find the equivalent sequential data or a pattern that leads to the event. The data is also multivariate so I want to consider multiple features rather than just a column. Help will be greatly appreciated.

$\endgroup$
2
  • $\begingroup$ Do you mean that an event corresponds to exactly one input sequence (i.e. you must have a "full" sequence for the event to appear), or that that an event may appear at any point of the sequence? $\endgroup$
    – Erwan
    Feb 21 '20 at 1:14
  • $\begingroup$ There is a sequence of inputs that result in an event to occur but I am focusing on the last 20 inputs before the event occurs because the size of the input space can be of varying size. $\endgroup$
    – Kyrios
    Feb 21 '20 at 1:47
1
$\begingroup$

I'm not entirely sure but it looks like sequence labeling might be what you need:

  • Sequences of varying length
  • Supervised: you would need to train a model with a sample of sequences annotated with a label at every step (not sure that this is your use case?)
  • Can handle any number of features

Conditional Random Fields is the state of the art method, there are many libraries available.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.