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.
Asked
Active
Viewed 311 times
2
-
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? – Erwan Feb 21 '20 at 01:14
-
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. – Kyrios Feb 21 '20 at 01:47
1 Answers
1
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.
Erwan
- 24,823
- 3
- 13
- 34