0
$\begingroup$

I'm trying to build a model that given the blue line in the graph below (my time series data) and the red lines (features manually labeled by myself) is able to detect similar features/patters in un-labeled data.

My question is whether someone can suggest an algorithm to use in order to achieve that.

enter image description here

$\endgroup$
1
$\begingroup$

The answer is: Yes.

If you are looking for pattern detection in time series, than you can find an efficient solution in this paper Stream Monitoring under the Time Warping Distance using an efficient dynamic time warping approach (DTW).

Yet I am not certain whether you are specifically looking for pattern recognition or whether you wish to detect spikes/increase within a time series. If you are going to the second target, other approaches are more suitably such as you find in this presentation. Change detection in monitoring time series.

If you enhance your question with detail and motivation, then this will have a strong positive effect on the quality and detail of answers.

| cite | improve this answer | |
$\endgroup$

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.