-2
$\begingroup$

I used to utilize dtwclust for my study and I got a problem. I have read a few studies and I understand the DTW distance like below picture.

enter image description here

source: A. Akl, S. Valaee, "Accelerometer-based gesture recognition via dynamic-time warping, affinity propagation, & compressive sensing," IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), pp.2270-2273, 2010.

and I read then "Comparing Time-Series Clustering Algorithms in R Using the dtwclust Package". The notation for DTW is different.

enter image description here

source: Alexis Sard´a-Espinosa, Comparing Time-Series Clustering Algorithms in R Using the dtwclust Package, pp.5-6

I don't understand what m_∅ is. They say it is a per-step weighting coefficient. anybody help me?? just give me clue. Thanks in advance.

$\endgroup$
0
$\begingroup$

That is a phi $\phi$, not a diameter ∅.

The small $m_\phi$ are weight terms, $M_\phi = \sum_i m_\phi$ is the normalization term to make them sum to 1.

This way, you can have different weights.

$\endgroup$

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.