# How to find the local similarity between an original data and a reconstructed data set ?

4 views (last 30 days)
Yapo OBOUE on 14 Aug 2019
Answered: Bhargavi Maganuru on 19 Aug 2019
Hi everyone, I am working on data reconstruction. I want to show the local similiraty between the original data and a reconstructed data set. The size of the original data is [300 12 12]=size(A), the size of the reconstructed data is [300 12 12]=size(B). Is there a way to find the similarity between the two data ? Any help could help me. Tahnks

KSSV on 14 Aug 2019
How about a regression coeffiecent/ corellation coefficient and mean square error between the two?
Yapo OBOUE on 14 Aug 2019
Hi dear KSSV thanks for your comment, I used mse like b = mse(A,B)
and got a coefficient b=6.9329e-05. So, I would like to know if there was a possibility of obtaining data with the same size like B after using regression coeffiecent/ correlation coefficient and mean square error between the two, so that the result of the similarity is then plotted.

Bhargavi Maganuru on 19 Aug 2019
You can use “corrcoef” function to find similarity between the two data.