-1
$\begingroup$

Let's say I have a file named accepted_D.dat that looks like this. This file gives 7247 and variables that describe them:

header: "model ID", "Z", "X", "M", "L", "T", "non-linear period"

1   0.00038 0.75053 0.50    35  6000    0.75346
2   0.00038 0.75053 0.50    35  6050    0.72079
3   0.00038 0.75053 0.50    35  6100    0.69229
4   0.00038 0.75053 0.50    35  6150    0.66689
5   0.00038 0.75053 0.50    35  6200    0.64382
...
7247    0.00061 0.74996 0.80    105 6800    0.61635

And I have a second file called FourierParams_SetD_I.dat. This file gives more information (more variables) about the 7247 models.

header: Model ID", "log(P)", "skewness", "acuteness", "mean mag", "err in mean mag", "A1", "A1_err", "phi1", "phi1_err", "R21", "R21_err", "R31", "R31_err", "P21", "P21_err", "P31", "P31_err

1   -0.123  -0.289  0.742   0.404   0.000   0.005   0.000   0.083   0.000   0.102   0.000   0.011   0.000   4.174   0.000   1.880   0.004
2   -0.142  -0.155  0.387   0.413   0.000   0.035   0.000   6.261   0.000   0.619   0.000   0.340   0.000   4.231   0.000   1.527   0.000
3   -0.160  -0.123  0.391   0.423   0.000   0.051   0.000   6.155   0.000   0.640   0.000   0.430   0.000   4.347   0.000   1.681   0.000
4   -0.176  -0.100  0.422   0.432   0.000   0.068   0.000   6.059   0.000   0.552   0.000   0.438   0.000   4.407   0.000   1.795   0.001
5   -0.191  -0.091  0.477   0.440   0.000   0.088   0.000   5.964   0.000   0.453   0.000   0.412   0.000   4.395   0.001   1.806   0.001
...
7247    -0.210  -0.337  inf -0.658  0.000   0.144   0.000   0.099   0.000   0.054   0.000   0.071   0.000   0.545   0.001   2.137   0.001

Using columns R21 and and logP in the 2nd file, we get this plot.

enter image description here

Do you see the various undulating sequences? I want to look for pattern and study how the Z, X, M, L, T variables affect these sequences (and their width/spread).

Is there a suitable technique for this, whether a clustering method or something completely different? I am completely new to clustering techniques and just learned about K-means clustering.

$\endgroup$

Your Answer

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

Browse other questions tagged or ask your own question.