I want to use Self organizing map technique of unsupervised learning to classify supernova 1a. How to use spectrum and light curves of Supernova 1a with Self organizing maps (SOM).
What kind of data do you have? Some years ago, I used SOMs to produce a map of some images based on some similarity measure. I guess you will have to visualize the data and use the pixel values and use some clustering technique on it to end up with a SOM. I hope that makes sense.
..
you might find the following article intersting
https://www.researchgate.net/publication/236168314_The_many_faces_of_a_Kohonen_Map
see also J. Vesanto's thesis (ref 10)
.
Article The many faces of a Kohonen Map
I agree with Dr. Varughese: it depends on the kind of data available. In my opinion, the first step could be the identification of the input vectors. E.g. the light curve are sequence of values: can you consider them as the vectors to compare, or, maybe, a part of them (i.e. a window)?
- Badges
- Science topic
- Similar topics
- Papers