Application of Self-Organizing Map to Stellar Spectral Classifications

TL;DR

This paper demonstrates the use of Self-Organizing Maps to classify stellar spectra into spectral types with high accuracy, providing an unsupervised and efficient method for astronomical object classification.

Contribution

It introduces an unsupervised SOM-based approach for stellar spectral classification that does not require training examples and effectively clusters high-dimensional spectral data.

Findings

Achieved approximately 92.4% classification accuracy.

Identified 7 spectral clusters corresponding to stellar types O to M.

Misclassifications occurred mainly into neighboring clusters.

Abstract

We present an automatic, fast, accurate and robust method of classifying astronomical objects. The Self Organizing Map (SOM) as an unsupervised Artificial Neural Network (ANN) algorithm is used for classification of stellar spectra of stars. The SOM is used to make clusters of different spectral classes of Jacoby, Hunter and Christian (JHC) library. This ANN technique needs no training examples and the stellar spectral data sets are directly fed to the network for the classification. The JHC library contains 161 spectra out of which, 158 spectra are selected for the classification. These 158 spectra are input vectors to the network and mapped into a two dimensional output grid. The input vectors close to each other are mapped into the same or neighboring neurons in the output space. So, the similar objects are making clusters in the output map and making it easy to analyze high dimensional data. After running the SOM algorithm on 158 stellar spectra, with 2799 data points each, the output map is analyzed and found that, there are 7 clusters in the output map corresponding to O to M stellar type. But, there are 12 misclassifications out of 158 and all of them are misclassified into the neighborhood of correct clusters which gives a success rate of about 92.4%.