Spectral Classification using DSS based on Weighted Random Forest and other mining techniques

NCIDCAS | Special Issue | December-2018 | Published Online: 22 December 2018    PDF ( 348 KB )
Author(s)
S.R. Gedam 1; R. A. Ingolikar 2

1IICC, RTMNU, Nagpur, (India)

2Department of Computer Science, SFS College, RTMNU, Nagpur (India)

Abstract

Spectral classification is an area of great importance for most of the astrologers. Different classification methods have been used for this purpose. Ensemble learning is a classification method which uses different classifiers and result of classification is based on the vote obtained by individual classifier. A decision support system based on ensemble learning technique for spectral classification is devised. The tree based ensemble learning classifier is called weighted random forest classifiers. The classification results obtained using weighted random forest and different mining techniques are compared. It is observed that weighted random forest performs better than other mining techniques.

Keywords
Ensemble learning; Weighted random forest; Decision Support System; Random forest
Statistics
Article View: 52