Predicting preeminent Machine Learning Approach on Stars
Soumobrata Manna, Vikas Jalodia, Keshav Kumar, Vikas Tripathi, Smita Sharma, Deepika Arora
Abstract
Numerous statistical methods, including “machine learning”, “predictive modeling” and “data mining” are included in predictive analysis.. One of the most intriguing and fascinating recent developments in artificial intelligence is machine learning. With the rise in technology the numbers of algorithms are also increasing for training models and based on the dataset the algorithms are being selected for training a good model with higher accuracy. In this paper I have used a stars dataset imported from Kaggle for predicting the spectral classes of the stars M and O based on the temperature, and have used regression algorithms for predicting it, since it contains continuous real values and regression algorithms work best for this type of cases for predictions and outputs with higher accuracy. By implementing the algorithms, I found that Random Forest Regressor works best with a higher R2_score.