Radio galaxies classification system using machine learning techniques in the IoT Era

Astronomy and astrophysics data sets have been increasing over the last decade as many new telescopes and detectors have been launched. High-redshift radio galaxies are powerful radio sources that are the ideal targets to discuss the evolution of Hi; thus, they are one of the key points to understand the universe evolution and formation. The cloud computing systems are applicable with IoT considering the processing time over training data using ANN. Machine learning is a subfield of AI and it is used by scientists for prediction or classification purposes considering the input data. Machine learning algorithms have become increasingly popular among astronomers and are now used for a wide range of astrophysical calculations and fields. This paper proposes five types of machine learning algorithms, namely back-propagation neural networks (BPNN), decision tree algorithm (DT), gradient boosting classifier algorithm (GBA), radial basis function neural network (RBFNN), and support vector machine (SVM). The machine learning models are implemented to classify and compare the results of high-redshift radio galaxies by their location in the sky in ELAIS-N1, ELAIS-N2, the Lockman hole, VIMOS fields, in order to increase the performance efficiency, accuracy and improve our confidence considering the critical nature of the calculations in redshift galaxies. When 100 instances were considered, back-propagation neural networks achieved an accuracy rate of 70%; however, when 200 instances were considered, radial basis function neural networks achieved an accuracy rate of 88.2%.

Automated summary

This paper discusses the use of machine learning algorithms in classifying and comparing high-redshift radio galaxies. The results show that different machine learning algorithms have different accuracy rates when applied to different datasets.