Perceptive Classification for Crystal Structure of ABO3 perovskite-type Oxides Using Machine Learning

The goal of machine learning (ML) is to teach a computer system to execute tasks on its own. A machine learning algorithm may be found practically everywhere these days. A frequent application of machine learning algorithms is to extract data and uncover patterns in the data that reveal correlations between different components. Supervised, unsupervised, semi-supervised, and reinforcement learning are all forms of machine learning. Classification, regression, and clustering issues may be solved using ML techniques. Machine learning techniques has been used in this work to predict the lattice type of ABO3 perovskite-type oxide lattices and to determine which approach does this job with the best accuracy. ABO3 perovskite-type oxides of 4,165 samples were used to train the system. There are 13 columns of data for each observation, with a single class column indicating whether the structure is cubic, tetragonal, orthorhombic, or rhombic. Based on the results of the experiments, the Random Forest model performs better than competitors.