A Novel Crystal Structure Prediction Using Hybrid Method

Abstract

Chemical compositions are used to predict the crystal structure in solid state of new materials. To finding the crystalline arrangements of materials for major unsolved problems in materials science for their chemical compositions. Crystal structure prediction is one of the foremost methods for discovering new materials. In this paper, we propose a deep and machine learning model for approach to classification of the crystal structure. The more than 5000 dataset were used to predict the crystal structure. In previous work, various machine learning models were used for predicting the crystal structure. In this work, we fuse the machine learning and deep neural network algorithm. The dataset were trained and tested to this crystal structure prediction. In this model we evaluate the dataset to classification the models with high accuracy. Our approach, ANB-NET(AlextNet with Naive Bayes) classifier get the best accuracy to predict the crystal structure and time complexity is less than other model.