A Systematic Review of Deep Learning Approaches for Natural Language Processing in Battery Materials Domain

Natural Language Processing (NLP) acts towards the processing of linguistics between human and computer. The application of NLP in the chemical industry has proven to be a boon in the past decade. This paper presents a survey of deep learning (DL) models on NLP fundamentals for battery materials domain-related research. Various DL models like convolutional neural networks, recursive neural networks, recurrent neural networks, long short-term memory networks, and attention networks have been discussed while comparing their efficacy for different NLP tasks. Innovations in presented DL models provide enhanced performance for the NLP-based tasks. This paper will enlighten researchers, to get an insight into the current practices and scope for future developments on battery materials. They can apply NLP techniques for various applications like structural recognition and extraction, understanding of chemical features without a domain expert, and understanding the chemical structure and summarization approaches. Moreover, this paper will help chemists and researchers in the chemical industry to explore various areas to work upon. To empower this, various applications and future scope is discussed in this paper.