Looking Through Glass: Knowledge Discovery from Materials Science Literature using Natural Language Processing

This is the authors' abstract. We don't add key points for this paper.

Most of the knowledge in materials science literature is in the form of data such as text and images. Here, we present a framework natural language processing, which automates text and image and precision knowledge extraction from inorganic glasses'. The abstracts are automatically categorized using latent Dirichlet (LDA), providing a way to classify and search semantically linked. Similarly, a comprehensive summary of images and plots are using the 'Caption Cluster Plot' (CCP), which provides direct access the images buried in the papers. Finally, we combine the LDA and CCP with chemical elements occurring in the manuscript to present an 'Elemental', a topical and image-wise distribution of chemical elements in the. Overall, the framework presented here can be a generic and powerful to extract and disseminate material-specific information on-structure-processing-property dataspaces, allowing insights into problems relevant to the materials science community and materials discovery.