An automated materials and processes identification tool for material informatics using deep learning approach
M. Saef Ullah Miah,
Junaida Sulaiman,
Talha Bin Sarwar,
Nur Ibrahim,
Md Masuduzzaman,
Rajan Jose
Affiliations
M. Saef Ullah Miah
Faculty of Computing, College of Computing and Applied Sciences, Universiti Malaysia Pahang, Pekan, Pahang, 26600, Malaysia; Department of Computer Science, FST, American International University-Bangladesh (AIUB), 1229, Dhaka, Bangladesh
Junaida Sulaiman
Faculty of Computing, College of Computing and Applied Sciences, Universiti Malaysia Pahang, Pekan, Pahang, 26600, Malaysia; Center for Data Science and Artificial Intelligence (Data Science Center), Universiti Malaysia Pahang, 26600, Pekan, Malaysia
Talha Bin Sarwar
Faculty of Computing, College of Computing and Applied Sciences, Universiti Malaysia Pahang, Pekan, Pahang, 26600, Malaysia
Nur Ibrahim
Faculty of Electrical Engineering, Telkom University, Bandung, West Java, 40257, Indonesia; Corresponding author at: Faculty of Electrical Engineering, Telkom University, Bandung, West Java, 40257, Indonesia.
Md Masuduzzaman
Kumoh National Institute of Technology, Gumi, 39076, Republic of Korea
Rajan Jose
Faculty of Industrial Sciences & Technology, Universiti Malaysia Pahang, 26300, Gambang, Malaysia; Center of Advanced Intelligent Materials, Universiti Malaysia Pahang, 26300, Kuantan, Malaysia
This article reports a tool that enables Materials Informatics, termed as MatRec, via a deep learning approach. The tool captures data, makes appropriate domain suggestions, extracts various entities such as materials and processes, and helps to establish entity-value relationships. This tool uses keyword extraction, a document similarity index to suggest relevant documents, and a deep learning approach employing Bi-LSTM for entity extraction. For example, materials and processes for electrical charge storage under an electric double layer capacitor (EDLC) mechanism are demonstrated herewith. A knowledge graph approach finds and visualizes different latent knowledge sets from the processed information. The MatRec received an F1 score of 9̃6% for entity extraction, 8̃3% for material-value relationship extraction, and 8̃7% for process-value relationship extraction, respectively. The proposed MatRec could be extended to solve material selection issues for various applications and could be an excellent tool for academia and industry.
