Data and Text Analytics on Material Data – Machine Learning and Material Science – Doctoral Research Topic

About the Project

This research introduces a Machine Learning technique to automatically extract sections from research documents to classify high-temperature materials and propose a new citation network. A deep learning method categorizes SEM images based on the crystal structure (crystalline, amorphous, cubic, and hexagon). The methods and characterization from the “Materials and Methods” section, people and organizations acknowledged from the “Acknowledgement” section were extracted from “Journal of Material Science” and website that revealed important insight. The extracted information extends the existing Citation Network to predict new links between authors & funding, material & methods, and authors & organizations. The proposed system used LDA to extract keywords and LSTM to summarize the text. As application, the computational simulation was performed to understand molecular dynamics and phase transformation on materials with high-temperature material. Shock tube experiments and numerical simulations were performed to understand the TPS material’s phase transformation and surface chemical reaction.

Department & Campus

Department of Computer Science Engineering, School of Engineering, Bengaluru

Skillsets Preferred from Applicants

Shock Tubes, XRD, SEM, QuantumATK – Material Simulation Software