Usage of machine learning and scraping to create a new dataset of refractive index of oxide glasses.

• Grant number: 25/02033-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: June 01, 2025
• End date: May 31, 2026
• Field of knowledge: Engineering - Materials and Metallurgical Engineering - Nonmetallic Materials
• Principal Investigator: Thiago Rodrigo Ramos
• Grantee: Thomaz Yeiden Busnardo Aguena
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Associated research grant: 13/07793-6 - CEPIV - Center for Teaching, Research and Innovation in Glass, AP.CEPID

Abstract

This project aims to: 1) Develop a computer code capable of collecting the chemical compositions of oxide glasses and their Refractive Indexes (RI) as reported in patents. It is worth noting that patents often lack comprehensive data about the patented glass, as companies typically measure only the properties relevant to their inventions. Consequently, the program must be fully adapted to extract additional information beyond the Refractive Index. Simultaneously, other students will be working on two additional projects, developing separate codes specifically designed to collect data on viscosity and Tliquidus, two crucial properties of oxide glasses. Eric Trevelato Costa will focus on viscosity, while Gustavo Laranja Thomaello will focus on Tliquidus. 2) Create a new dataframe with the collected data to enhance the accuracy of machine learning models when combined with existing datasets such as SciGlass. 3) Utilize machine learning that leverages data-driven predictive models to predict oxide glass compositions and their corresponding Refractive Index.