Thanks to partial funding from the University of Oklahoma’s Data Institute for Societal Challenges (DISC), a paper from Kun Lu, Ph.D., associate professor in the School of Library and Information Studies, Dodge Family College of Arts and Sciences, and doctoral student Satanu Ghosh has been published in the Aslib Journal of Information Management.
According to the U.S. Energy Information Administration, global energy consumption is expected to increase nearly 50% by 2050 due to economic and population growth. Renewable and clean energy will play a large role in meeting the demand.
Photovoltaic (PV) material that converts light into electricity will allow scientists to harness renewable and clean solar energy. For PV material to accomplish this, photons in the light need to carry enough energy to excite electrons into a free state to create electric current. Band gap is the minimum amount of energy required to excite an electron in a material into such a free state. Knowing the band gap information is very important for material scientists to determine candidate materials for PV cells.
“Band gap information has been widely reported in scientific literature, but the volume of the literature prevents scientists from gaining a complete view of the band gaps of various materials,” Lu said. “Manually collecting this information is inefficient and unable to keep up with the ever-increasing volume. As a result, most scientific decisions are made based on partial information, which can lead to missed opportunities for discovering novel solar materials. Automated methods to extract band gap information from literature will overcome this barrier.”
The paper, “Band gap information extraction from materials science literature – a pilot study,” presents preliminary work on extracting information related to materials and their band gap from nearly 12,000 academic papers out of an original collection of 1.44 million titles.
Future work will focus on improving the performance of Lu’s model and addressing errors in information extraction. "I hope to collaborate with a domain scientist to pursue further funding to improve the program we have developed,” Lu said.
