Aarthi Koripelly
Project: Extracting Scientific Information from Free Text Articles
Mentor: Kyle Chard, Globus Labs
Research Area Keywords: Machine Learning & AI // Systems // Medicine & Health
Project Description: Aarthi Koripelly is an incoming freshman at the University of Chicago studying computer science and statistics, a 2020 Coca Cola Scholar, and a previous intern in the 2019 program. This summer, Aarthi worked with the Globus Labs research group with Dr. Kyle Chard and Zhi Hong, on a project that explored scalable approaches for automatically extracting relations from scientific papers (e.g., melting point of a polymer). The project implemented a dependency parser-based relation extraction model to understand relationships without the need for a Named Entity tagger and integrated several word embeddings models and custom tokenization to boost learning performance for scientific text.