May 14, 2020 By Victoria Grdina
The Department of Computer Science and Engineering has added a new summer course in which students can study how natural language processing could be used to combat COVID-19.
Associate professor Stephen Scott will be teaching CSCE 496/896, Section 200: Combating COVID-19 with Natural Language Processing. The online course will introduce fundamental concepts and techniques in natural language processing (NLP) and assign exercises to students to implement NLP solutions using the machine crane at the Holland Computing Center.
“Concepts taught in this course will build a foundation for the NLP side,” Scott said. “We'll use machine learning techniques such as naive Bayes and deep learning with recurrent artificial neural networks.”
While the idea of natural language processing (NLP) has been around for decades, the past several years have witnessed significant advances in numerous applications such as information extraction, question answering, sentiment analysis, machine translation, and many more. NLP has a significant opportunity to help combat the COVID-19 pandemic, particularly in the Kaggle competition COVID-19 Open Research Dataset Challenge (CORD-19).
Course projects will focus on addressing the requirements of the CORD-19 competition: analyzing a dataset of over 59,000 scholarly articles and answering questions related to COVID-19 spread, risk factors, vaccines, genetics, and more.
“The competition's goal is to automatically analyze that literature to answer fundamental questions,” Scott said. “Mining answers of such questions out of the literature might help researchers and clinicians better combat the disease.”
Direct submissions to the competition itself by students will be welcome, but not required, since the competition's round two deadline is a month before the course ends.
Students interested in enrolling the course (which begins May 18 during the 8-week session) can still register online and contact Stephen Scott (email@example.com) with any questions.