Dr. Catherine Shi has been awarded funding by the Strategic Research Initiative (SRI) of the College of Arts and Sciences for her project titled "Developing Automatic Pipelines and Algorithms to Enable Rapid Learning of COVID-19 Publications". A description of the project follows.
During the COVID-19 pandemic, publications have been proliferating. Toward efficient learning of massive publications, we plan to develop automatic pipelines and algorithms to enable the rapid extraction of key knowledge and entities from COVID-related publications. Aside from the pipeline development based on Natural Language Processing (NLP) technologies, we intend to bridge the gap between cutting-edge statistical methods and existing NLP algorithms, and develop new NLP algorithms to improve the accuracy, generalizability, and scalability at a population level.
This entire project is a direct response to the COVID-19 pandemic. However, the outcomes will be valuable not only for the COVID-19 research but, more importantly, for all similar knowledge extraction needs in the future, while the number of publications and other free-text materials is increasing rapidly in the Big Data era.