Sun’s NSF CAREER award will improve question answering certainty
Huan Sun has received a five-year, $499,965 Faculty Early Career Development (CAREER) award from the National Science Foundation (NSF) for her research on interactive and transparent natural language question answering models. She is an assistant professor at Ohio State in the Department of Computer Science and Engineering.
The CAREER award is the National Science Foundation’s (NSF) most prestigious award in support of junior faculty who exemplify the role of teacher-scholars through outstanding research, excellent education and the integration of both.
The overall goal of Sun’s project, “Towards Interactive and Transparent Question Answering with Applications in the Clinical Domain” is to develop a new intelligent question answering (QA) model that can interact with users to resolve ambiguity and uncertainty during the answering process.
“Finding relevant information quickly is integral to effective and efficient decision making, but this becomes increasingly difficult as the scale and heterogeneity of data continue to grow rapidly,” explained Sun.
QA systems, which aim to find precise answers to natural language questions from users, have shown great potential to address this problem. However, state-of-the-art QA systems still largely fall short in various scenarios: when questions are ambiguous or complex; when answering questions requires background knowledge not readily available in the data; and when users need to understand the system’s answering process in order to judge trustworthiness.
“These scenarios are prevalent in real application domains of QA, including healthcare, finance and sciences, and must be addressed when building practical systems,” said Sun.
Her project will develop a novel interactive QA model, which will detect the ambiguities and uncertainties during the answering process and interact with users in a natural fashion to seek clarifications. Additionally, the QA model will learn from such interactions to simultaneously improve answer quality and reduce human intervention over time, using imitation and reinforcement learning-based frameworks. Sun’s project further aims to improve the QA model’s transparency by decomposing a complex question into several intermediate sub-questions and allowing users to validate them.
“We expect our results will contribute to future human-technology partnership by enabling QA models to be more interactive, more transparent and therefore more trustworthy,” she said.
The proposed QA model will be tested in the clinical domain, where doctors often ask questions about a patient and look for answers in electronic medical records (EMRs). Sun’s model may enable doctors to effectively and efficiently query EMRs and gather relevant evidence for critical decision making.
The project also includes plans for integrating research and education. The research team will work closely with healthcare providers for model evaluation and actively seek technology transfer opportunities. All datasets, software and demos will be publicly accessible. Research findings will be disseminated in computer science and medical informatics-related venues, and will be integrated into existing and new courses.
Sun’s research team will engage high school students and undergraduates, particularly those from underrepresented groups, and prepare them for future education and employment opportunities.
Her previous QA research, which utilizes human-machine collaboration to improve question answering, has been funded by the Army Research Office (ARO), an element of the U.S. Army Combat Capabilities Development Command’s Army Research Laboratory.
Sun joined The Ohio State University in 2016. Prior to that, she was a visiting scientist at the University of Washington. She earned her PhD in computer science from the University of California, Santa Barbara, and her bachelor’s in electronic engineering and information science from the University of Science and Technology of China.
by Meggie Biss, College of Engineering Communications | firstname.lastname@example.org