Huan He, Sunyang Fu, Liwei Wang, Andrew Wen, Sijia Liu, Sungrim Moon, Kurt Miller, Hongfang Liu
Abstract: With the rapid development of big data and deep learning techniques in recent years, natural language processing (NLP) systems have been widely used in the healthcare domain. Due to the low error tolerance for clinical use cases, such as clinical decision support and precision medicine, performance evaluation is a crucial task in developing and refining an NLP system. However, evaluating an NLP system for a specific clinical use case is challenging due to the complexity of the gold standard corpus creation and the evaluation process. In response, we proposed using web-based data visualization techniques to promote the whole process and developed several visual analysis modules based on a text annotation tool.