Affiliations: [a] Information Resources Research Center, Wuhan University, Wuhan, China
| [b] School of Information Management, Wuhan University, Wuhan, China
Abstract: Medical literature research results are more accurate and representative than patient medical record data. The medical literature intelligently extracts research objects, research areas, and research result entities, identifies their indexing features in the abstract, and finds risk factors for chronic illness and its association with regions and populations. This study proposes a literature knowledge extraction model for chronic disease risk factors. Based on dictionaries and rules, manual annotation extraction methods and gCLUTO dual cluster analysis, the Chinese biomedical literature database was published to correlate with chronic disease risk factors. The literature is a corpus, which intelligently identifies extracts and clusters the literature abstracts. The discovery of chronic disease risk factors from the perspective of human health was explored by taking the literature of hypertension risk factor research as an example; the literature knowledge extraction model for chronic disease risk factors was verified to construct a chronic disease risk factor set. It also reveals the relationship between chronic disease risk factors and regions/populations, and provides reference and reference for the research of chronic disease risk factors.
