Abstract：Objective To transform the unstructured address information of patients in electronic medical record into structured
address information, and supplement the missing address element in the address. Methods A standard address library for storing
standard address data sets and a custom address matching rule library were built in this paper. Based on the standard address library,
the address was segmented by a forward adaptive matching algorithm based on address elements. Then the address elements obtained
by word segmentation were looked up from back to front according to the custom address matching rule base constructed to obtain
the complete address. Results The automatic word segmentation of address data in medical records was realized, and the missing
address elements in address data was complemented to complete the work of address standardization. Conclusion This study not
only greatly facilitates the automatic acquisition of address information on the first page of clinical medical records, but also facilitates the
data reporting and statistical analysis of various institutions. It can greatly reduce the workload of manual data processing and lay a
solid foundation for subsequent extraction and standardization of other information.
李净，朱贵鲜，周亮，郑西川. 基于标志词的正向自适应长度匹配的地址分词算法与缺失地址要素补充方法[J]. 中国医疗设备, 2019, 34(4): 112-114.
LI Jing, ZHU Guixian, ZHOU Liang, ZHENG Xichuan. Address Segmentation Algorithm Based on Forward Adaptive Length Matching
by Mark Words and Supplementary Method of Missing Address Elements. China Medical Devices, 2019, 34(4): 112-114.
