A Childhood Disease Database Based on Word Segmentation Technology: Research and Practice

Most Chinese EMRs, text mining over which enable compilation of specific disease databases, are stored as unstructured text, natural language in the field of medicine is distinct and differs from general Chinese. word segmentation of EMRs, pattern matching and feature mining methods were used to find the quantity of disease characteristics in case histories. a full-scale, comprehensive medical database was compiled. After construction of a semantic analysis model based on clinical natural language processing, this study perform pattern matching and disease feature mining and complete mining and analysis of diseases, realizing a semantic parsing and analysis system for diseases.