Named Entity Recognition of Electronic Medical Record Based on Improved HMM Algorithm

Electronic medical record is important clinical diagnosis information generated in the curative activity which contains a large amount of medical data related with health condition of patients. Recognition of medical named entity is based on the data mining on electronic medical record. This paper takes ophthalmic electronic medical record as research object. At first, training corpus is annotated under the guidance of specialist; and then HMM algorithm is improved to be applied in named entity recognition, and annotated training corpus is used to train HMM model; after that, trained HMM model is used in test set for entity recognition. At last, experiment is conducted to make a contrast between the algorithm proposed in this paper and the algorithm based on word segmentation model. The experimental findings show that the algorithm proposed in this paper achieves good results in the named entity recognition of electronic medical record.