This code repository is used for Open Enzymes for the medical natrual language process tool in Chinese. Especially codes in cardiology. Also could be reused in other domain.
Recognize the hematoma information from free-text clinical reports, on scope-level and patient-level.
The inputs are stored in test.pos.txt and test.neg.txt. Each line for one input. The prediction results are list in test_reult.txt in following format: TYPE-LineNumber Predict.
Running eval.py will give out the system performance.