Purpose/Significance To establish the off-label evaluation system of antibacterial drugs and to improve the management level of antibacterial drugs for children. Method/Process Based on natural language processing (NLP) technology, post-structured processing of unstructured data such as electronic medical record (EMR) is carried out. Based on the deep learning algorithm model and medication rule knowledge base, the medication characteristics data of children are analyzed, and an evaluation model for off-label use of antibacterial drugs is constructed. Additionally, the off-label evaluation system of antibacterial drugs is designed and implemented, providing physicians with accurate intelligent prompts for reviewing antimicrobial prescriptions. Result/Conclusion This system evaluates the antibacterial drugs use of 37 372 children, provides effective references for clinicians to make decisions antibacterial drugs, and meets the requirements of hospitals for off-label administration of children’s antibacterial drugs.