| In recent years,with the rapid development of related technologies such as artificial intelligence and machine learning,the court office system is being upgraded in the direction of intelligence.In order to comply with the trend of national science and technology upgrading,the Tianjin High People’s Court is increasing the research and development of intelligent systems to improve work efficiency.At the same time,electronic file management is receiving more and more attention.The realization of the cataloging task of its intelligent cataloging system requires labeling data,but the lack of corresponding data sets requires the construction of a system for labeling.Based on this background,this thesis builds an online electronic file labeling system for collecting and managing labeling data.In this thesis,based on full-text retrieval technology and image content recognition technology,a combination of automatic annotation and manual annotation is used to establish an annotation data set.Scan the electronic file pictures through OCR image recognition technology to obtain the picture content,and use the Elasticsearch search engine to match the relevant label keywords to complete the automatic labeling.Regarding the case of matching failure,supplement the annotation through manual annotation to ensure the accuracy of the image annotation.This thesis divides the electronic file labeling system into seven functional modules,namely file file acquisition,file data preprocessing,electronic file labeling,electronic file task management,data statistics,system management,and data export.The system adopts Spring and Spring Boot as the back-end framework of the system;uses My Batis as the data persistence layer framework of the system;the electronic file data uses the FTP file transfer protocol for file transmission;combined with the demand analysis and module design of the electronic file labeling system,in turn to the above Function modules are implemented and tested.At present,the electronic file labeling system has been put into use in the Tianjin Higher People’s Court.The feedback system runs smoothly after use by the staff,and the operation is simple and convenient. |