Entity Recognition Method for Key Information of Police Records Based on Bert-Bilstm-Selfatt-Crf

At present, the public security department has a large number of case records, which contain information such as the alarm people, the time of the crime, the place of the crime, the modus operandi, the carrier involved, and the amount of money involved. Maximizing the use of this information is one of the key factors for the police to quickly solve the case and prevent crime. Using information extraction[1] technology to automatically classify police information and extract entities from case records can not only improve the efficiency of information analysis and enable the police to clarify the context of the case more quickly, but also have more significance for the further use of these information.