Research On User Profile Based On Mapping Knowledge Domain Of Public Security

The Mapping Knowledge Domain,also known as semantic web knowledge bases,are a description of knowledge and can better describe the relationships between entities.At present,the application of big data mining technology in the field of public security has not been fully popularized,and more basic platforms of big data are established.In this paper,the public security public transportation domain Mapping Knowledge Domain of public security is established,involving data mainly including data of subway,data of bus,data of bus GPS,data of electronic identity information.The purpose of establishing Mapping Knowledge Domain of public security is to query the relationship between entity attributes and entities quickly and provide the fact label for user portrait.Based on the study of user portrait based on knowledge graph,this paper proposes the establishment of user portrait based on knowledge graph and its application in police intelligence work,which mainly includes the following aspects:1)This paper firstly establishes the Mapping Knowledge Domain of public security of public transportation police,including knowledge fusion,knowledge extraction and knowledge storage.The data used in this paper are mainly public transportation data in the field of public security,electronic identity information,etc.Due to different primary keys,multi-source data cannot be fused through a certain key word.Knowledge extraction mostly comes from relational database,and there is no semantic analysis.Knowledge storage USES the graph database Neo4 j.The establishment of Mapping Knowledge Domain of public security is to provide information input for the establishment of user portrait label.2)The establishment of user portrait requires two parts of data,one is the public security Mapping Knowledge Domain of public security data,and the other is the cleaned public transportation data.Rely on Hadoop big data platform to clean public transport data.To clean the public transportation data,the most important thing is to predict the bus exit station.Since the bus only swipes the card when getting on the bus,and the user portrait focuses on the user’s personal travel habits,it is necessary to predict the bus exit track.In this paper,three methods,namely travel chain,travel habit and site attractiveness,are used in tandem to predict bus drop-off stations.3)User portrait tag is composed of fact tag and model tag.The user fact label was obtained by Mapping Knowledge Domain of public security,and the user behavior data was selected as input and output model label,including occupation and residence label,accompanying travel label,abnormal identity label and travel purpose label.The first three categories of tags adopt rules for statistical calculation.The travel purpose tag USES k-means clustering algorithm to cluster the sites through POI data and infer the travel purpose according to the site attributes.4)The usability of the user portrait based on knowledge graph is tested by the application of user portrait in the public transportation integral model.The integral model in this paper is designed for the monitoring and control of public transportation personnel.The integral model includes personnel category,integral term,integral value,control score,disposal score,etc.The integral term is composed of basic score and dynamic grouping.Through the integral model early warning personnel has the crime suspect.In this paper,the establishment of user portraits from multi-source data plays an auxiliary role in decision-making.