WO2021114615A1 - 行为风险识别的可视化方法、装置、设备及存储介质 - Google Patents

行为风险识别的可视化方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2021114615A1
WO2021114615A1 PCT/CN2020/098880 CN2020098880W WO2021114615A1 WO 2021114615 A1 WO2021114615 A1 WO 2021114615A1 CN 2020098880 W CN2020098880 W CN 2020098880W WO 2021114615 A1 WO2021114615 A1 WO 2021114615A1
Authority
WO
WIPO (PCT)
Prior art keywords
behavior
risk
data
target object
preset
Prior art date
Application number
PCT/CN2020/098880
Other languages
English (en)
French (fr)
Inventor
赵喆
曾永理
刘言曌
Original Assignee
平安科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2021114615A1 publication Critical patent/WO2021114615A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0635Risk analysis of enterprise or organisation activities
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24Querying
    • G06F16/245Query processing
    • G06F16/2458Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2465Query processing support for facilitating data mining operations in structured databases
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/23Clustering techniques
    • G06F18/232Non-hierarchical techniques
    • G06F18/2321Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions

Definitions

  • This application relates to the data visualization field of big data technology, and in particular to a visualization method, device, equipment, and storage medium for behavioral risk identification.
  • the common method of identifying personnel behavioral risks in the industry is usually to determine the risk attribute entries for employee behavioral risk early warning, collect target employees’ data, and The attribute characteristics of the target employee are input into the employee behavior risk early warning model and the risk probability score of the target employee is obtained.
  • the inventor realizes that the existing risk identification technology generally mines a single feature of behavioral data and corresponds to the same type of behavioral risk scenarios. For different types of behavioral risk scenarios, a new risk identification process needs to be re-developed, resulting in multi-dimensional risk identification. The efficiency is low. In addition, risk identification results usually aim to obtain a certain risk scoring index. The scoring results cannot cover all the meanings of risks, nor can it be traced back to the history of personnel behavior, which leads to the problem of whether the risk identification results are reliable.
  • the main purpose of this application is to solve the problem that the existing risk identification technology is low in the recognition of multi-dimensional behavior characteristics, and it is difficult to trace back the behavior history of personnel, which makes it impossible to determine whether the risk determination result is reliable.
  • the first aspect of the present application provides a visualization method for behavior risk identification, including: collecting first behavior data and identification data of a target object from a terminal with a preset burying mechanism, wherein the identification data Including a buried point identifier and a target object identifier; performing data mining on the first behavior data of the target object according to the buried point identifier, the target object identifier, and preset feature types to obtain the multi-dimensional behavior characteristics of the target object;
  • the preset risk recognition engine performs risk assessment on the multi-dimensional behavior characteristics of the target object, and obtains multiple assessment results; when it is detected that at least one of the assessment results is at risk, the second behavior data of the target object is obtained, and the second behavior data of the target object is obtained through the preset two-dimensional behavior.
  • the map application processes the first behavior data of the target object and the second behavior data of the target object to generate and display a target visualization interface, wherein the second behavior data is generated before the first behavior data is generated
  • the target visualization interface is used to display the risk behavior represented by the first behavior data and/or the second behavior data in a two-dimensional map.
  • a second aspect of the present application provides a visualization device for behavioral risk identification, including: a collection module for collecting first behavioral data and identification data of a target object from a terminal with a preset burying mechanism, wherein the identification data Including buried point identification and target object identification; a mining module for data mining on the first behavior data of the target object according to the buried point identification, the target object identification and preset feature types, to obtain the number of target objects Dimensional behavior characteristics; an evaluation module for risk assessment of the target object’s multi-dimensional behavioral characteristics through a preset risk recognition engine to obtain multiple evaluation results; a display module, when it is detected that at least one evaluation result is at risk, use
  • the first behavior data of the target object and the second behavior data of the target object are processed through a preset two-dimensional map application to generate and display the target visualization interface.
  • the second behavior data is generated before the first behavior data is generated, and the target visualization interface is used to display the first behavior data and/or the risk behavior represented by the second behavior data in
  • a third aspect of the present application provides a visualization device for behavioral risk identification.
  • the visualization device for behavioral risk identification includes a memory and a processor, and the processor and the memory are connected to each other, wherein the memory is used for storing
  • a computer program the computer program includes program instructions, and the processor is configured to execute the program instructions of the memory, wherein: first behavior data and identification data of the target object are collected from a terminal with a preset embedding mechanism, wherein The identification data includes a buried spot identification and a target object identification; data mining is performed on the first behavior data of the target object according to the buried spot identification, the target object identification and the preset feature type, and the number of target objects is obtained.
  • the fourth aspect of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a processor, they are used to implement the following Step: Collect the first behavior data and identification data of the target object from the terminal of the preset burying mechanism, where the identification data includes the burying point identifier and the target object identifier; according to the burying point identifier, the target object identifier Data mining is performed on the first behavior data of the target object with the preset feature types to obtain the multi-dimensional behavior characteristics of the target object; the risk assessment is performed on the multi-dimensional behavior characteristics of the target object through the preset risk recognition engine, and the multi-dimensional behavior characteristics of the target object are obtained.
  • An evaluation result when it is detected that at least one evaluation result is at risk, the second behavior data of the target object is acquired, and the first behavior data of the target object and the second behavior data of the target object are applied through a preset two-dimensional map
  • the data is processed to generate and display a target visualization interface, wherein the second behavior data is generated before the first behavior data is generated, and the target visualization interface is used to display the first behavior data and the first behavior data in a two-dimensional map. /Or the risk behavior represented by the second behavior data.
  • the first behavior data and identification data of the target object are collected from a terminal with a preset burying mechanism, wherein the identification data includes a burying point identifier and a target object identifier; according to the burying point identifier ,
  • the target object identifier and the preset feature type perform data mining on the first behavior data of the target object to obtain the multi-dimensional behavior characteristics of the target object;
  • the multi-dimensional behavior characteristics of the target object are determined by the preset risk recognition engine Perform risk assessment to obtain multiple assessment results; when it is detected that at least one assessment result is at risk, obtain the second behavior data of the target object, and apply the first behavior data and the first behavior data of the target object through a preset two-dimensional map.
  • the second behavior data of the target object is processed to generate and display a target visualization interface, wherein the second behavior data is generated before the first behavior data is generated, and the target visualization interface is used to display all objects in a two-dimensional map.
  • the risk identification engine is used to decouple the risk assessment model from the risk scenario, and the multi-dimensional behavior characteristics of the target object are identified and evaluated, which not only makes the risk identification rules scalable, but also improves the efficiency of risk identification
  • the visualization application based on the two-dimensional map intuitively displays the risk behavior of the target object, which improves the intuitiveness and speed of behavior risk identification.
  • FIG. 1 is a schematic diagram of an embodiment of a visualization method for behavior risk identification in an embodiment of this application.
  • FIG. 2 is a schematic diagram of another embodiment of the visualization method for behavior risk identification in an embodiment of this application.
  • FIG. 3 is a schematic diagram of an embodiment of a visualization device for behavior risk identification in an embodiment of the application.
  • FIG. 4 is a schematic diagram of another embodiment of a visualization device for behavior risk identification in an embodiment of this application.
  • Fig. 5 is a schematic diagram of an embodiment of a visualization device for behavior risk identification in an embodiment of the application.
  • the embodiments of the present application provide a visualization method, device, equipment, and storage medium for behavioral risk identification, which are used to decouple the risk assessment model from the risk scenario through a risk identification engine to identify and evaluate the multi-dimensional behavioral characteristics of the target object , Improve the efficiency of risk identification.
  • An embodiment of the visualization method for behavioral risk identification in the embodiment of the present application includes.
  • the preset burying mechanism is a data collection mechanism set in advance to satisfy fast, efficient, and rich data applications.
  • the preset burying mechanism is used to collect and record the process and results of the target object’s behavior on the terminal, and integrate the target The process and result of the object's behavior are reported to the server.
  • the terminals with the preset burying mechanism include mobile phones, on-vehicle automatic diagnostic devices and attendance equipment.
  • the server collects the first behavior data and identification data of the target object through the mobile phones, on-vehicle automatic diagnostic devices and attendance equipment, where the identification data includes embedded Point identifier and target object identifier, for example, the buried point identifier pay_fail is used to identify the buried point event of the order payment failure.
  • the target object identifier can be user_A.
  • the buried point identifier and target object identifier can be in uppercase and lowercase English, numbers, and underscores, but cannot Start with a number.
  • the first behavior data of the target object includes target positioning system data and related business data, and the target positioning system data is global positioning system GPS data.
  • the server stores the first behavior data of the target object, the embedded point identifier, and the target object identifier in a preset database.
  • the execution subject of the present application may be a visualization device for behavior risk identification, and may also be a terminal or a server, which is not specifically limited here.
  • the embodiment of the present application takes the server as the execution subject as an example for description.
  • the server stores the first behavior data of the target object in the distributed database hbase according to the embedded point identifier and the target object identifier, and uses the massive data provided by the distributed system infrastructure hadoop to process short hair and database space computing capabilities.
  • the built-in feature type real-time mining of the multi-dimensional behavior characteristics of the target object.
  • the multi-dimensional behavior characteristics of the target object include the target positioning system density feature, trajectory composite feature, detention time feature, detention location feature, and preset business feature.
  • preset The business characteristics include the characteristics of over-dispatch area and lateness.
  • the server stores the density characteristics, trajectory composite characteristics, detention time characteristics, and detention location characteristics in the object-relational database postgre, and stores them in a binary manner.
  • the preset business features are generally stored in the preset field table of the object-relational database mongdb or the preset field table of the relational database.
  • the preset risk recognition engine uses at least one risk recognition application interface to perform risk recognition and evaluation processing on the multi-dimensional behavior characteristics of the target object.
  • the server recognizes the multi-dimensional behavior characteristics of the target object through the preset risk recognition engine, and obtains multiple standardized interfaces for risk judgment rules, and sequentially processes the multi-dimensional behavior characteristics of the target object based on the multiple risk judgment rule standardized interfaces to obtain Multiple evaluation results, for example, three evaluation results A, B, and C.
  • the server uses the target object identifier to associate the mapping relationship between the multi-dimensional behavior characteristics of the target object and the multiple evaluation results.
  • the server can directly call the preset risk identification engine for identification processing after the terminal uploads the first behavior data of the target object, or it can call the preset risk identification engine for identification processing regularly according to the preset duration.
  • the duration can be 1 minute or 5 minutes, which is not limited here.
  • the server can read the second behavior data of the target object from the preset database according to the target object identifier, for example, three evaluation results A, B, and C, evaluation result A and evaluation result C has no risk, but the assessment result A is risky, the server obtains the target positioning system data.
  • the server encapsulates the first behavior data of the target object and the second behavior data of the target object, and renders the encapsulated data through the preset two-dimensional map application corresponding to the preset two-dimensional map interface to obtain the target visualization interface , And show the target visualization interface.
  • the risk behavior represented by the first behavior data and/or the second behavior data of the target object is visually displayed based on the target visualization interface, so that the risk behavior represented by the first behavior data and/or the second behavior data of the target object can be displayed.
  • Backtracking and visualization, and feedback to the preset risk identification engine through the preset confirmation mechanism optimize the preset risk identification engine, and quickly and efficiently improve the risk identification rate.
  • the risk identification engine is used to decouple the risk assessment model from the risk scenario, and the multi-dimensional behavior characteristics of the target object are identified and evaluated, which not only makes the risk identification rules scalable, but also improves the efficiency of risk identification
  • the visualization application based on the two-dimensional map intuitively displays the risk behavior of the target object, which improves the intuitiveness and speed of behavior risk identification. This solution can be applied in the field of smart security, thereby promoting the construction of smart cities.
  • FIG. 2 another embodiment of the visualization method for behavior risk identification in the embodiment of the present application includes.
  • 201 Collect first behavior data and identification data of a target object from a terminal with a preset burial mechanism, where the identification data includes a burial spot identification and a target object identification.
  • the server receives the burial data sent by the terminal with the preset burial mechanism, where the burial data is data collected based on the preset burial event.
  • the burial data is data collected based on the preset burial event.
  • the burying point data of the operation case what pictures are uploaded at what time, how to adjust the payment amount when the case is closed, account login records, and survey photo shooting data
  • terminals with preset burying mechanism include mobile phones, on-board automatic diagnosis devices and Attendance equipment
  • the server sets a burying point for the terminal to obtain a terminal with a preset burying point mechanism.
  • the server regularly receives the burial point data sent by the terminal with the preset burial point mechanism, for example, collects target positioning system data or attendance data every 10 seconds or 15 seconds, and the attendance data includes the check-in time, the check-in location, and the check-in object.
  • the server performs data analysis on the buried point data according to the first preset data format to obtain the first behavior data and identification data of the target object.
  • the identification data includes the buried point identifier and the target object identifier.
  • the first preset data format is JS object notation data exchange format (javaScriptobject notation, JSON), the fields in the buried point data include the buried point identifier buryId, the target object identifier userId, the latitude and longitude upload time gpsTime, the latitude and longitude gps and the moving speed speed, and the latitude and longitude can be separated by commas, for example , Identification data ⁇ "ids_data”:["buryId”:"23”,”userId”:"USER_A” ⁇ , the first line data of the target object ⁇ "gpsTime”:"2020-1-12 23:54:11",”speed”:"21.1”,”gps":"112.3232,28.322423” ⁇ .
  • the server associates the embedded point identifier in the embedded point data with the identifier of the preset business scenario for associated query, Obtain the preset business scenario.
  • the server encapsulates the target behavior data based on the preset business scenario and saves it in the preset database.
  • the server determines whether the first behavior data of the target object belongs to the target positioning system data based on the buried point identifier. If the first behavior data of the target object does not belong to the target positioning system data, the server performs business feature mining on the first behavior data of the target object according to the preset feature type to obtain the first behavior feature, where the first behavior feature includes whether it exceeds
  • the characteristics of the dispatched area, the characteristics of whether they are late, and whether the characteristics of the case are inquired during driving are selected and determined according to the preset characteristic types, which are not limited here.
  • the server performs feature extraction on the first behavior data of the target object, and stores the extracted features in a preset database to obtain the second behavior feature, the second behavior Features include target positioning system density features, trajectory composite features, detention location features, and detention time characteristics.
  • the server can use the density-based noise application spatial clustering DBSCAN algorithm to perform clustering calculations on the first behavior data of the target object. Then, the number of target positioning systems in each cluster is taken as the cluster density to obtain the density characteristics of the target positioning system.
  • the distance between points is the distance from the surface of the earth, and the clustering critical value and the number of cluster points are based on actual business requirements. Setting, the default is 20 points within 50 meters.
  • the server can conveniently traverse each trajectory to find a trajectory whose speed is less than 5km/h and the distance between the start and end points of the target positioning system within 5 minutes is less than 500 meters.
  • Two points are used as the staying point, and the trajectory duration is used as the staying time.
  • the staying feature is ⁇ behaviorFeature: ⁇ "stayGps":"112.3232,28.322423","stayDuration":"23" ⁇ , where stayGps is used to indicate the characteristics of the staying place And stayDuration are used to indicate the characteristics of the length of stay.
  • the server sorts the target positioning system data in the first behavior data of the target object in order of time to obtain the trajectory synthesis feature.
  • the server stores the trajectory composite feature into the database, where the preset duration may be 10 minutes. Finally, the server combines the first behavior feature and the second behavior feature, and maps the combined feature data with the target object identifier to obtain the multi-dimensional behavior feature of the target object.
  • a preset risk recognition engine uses a preset risk recognition engine to match and recognize the multi-dimensional behavior characteristics of the target object to obtain a risk rule data set, where the risk rule data set includes multiple risk rules.
  • the multiple risk rules are pre-configured in the preset risk identification engine according to the business, and further, the server obtains the preset business types corresponding to the multi-dimensional behavior characteristics of the target object.
  • the server queries the preset rule data table according to the respective preset business types to obtain a risk rule data set.
  • the risk rule data set includes multiple risk rules, for example, a risk rule for whether to arrive late and a risk rule for whether to shift shifts on time.
  • the server respectively calls multiple preset risk assessment models according to multiple risk rules, so that the multiple preset risk assessment models perform risk assessment on the multi-dimensional behavior characteristics of the target object, and obtain multiple risk scores. Further, the server determines multiple risk levels according to the risk scores, and sets the multiple risk scores or multiple risk levels as multiple evaluation results.
  • the server transmits the target object identifier userId to the preset risk recognition engine, and the preset risk recognition engine determines that the corresponding risk rule is a rule for judging whether you are late, and the server passes the corresponding preset risk
  • the evaluation model and the target object identifier userId look up the shift schedule of the day and the geographical boundary of the dispatch area, and obtain the punch card positioning data from the preset database according to the target object identifier userId, and then proceed according to the punch card positioning data and the geographic boundary of the dispatch area
  • the space contains calculations. Compare the clocking time with the shift schedule.
  • the evaluation result can be marked with a risk score or a risk level, for example, The risk level is 3 and the risk score is 10.
  • the server obtains behavior risk scenarios according to preset business requirements, and sets behavior risk standards for behavior risk scenarios.
  • the server obtains sample data, and trains an initial behavioral risk model based on the sample data and the established behavioral risk standards to obtain a preset behavioral risk model.
  • the server adds the preset behavior risk model to the preset risk identification engine, and configures the corresponding risk rules for the preset behavior risk model.
  • the preset behavior risk model is used to accept calls from the preset risk identification engine and output the corresponding evaluation results .
  • the server obtains the second behavior data of the target object according to the target object identifier, where the second behavior data is generated before the first behavior data is generated.
  • the server performs data encapsulation on the first behavior data of the target object and the second behavior data of the target object according to the second preset data format to obtain the encapsulated data.
  • the server obtains the encapsulated data from the first behavior data of the target object and the target object’s second behavior data.
  • the second line of data extracts multiple target positioning system data, and each target positioning system data includes latitude and longitude coordinates.
  • the server performs data encapsulation based on the latitude and longitude coordinates.
  • the second preset data format is geospatial information based on script object representation.
  • the data exchange format GeoJSON is a format for encoding various geographic data structures.
  • the server calls a preset two-dimensional map application to render and display the encapsulated data to obtain a target visualization interface.
  • the target visualization interface is used to display the first behavior data and/or second behavior data of the target object in a two-dimensional map mode Characterized risk behavior.
  • a two-dimensional map is used to display the vehicle trajectory information of the target person A when processing the auto insurance business on 2020-03-03, and the vehicle trajectory information is at risk of not conforming to the preset driving route for processing the auto insurance business.
  • the server optimizes the preset risk identification engine through the preset confirmation mechanism, and quickly improves the efficiency of risk identification.
  • the server receives the target picture sent by the terminal, and stores the target picture in a preset directory.
  • the target picture is a picture containing the target visualization interface.
  • the server reads the multi-dimensional behavior characteristics and multiple evaluation results of the target object from the preset database according to the target object identifier.
  • the server combines the multi-dimensional behavior characteristics of the target object, multiple evaluation results, and target pictures based on the preset template to obtain a risk evaluation report.
  • the server obtains the terminal identifier of the target terminal by querying the preset database according to the target object identifier.
  • the server determines a preset mode according to the terminal identifier of the target terminal, and the preset mode includes an email mode and a message push mode.
  • the server sends the risk assessment report to the target terminal according to the preset mode and terminal identification.
  • the risk identification engine is used to decouple the risk assessment model from the risk scenario, and the multi-dimensional behavior characteristics of the target object are identified and evaluated, which not only makes the risk identification rules scalable, but also improves the efficiency of risk identification
  • the visualization application based on the two-dimensional map intuitively displays the risk behavior of the target object, which improves the intuitiveness and speed of behavior risk identification.
  • the collection module 301 is configured to collect first behavior data and identification data of a target object from a terminal with a preset burial mechanism, where the identification data includes a burial spot identification and a target object identification.
  • the mining module 302 is configured to perform data mining on the first behavior data of the target object according to the buried point identifier, the target object identifier and the preset feature type to obtain the multi-dimensional behavior characteristics of the target object.
  • the evaluation module 303 is configured to perform risk evaluation on the multi-dimensional behavior characteristics of the target object through a preset risk recognition engine, and obtain multiple evaluation results.
  • the display module 304 is used to obtain the second behavior data of the target object when it is detected that at least one evaluation result is at risk, and apply the first behavior data of the target object and the second behavior data of the target object through a preset two-dimensional map Process, generate and display the target visualization interface, where the second behavior data is generated before the first behavior data is generated, and the target visualization interface is used to display the first behavior data and/or the risk represented by the second behavior data in a two-dimensional map behavior.
  • the risk identification engine is used to decouple the risk assessment model from the risk scenario, and the multi-dimensional behavior characteristics of the target object are identified and evaluated, which not only makes the risk identification rules scalable, but also improves the efficiency of risk identification
  • the visualization application based on the two-dimensional map intuitively displays the risk behavior of the target object, which improves the intuitiveness and speed of behavior risk identification. This solution can be applied in the field of smart security, thereby promoting the construction of smart cities.
  • FIG. 4 another embodiment of the visualization device for behavior risk identification in the embodiment of the present application includes.
  • the collection module 301 is configured to collect first behavior data and identification data of a target object from a terminal with a preset burial mechanism, and the identification data includes a burial spot identification and a target object identification.
  • the mining module 302 is configured to perform data mining on the first behavior data of the target object according to the buried point identifier, the target object identifier and the preset feature type to obtain the multi-dimensional behavior characteristics of the target object.
  • the evaluation module 303 is configured to perform risk evaluation on the multi-dimensional behavior characteristics of the target object through a preset risk recognition engine, and obtain multiple evaluation results.
  • the display module 304 is used to obtain the second behavior data of the target object when it is detected that at least one evaluation result is at risk, and apply the first behavior data of the target object and the second behavior data of the target object through a preset two-dimensional map Process, generate and display the target visualization interface, where the second behavior data is generated before the first behavior data is generated, and the target visualization interface is used to display the first behavior data and/or the risk represented by the second behavior data in a two-dimensional map behavior.
  • the collection module 301 may also be specifically used.
  • the terminal with the preset burying mechanism includes a mobile phone, a vehicle-mounted automatic diagnosis device, and attendance equipment.
  • the mining module 302 can also be specifically used.
  • first behavior data of the target object does not belong to the target positioning system data
  • business feature mining is performed on the first behavior data of the target object according to the preset feature type to obtain the first behavior feature.
  • the first behavior data of the target object belongs to the target positioning system data
  • feature extraction is performed on the first behavior data of the target object, and the extracted features are stored in a preset database to obtain the second behavior feature, the second behavior feature Including target positioning system density characteristics, trajectory synthesis characteristics and detention location characteristics.
  • the first behavior characteristic and the second behavior characteristic are merged, and the merged characteristic data is mapped to the target object identifier to obtain the multi-dimensional behavior characteristic of the target object.
  • the evaluation module 303 may also be specifically used.
  • the preset risk recognition engine is used to match and recognize the multi-dimensional behavior characteristics of the target object to obtain a risk rule data set, where the risk rule data set includes multiple risk rules.
  • multiple preset risk assessment models are called respectively, so that multiple preset risk assessment models perform risk assessment on the multi-dimensional behavior characteristics of the target object, and obtain multiple risk scores or multiple risk levels.
  • One risk score or multiple risk levels are set as multiple evaluation results.
  • the display module 304 can also be specifically used.
  • the second behavior data of the target object is acquired according to the target object identifier, where the second behavior data is generated before the first behavior data is generated.
  • Data encapsulation is performed on the first behavior data of the target object and the second behavior data of the target object according to the second preset data format to obtain the encapsulated data.
  • the target visualization interface is used to display the first behavior data and/or the second behavior data characterization in a two-dimensional map mode Risky behavior.
  • the visualization device for behavior risk identification further includes.
  • the setting module 305 is used to obtain behavior risk scenarios according to preset business requirements and set behavior risk standards for the behavior risk scenarios to obtain the established behavior risk standards.
  • the training module 306 is configured to obtain sample data, and train an initial behavior risk model based on the sample data and the established behavior risk standard to obtain a preset behavior risk model.
  • the configuration module 307 is used to add the preset behavior risk model to the preset risk identification engine and configure corresponding risk rules for the preset behavior risk model.
  • the preset behavior risk model is used to accept the call of the preset risk identification engine and Output the corresponding evaluation results.
  • the visualization device for behavior risk identification further includes.
  • the combination module 308 is configured to obtain a target picture, and combine the multi-dimensional behavior characteristics of the target object, multiple evaluation results, and the target picture based on a preset template to obtain a risk assessment report.
  • the target picture is a picture containing a target visualization interface.
  • the sending module 309 is configured to send the risk assessment report to the target terminal in a preset manner, and the preset manner includes an email method and a message push method.
  • the risk identification engine is used to decouple the risk assessment model from the risk scenario, and the multi-dimensional behavior characteristics of the target object are identified and evaluated, which not only makes the risk identification rules scalable, but also improves the efficiency of risk identification
  • the visualization application based on the two-dimensional map intuitively displays the risk behavior of the target object, which improves the intuitiveness and speed of behavior risk identification.
  • FIG. 5 is a schematic structural diagram of a visualization device for behavioral risk recognition provided by an embodiment of the present application.
  • the visualization device 500 for behavioral risk recognition may have relatively large differences due to different configurations or performance, and may include one or more processors (Central Processing units (CPU) 510 (for example, one or more processors) and memory 520, and one or more storage media 530 for storing application programs 533 or data 532 (for example, one or one storage device with a large amount of storage).
  • the memory 520 and the storage medium 530 may be short-term storage or persistent storage.
  • the program stored in the storage medium 530 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the visualization device 500 for identifying behavioral risks.
  • the processor 510 may be configured to communicate with the storage medium 530, and execute a series of instruction operations in the storage medium 530 on the visualization device 500 for behavior risk identification.
  • the visualization device 500 for behavioral risk identification may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input and output interfaces 560, and/or one or more operating systems 531, for example Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc.
  • operating systems 531 for example Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc.
  • the present application also provides a computer-readable storage medium.
  • the computer-readable storage medium may be non-volatile or volatile. Instructions are stored in the computer-readable storage medium, and when the instructions are run on a computer, the computer executes the steps of the visualization method for behavioral risk identification.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of the present application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in each embodiment of the present application.
  • the aforementioned storage medium includes. U disk, mobile hard disk, read-only memory (read-only memory, ROM), random access memory (random access memory, RAM), magnetic disk or optical disk and other media that can store program code.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Engineering & Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Databases & Information Systems (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Operations Research (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种行为风险识别的可视化方法、装置、设备及存储介质,可在智慧城市的智慧安防领域中实现。该方法包括:从预设埋点机制的终端中采集目标对象的第一行为数据、埋点标识和目标对象标识;按照埋点标识、目标对象标识和预置特征类型对目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征(102);通过预置风险识别引擎对目标对象的多维度行为特征进行风险评估,得到多个评估结果(103);当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用处理第一行为数据和第二行为数据,得到目标可视化界面。采用该方法,可提高行为风险识别的直观性和快捷性。

Description

行为风险识别的可视化方法、装置、设备及存储介质
本申请要求于2020年05月27日提交中国专利局、申请号为202010461003.0,发明名称为“行为风险识别的可视化方法、装置、设备及存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及大数据技术的数据可视化领域,尤其涉及一种行为风险识别的可视化方法、装置、设备及存储介质。
背景技术
随着互联网的日益发展,行为风险识别和评估越来越受到业务的重视,业内识别人员行为风险的通用方式通常都是确定用于员工行为风险预警的风险属性条目、采集目标员工的数据、将目标员工属性特征输入员工行为风险预警模型和获取目标员工的风险概率分数。
发明人意识到,现有的风险识别技术一般对行为数据特征挖掘单一,并对应同一类型的行为风险场景,针对不同类型的行为风险场景需重新开发新的风险识别流程,导致对多维度风险识别效率较低。另外,风险识别结果通常以获取某个风险评分指标为目标,评分结果难以涵盖所有风险含义,也无法回溯人员行为历史,导致无法确定风险识别结果是否可靠的问题。
技术问题
本申请的主要目的在于解决了现有的风险识别技术对多维度行为特征识别效率较低,以及很难回溯人员行为历史导致无法确定风险判定结果是否可靠的问题。
技术解决方案
为实现上述目的,本申请第一方面提供了一种行为风险识别的可视化方法,包括:从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
本申请第二方面提供了一种行为风险识别的可视化装置,包括:采集模块,用于从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;挖掘模块,用于根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;评估模块,用于通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;展示模块,当检测到至少一个评估结果存在风险时,用于获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
本申请第三方面提供了一种行为风险识别的可视化设备,所述行为风险识别的可视化设备包括存储器和处理器,所述处理器、和所述存储器相互连接,其中,所述存储器用于存储计算机程序,所述计算机程序包括程序指令,所述处理器用于执行所述存储器的所述程序指令,其中:从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
本申请的第四方面提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,所述计算机程序包括程序指令,所述程序指令被处理器执行时,用于实现以下步骤:从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
有益效果
本申请提供的技术方案中,从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。本申请实施例中,通过风险识别引擎将风险评估模型与风险场景解耦,并对目标对象的多维度行为特征进行识别与评估,不仅使风险识别规则具备可扩展性,还提高了风险识别效率,另外,基于二维地图的可视化应用直观展示目标对象的风险行为,提高了对行为风险识别的直观性和快捷性。
附图说明
图1为本申请实施例中行为风险识别的可视化方法的一个实施例示意图。
图2为本申请实施例中行为风险识别的可视化方法的另一个实施例示意图。
图3为本申请实施例中行为风险识别的可视化装置的一个实施例示意图。
图4为本申请实施例中行为风险识别的可视化装置的另一个实施例示意图。
图5为本申请实施例中行为风险识别的可视化设备的一个实施例示意图。
本发明的实施方式
本申请实施例提供了一种行为风险识别的可视化方法、装置、设备及存储介质,用于通过风险识别引擎将风险评估模型与风险场景解耦,对目标对象的多维度行为特征进行识别与评估,提高了风险识别效率。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外,术语“包括”或“具有”及其任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、***、产品或设备不必限于清楚地列出的那些步骤或单元,而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
为便于理解,下面对本申请实施例的具体流程进行描述,请参阅图1,本申请实施例中行为风险识别的可视化方法的一个实施例包括。
101、从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,标识数据包括埋点标识和目标对象标识。
其中,预设埋点机制是为了满足快捷、高效、丰富的数据应用而预先设置的数据采集机制,预设埋点机制用于在终端采集并记录目标对象行为的过程及结果,并将该目标对象行为的过程及结果上报到服务器。预设埋点机制的终端包括手机、车载自动诊断装置和考勤设备,进一步地,服务器通过手机、车载自动诊断装置和考勤设备采集目标对象的第一行为数据和标识数据,其中,标识数据包括埋点标识和目标对象标识,例如,埋点标识pay_fail用于标识订单支付失败的埋点事件,目标对象标识可以为user_A,埋点标识和目标对象标识可以采用大小写英文、数字以及下划线,但是不能以数字开头。目标对象的第一行为数据包括目标定位***数据和相关的业务数据,目标定位***数据为全球定位***GPS数据。服务器将目标对象的第一行为数据、埋点标识和目标对象标识存入预置数据库中。
可以理解的是,本申请的执行主体可以为行为风险识别的可视化装置,还可以是终端或者服务器,具体此处不做限定。本申请实施例以服务器为执行主体为例进行说明。
102、根据埋点标识、目标对象标识和预置特征类型对目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征。
进一步地,服务器按照埋点标识和目标对象标识将目标对象的第一行为数据存入分布式数据库hbase中,利用分布式***基础架构hadoop提供的海量数据处理短发和数据库空间计算能力,服务器按照预置特征类型实时挖掘目标对象的多维度行为特征,其中,目标对象的多维度行为特征包括目标定位***密度特征、轨迹合成特征、滞留时长特征、滞留地点特征以及预置业务特征,其中,预置业务特征包括是否超派工区域特征和是否迟到特征,进一步地,服务器将密度特征、轨迹合成特征、滞留时长特征、滞留地点特征存储在对象-关系型数据库postgre中,并按照二进制方式存储,而预置业务特征一般采用对象-关系型数据库mongdb的预置字段表或关系型数据库的预置字段表进行存储。
103、通过预置风险识别引擎对目标对象的多维度行为特征进行风险评估,得到多个评估结果。
其中,预置风险识别引擎采用至少一个风险识别应用接口对目标对象的多维度行为特征进行风险识别与评估处理。具体的,服务器通过预置风险识别引擎对目标对象的多维度行为特征进行识别,得到多个风险判断规则标准化接口,并基于多个风险判断规则标准化接口依次处理目标对象的多维度行为特征,得到多个评估结果,例如,3个评估结果A、B和C,进一步地,服务器采用目标对象标识进行关联目标对象的多维度行为特征与多个评估结果之间的映射关系。
需要说明的是,服务器可以对终端上传目标对象的第一行为数据后直接调用预置风险识别引擎进行识别处理,也可以按照预设时长定时调用预置风险识别引擎进行识别处理,其中,预设时长可以为1分钟,也可以为5分钟,具体此处不做限定。
104、当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对目标对象的第一行为数据和目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,第二行为数据生成于第一行为数据生成之前,目标可视化界面用于按照二维地图方式显示第一行为数据和/或第二行为数据表征的风险行为。
当检测到至少一个评估结果存在风险时,服务器可以根据目标对象标识从预置数据库中读取目标对象的第二行为数据,例如,3个评估结果A、B和C,评估结果A和评估结果C不存在风险,但是评估结果A存在风险,则服务器获取目标定位***数据。服务器对目标对象的第一行为数据和目标对象的第二行为数据进行数据封装,并通过预置二维地图应用对应的预置二维地图接口对已封装的数据进行渲染处理,得到目标可视化界面,并展示目标可视化界面。
进一步地,基于目标可视化界面直观展示目标对象的第一行为数据和/或第二行为数据表征的风险行为,以使得对目标对象的第一行为数据和/或第二行为数据表征的风险行为可回溯和可视化,并且通过预置确认机制反馈给预置风险识别引擎,优化预置风险识别引擎,快速高效地提升风险识别率。
本申请实施例中,通过风险识别引擎将风险评估模型与风险场景解耦,并对目标对象的多维度行为特征进行识别与评估,不仅使风险识别规则具备可扩展性,还提高了风险识别效率,另外,基于二维地图的可视化应用直观展示目标对象的风险行为,提高了对行为风险识别的直观性和快捷性。本方案可应用于智慧安防领域中,从而推动智慧城市的建设。
请参阅图2,本申请实施例中行为风险识别的可视化方法的另一个实施例包括。
201、从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,标识数据包括埋点标识和目标对象标识。
具体的,首先,服务器接收预设埋点机制的终端发送的埋点数据,其中,埋点数据为基于预设的埋点事件采集的数据,例如,对于车险业务,从查勘员出险时在手机业务应用中操作案件的埋点数据,什么时刻上传了哪些图片、结案时如何将赔付金额进行调整、账号登录记录以及查勘照片拍摄数据,预设埋点机制的终端包括手机、车载自动诊断装置和考勤设备,可选的,服务器对终端设置埋点,得到预设埋点机制的终端。服务器定时接收预设埋点机制的终端发送的埋点数据,例如,每隔10秒或者15秒采集一次目标定位***数据或者考勤数据,该考勤数据包括打卡时刻、打卡地点和打卡对象。
其次,服务器对埋点数据按照第一预设数据格式进行数据解析,得到目标对象的第一行为数据和标识数据,标识数据包括埋点标识和目标对象标识,其中,第一预设数据格式为JS 对象简谱的数据交换格式(javaScriptobject notation,JSON),埋点数据中的字段包括埋点标识buryId,目标对象标识userId、经纬度上传时刻gpsTime、经纬度gps和移动速度speed,而经纬度可以逗号分隔,例如,标识数据{"ids_data":["buryId":"23","userId":"USER_A"}},目标对象的第一行为数据 {"gpsTime":"2020-1-12 23:54:11","speed":"21.1","gps":"112.3232,28.322423"}。进一步地,服务器将埋点数据中的埋点标识与预置业务场景的标识进行关联查询,得到预置业务场景。服务器基于预置业务场景对目标行为数据进行数据封装并保存到预置数据库中。
202、根据埋点标识、目标对象标识和预置特征类型对目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征。
具体的,服务器基于埋点标识判断目标对象的第一行为数据是否属于目标定位***数据。若目标对象的第一行为数据不属于目标定位***数据,则服务器按照预置特征类型对目标对象的第一行为数据进行业务特征挖掘,得到第一行为特征,其中,第一行为特征包括是否超派工区域特征、是否迟到特征和行驶中是否查询案件特征,具体根据预置特征类型进行筛选确定,具体此处不做限定。若目标对象的第一行为数据属于目标定位***数据,则服务器对目标对象的第一行为数据进行特征提取,并将提取到的特征存储到预置数据库中,得到第二行为特征,第二行为特征包括目标定位***密度特征、轨迹合成特征和滞留地点特征和滞留时长特征,例如,服务器可以采用基于密度的噪声应用空间聚类DBSCAN算法对目标对象的第一行为数据进行聚类计算,聚类后将每个簇的目标定位***的数量作为簇的密度,得到目标定位***密度特征,其中,点与点之间的距离采用地球表面距离,聚类临界值和簇的点数根据实际业务需求预先设置,默认为50米内有20个点。而滞留地点特征和滞留时长特征,则通过服务器便利遍历每条轨迹,找出速度小于5km/h且5分钟内目标定位***的起点和终点距离小于500米的一段轨迹,服务器将该轨迹第一个点作为滞留点,轨迹时长作为滞留时长,例如,滞留特征为{behaviorFeature:{"stayGps":"112.3232,28.322423","stayDuration":"23"}},其中,stayGps用于指示滞留地点特征和stayDuration用于指示滞留时长特征。进一步地,服务器将目标对象的第一行为数据中的目标定位***数据按照时刻先后顺序进行排序,得到轨迹合成特征,当在预置时长内未接收到新的目标对象的第一行为数据时,则服务器将该轨迹合成特征入库,其中预置时长可以为10分钟。最后,服务器将第一行为特征和第二行为特征进行合并,并将合并后的特征数据与目标对象标识进行关系映射,得到目标对象的多维度行为特征。
203、通过预置风险识别引擎对目标对象的多维度行为特征进行匹配识别,得到风险规则数据集,其中,风险规则数据集包括多个风险规则。
其中,多个风险规则为按照业务预先配置在预置风险识别引擎中的,进一步地,服务器获取目标对象的多维度行为特征各自对应的预置业务类型。服务器按照各自对应的预置业务类型查询预置规则数据表,得到风险规则数据集,风险规则数据集包括多个风险规则,例如,是否迟到的风险规则和是否按时交接班的风险规则。
204、按照多个风险规则分别调用多个预置风险评估模型,以使得多个预置风险评估模型对目标对象的多维度行为特征进行风险评估,得到多个风险分数或多个风险等级,并将多个风险分数或多个风险等级设置为多个评估结果。
服务器按照多个风险规则分别调用多个预置风险评估模型,以使得多个预置风险评估模型对目标对象的多维度行为特征进行风险评估,得到多个风险分数。进一步地,服务器按照风险分数确定多个风险等级,并将多个风险分数或多个风险等级设置为多个评估结果。例如,每当打卡设备上传一次考勤数据时,服务器将目标对象标识userId传给预置风险识别引擎,预置风险识别引擎确定对应的风险规则为判断是否迟到的规则,服务器通过对应的预置风险评估模型和目标对象标识userId查找当天排班班次表和派工区域地理边界,并根据目标对象标识userId从预置数据库中获取出打卡定位数据,再按照打卡定位数据和派工区域地里边界进行空间包含运算,将打卡时刻与班次表做比较,如果满足不在派工区域内某班次前打卡,则确定存在迟到风险,也就是评估结果,评估结果可以采用风险分数或者风险等级进行标记,例如,风险等级为3级,风险分数为10。
可选的,服务器按照预置业务需求获取行为风险场景,并对行为风险场景设置行为风险标准。服务器获取样本数据,并基于样本数据和已制定的行为风险标准训练初始行为风险模型,得到预置行为风险模型。服务器将预置行为风险模型添加到预置风险识别引擎中,并对预置行为风险模型配置对应的风险规则,预置行为风险模型用于接受预置风险识别引擎的调用并输出对应的评估结果。
205、当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对目标对象的第一行为数据和目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,第二行为数据生成于第一行为数据生成之前,目标可视化界面用于按照二维地图方式显示第一行为数据和/或第二行为数据表征的风险行为。
具体的,当检测到至少一个评估结果存在风险时,服务器按照目标对象标识获取目标对象的第二行为数据,其中,第二行为数据生成于第一行为数据生成之前。服务器按照第二预设数据格式对目标对象的第一行为数据和目标对象的第二行为数据进行数据封装,得到已封装的数据,进一步地,服务器从目标对象的第一行为数据和目标对象的第二行为数据中提取多个目标定位***数据,多每个目标定位***数据包括经纬度坐标,服务器基于经纬度坐标进行数据封装,其中,第二预设数据格式为基于脚本对象表示法的地理空间信息数据交换格式GeoJSON,也就是对各种地理数据结构进行编码的格式。服务器调用预置二维地图应用对已封装的数据进行图层渲染并展示,得到目标可视化界面,目标可视化界面用于按照二维地图方式显示目标对象的第一行为数据和/或第二行为数据表征的风险行为。例如,采用二维地图展示目标人员A在2020-03-03处理车险业务时的车辆轨迹信息,车辆轨迹信息存在不符合处理车险业务的预置行驶路径的风险。进一步地,服务器通过预置确认机制优化预置风险识别引擎,快速提高风险识别效率。
206、获取目标图片,基于预设模板对目标对象的多维度行为特征、多个评估结果和目标图片进行组合,得到风险评估报告,其中,目标图片为包含目标可视化界面的图片。
具体的,服务器接收终端发送的目标图片,并将目标图片存储到预置目录中,目标图片为包含目标可视化界面的图片。服务器按照目标对象标识从预置数据库中读取目标对象的多维度行为特征和多个评估结果。服务器基于预设模板对目标对象的多维度行为特征、多个评估结果和目标图片进行组合,得到风险评估报告。
207、按照预置方式将风险评估报告发送到目标终端,其中,预置方式包括邮件方式和消息推送方式。
具体的,服务器按照目标对象标识从预置数据库中查询得到目标终端的终端标识。服务器按照目标终端的终端标识确定预置方式,该预置方式包括邮件方式和消息推送方式。服务器按照预置方式和终端标识将风险评估报告发送到目标终端。
本申请实施例中,通过风险识别引擎将风险评估模型与风险场景解耦,并对目标对象的多维度行为特征进行识别与评估,不仅使风险识别规则具备可扩展性,还提高了风险识别效率,另外,基于二维地图的可视化应用直观展示目标对象的风险行为,提高了对行为风险识别的直观性和快捷性。本方案属于智慧安防领域,通过本方案能够推动智慧城市的建设。
上面对本申请实施例中行为风险识别的可视化方法进行了描述,下面对本申请实施例中行为风险识别的可视化装置进行描述,请参阅图3,本申请实施例中行为风险识别的可视化装置的一个实施例包括。
采集模块301,用于从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,标识数据包括埋点标识和目标对象标识。
挖掘模块302,用于根据埋点标识、目标对象标识和预置特征类型对目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征。
评估模块303,用于通过预置风险识别引擎对目标对象的多维度行为特征进行风险评估,得到多个评估结果。
展示模块304,当检测到至少一个评估结果存在风险时,用于获取目标对象的第二行为数据,并通过预置二维地图应用对目标对象的第一行为数据和目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,第二行为数据生成于第一行为数据生成之前,目标可视化界面用于按照二维地图方式显示第一行为数据和/或第二行为数据表征的风险行为。
本申请实施例中,通过风险识别引擎将风险评估模型与风险场景解耦,并对目标对象的多维度行为特征进行识别与评估,不仅使风险识别规则具备可扩展性,还提高了风险识别效率,另外,基于二维地图的可视化应用直观展示目标对象的风险行为,提高了对行为风险识别的直观性和快捷性。本方案可应用于智慧安防领域中,从而推动智慧城市的建设。
请参阅图4,本申请实施例中行为风险识别的可视化装置的另一个实施例包括。
采集模块301,用于从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,标识数据包括埋点标识和目标对象标识。
挖掘模块302,用于根据埋点标识、目标对象标识和预置特征类型对目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征。
评估模块303,用于通过预置风险识别引擎对目标对象的多维度行为特征进行风险评估,得到多个评估结果。
展示模块304,当检测到至少一个评估结果存在风险时,用于获取目标对象的第二行为数据,并通过预置二维地图应用对目标对象的第一行为数据和目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,第二行为数据生成于第一行为数据生成之前,目标可视化界面用于按照二维地图方式显示第一行为数据和/或第二行为数据表征的风险行为。
可选的,采集模块301还可以具体用于。
接收预设埋点机制的终端发送的埋点数据,预设埋点机制的终端包括手机、车载自动诊断装置和考勤设备。
对埋点数据按照第一预设数据格式进行数据解析,得到目标对象的第一行为数据和标识数据。
可选的,挖掘模块302还可以具体用于。
基于埋点标识判断目标对象的第一行为数据是否属于目标定位***数据。
若目标对象的第一行为数据不属于目标定位***数据,则按照预置特征类型对目标对象的第一行为数据进行业务特征挖掘,得到第一行为特征。
若目标对象的第一行为数据属于目标定位***数据,则对目标对象的第一行为数据进行特征提取,并将提取到的特征存储到预置数据库中,得到第二行为特征,第二行为特征包括目标定位***密度特征、轨迹合成特征和滞留地点特征。
将第一行为特征和第二行为特征进行合并,并将合并后的特征数据与目标对象标识进行关系映射,得到目标对象的多维度行为特征。
可选的,评估模块303还可以具体用于。
通过预置风险识别引擎对目标对象的多维度行为特征进行匹配识别,得到风险规则数据集,其中,风险规则数据集包括多个风险规则。
按照多个风险规则分别调用多个预置风险评估模型,以使得多个预置风险评估模型对目标对象的多维度行为特征进行风险评估,得到多个风险分数或多个风险等级,并将多个风险分数或多个风险等级设置为多个评估结果。
可选的,展示模块304还可以具体用于。
当检测到至少一个评估结果存在风险时,按照目标对象标识获取目标对象的第二行为数据,其中,第二行为数据生成于第一行为数据生成之前。
按照第二预设数据格式对目标对象的第一行为数据和目标对象的第二行为数据进行数据封装,得到已封装的数据。
调用预置二维地图应用对已封装的数据进行图层渲染并展示,得到目标可视化界面,其中,目标可视化界面用于按照二维地图方式显示第一行为数据和/或第二行为数据表征的风险行为。
可选的,行为风险识别的可视化装置还包括。
设置模块305,用于按照预置业务需求获取行为风险场景,并对行为风险场景设置行为风险标准,得到已制定的行为风险标准。
训练模块306,用于获取样本数据,并基于样本数据和已制定的行为风险标准训练初始行为风险模型,得到预置行为风险模型。
配置模块307,用于将预置行为风险模型添加到预置风险识别引擎中,并对预置行为风险模型配置对应的风险规则,预置行为风险模型用于接受预置风险识别引擎的调用并输出对应的评估结果。
可选的,行为风险识别的可视化装置还包括。
组合模块308,用于获取目标图片,基于预设模板对目标对象的多维度行为特征、多个评估结果和目标图片进行组合,得到风险评估报告,目标图片为包含目标可视化界面的图片。
发送模块309,用于按照预置方式将风险评估报告发送到目标终端,预置方式包括邮件方式和消息推送方式。
本申请实施例中,通过风险识别引擎将风险评估模型与风险场景解耦,并对目标对象的多维度行为特征进行识别与评估,不仅使风险识别规则具备可扩展性,还提高了风险识别效率,另外,基于二维地图的可视化应用直观展示目标对象的风险行为,提高了对行为风险识别的直观性和快捷性。本方案属于智慧安防领域,通过本方案能够推动智慧城市的建设。
上面图3和图4从模块化功能实体的角度对本申请实施例中的行为风险识别的可视化装置进行详细描述,下面从硬件处理的角度对本申请实施例中行为风险识别的可视化设备进行详细描述。
图5是本申请实施例提供的一种行为风险识别的可视化设备的结构示意图,该行为风险识别的可视化设备500可因配置或性能不同而产生比较大的差异,可以包括一个或一个以上处理器(central processing units,CPU)510(例如,一个或一个以上处理器)和存储器520,一个或一个以上存储应用程序533或数据532的存储介质530(例如一个或一个以上海量存储设备)。其中,存储器520和存储介质530可以是短暂存储或持久存储。存储在存储介质530的程序可以包括一个或一个以上模块(图示没标出),每个模块可以包括对行为风险识别的可视化设备500中的一系列指令操作。更进一步地,处理器510可以设置为与存储介质530通信,在行为风险识别的可视化设备500上执行存储介质530中的一系列指令操作。
行为风险识别的可视化设备500还可以包括一个或一个以上电源540,一个或一个以上有线或无线网络接口550,一个或一个以上输入输出接口560,和/或,一个或一个以上操作***531,例如Windows Serve,Mac OS X,Unix,Linux,FreeBSD等等。本领域技术人员可以理解,图5示出的行为风险识别的可视化设备结构并不构成对行为风险识别的可视化设备的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置。
本申请还提供一种计算机可读存储介质,所述计算机可读存储介质可以是非易失性,也可以是易失性。所述计算机可读存储介质中存储有指令,当所述指令在计算机上运行时,使得计算机执行所述行为风险识别的可视化方法的步骤。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的***,装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括。U盘、移动硬盘、只读存储器(read-only memory, ROM)、随机存取存储器(random access memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应以所述权利要求的保护范围为准。

Claims (20)

  1. 一种行为风险识别的可视化方法,其中,所述行为风险识别的可视化方法包括:
    从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;
    根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;
    通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;
    当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
  2. 根据权利要求1所述的行为风险识别的可视化方法,其中,所述从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,包括:
    接收预设埋点机制的终端发送的埋点数据,所述预设埋点机制的终端包括手机、车载自动诊断装置和考勤设备;
    对所述埋点数据按照第一预设数据格式进行数据解析,得到目标对象的第一行为数据和标识数据。
  3. 根据权利要求1所述的行为风险识别的可视化方法,其中,所述根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征,包括:
    基于所述埋点标识判断所述目标对象的第一行为数据是否属于目标定位***数据;
    若所述目标对象的第一行为数据不属于目标定位***数据,则按照所述预置特征类型对所述目标对象的第一行为数据进行业务特征挖掘,得到第一行为特征;
    若所述目标对象的第一行为数据属于目标定位***数据,则对所述目标对象的第一行为数据进行特征提取,并将提取到的特征存储到预置数据库中,得到第二行为特征,所述第二行为特征包括目标定位***密度特征、轨迹合成特征和滞留地点特征;
    将所述第一行为特征和所述第二行为特征进行合并,并将合并后的特征数据与所述目标对象标识进行关系映射,得到目标对象的多维度行为特征。
  4. 根据权利要求1所述的行为风险识别的可视化方法,其中,所述通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果,包括:
    通过预置风险识别引擎对所述目标对象的多维度行为特征进行匹配识别,得到风险规则数据集,其中,所述风险规则数据集包括多个风险规则;
    按照所述多个风险规则分别调用多个预置风险评估模型,以使得所述多个预置风险评估模型对所述目标对象的多维度行为特征进行风险评估,得到多个风险分数或多个风险等级,并将所述多个风险分数或所述多个风险等级设置为多个评估结果。
  5. 根据权利要求1所述的行为风险识别的可视化方法,其中,所述当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为,包括:
    当检测到所述至少一个评估结果存在风险时,按照所述目标对象标识获取目标对象的第二行为数据,其中,所述第二行为数据生成于所述第一行为数据生成之前;
    按照第二预设数据格式对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行数据封装,得到已封装的数据;
    调用预置二维地图应用对所述已封装的数据进行图层渲染并展示,得到目标可视化界面,其中,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
  6. 根据权利要求1-5中任一项所述的行为风险识别的可视化方法,其中,在所述当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为之前,所述行为风险识别的可视化方法还包括:
    按照预置业务需求获取行为风险场景,并对所述行为风险场景设置行为风险标准,得到已制定的行为风险标准;
    获取样本数据,并基于所述样本数据和所述已制定的行为风险标准训练初始行为风险模型,得到预置行为风险模型;
    将所述预置行为风险模型添加到所述预置风险识别引擎中,并对所述预置行为风险模型配置对应的风险规则,所述预置行为风险模型用于接受所述预置风险识别引擎的调用并输出对应的评估结果。
  7. 根据权利要求1-5中任一项所述的行为风险识别的可视化方法,其中,在所述当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为之后,所述行为风险识别的可视化方法还包括:
    获取目标图片,基于预设模板对所述目标对象的多维度行为特征、所述多个评估结果和所述目标图片进行组合,得到风险评估报告,其中,所述目标图片为包含所述目标可视化界面的图片;
    按照预置方式将所述风险评估报告发送到目标终端,其中,所述预置方式包括邮件方式和消息推送方式。
  8. 一种行为风险识别的可视化装置,其中,所述行为风险识别的可视化装置包括:
    采集模块,用于从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;
    挖掘模块,用于根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;
    评估模块,用于通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;
    展示模块,当检测到至少一个评估结果存在风险时,用于获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
  9. 一种行为风险识别的可视化设备,其中,所述行为风险识别的可视化设备包括存储器和处理器,所述处理器、和所述存储器相互连接,其中,所述存储器用于存储计算机程序,所述计算机程序包括程序指令,所述处理器用于执行所述存储器的所述程序指令,其中:
    从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;
    根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;
    通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;
    当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
  10. 根据权利要求9所述的行为风险识别的可视化设备,其中,所述处理器用于:
    接收预设埋点机制的终端发送的埋点数据,所述预设埋点机制的终端包括手机、车载自动诊断装置和考勤设备;
    对所述埋点数据按照第一预设数据格式进行数据解析,得到目标对象的第一行为数据和标识数据。
  11. 根据权利要求9所述的行为风险识别的可视化设备,其中,所述处理器用于:
    基于所述埋点标识判断所述目标对象的第一行为数据是否属于目标定位***数据;
    若所述目标对象的第一行为数据不属于目标定位***数据,则按照所述预置特征类型对所述目标对象的第一行为数据进行业务特征挖掘,得到第一行为特征;
    若所述目标对象的第一行为数据属于目标定位***数据,则对所述目标对象的第一行为数据进行特征提取,并将提取到的特征存储到预置数据库中,得到第二行为特征,所述第二行为特征包括目标定位***密度特征、轨迹合成特征和滞留地点特征;
    将所述第一行为特征和所述第二行为特征进行合并,并将合并后的特征数据与所述目标对象标识进行关系映射,得到目标对象的多维度行为特征。
  12. 根据权利要求9所述的行为风险识别的可视化设备,其中,所述处理器用于:
    通过预置风险识别引擎对所述目标对象的多维度行为特征进行匹配识别,得到风险规则数据集,其中,所述风险规则数据集包括多个风险规则;
    按照所述多个风险规则分别调用多个预置风险评估模型,以使得所述多个预置风险评估模型对所述目标对象的多维度行为特征进行风险评估,得到多个风险分数或多个风险等级,并将所述多个风险分数或所述多个风险等级设置为多个评估结果。
  13. 根据权利要求9所述的行为风险识别的可视化设备,其中,所述处理器用于:
    当检测到所述至少一个评估结果存在风险时,按照所述目标对象标识获取目标对象的第二行为数据,其中,所述第二行为数据生成于所述第一行为数据生成之前;
    按照第二预设数据格式对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行数据封装,得到已封装的数据;
    调用预置二维地图应用对所述已封装的数据进行图层渲染并展示,得到目标可视化界面,其中,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
  14. 根据权利要求9-13中任一项所述的行为风险识别的可视化设备,其中,所述处理器用于:
    按照预置业务需求获取行为风险场景,并对所述行为风险场景设置行为风险标准,得到已制定的行为风险标准;
    获取样本数据,并基于所述样本数据和所述已制定的行为风险标准训练初始行为风险模型,得到预置行为风险模型;
    将所述预置行为风险模型添加到所述预置风险识别引擎中,并对所述预置行为风险模型配置对应的风险规则,所述预置行为风险模型用于接受所述预置风险识别引擎的调用并输出对应的评估结果。
  15. 根据权利要求9-13中任一项所述的行为风险识别的可视化设备,其中,所述处理器用于:
    获取目标图片,基于预设模板对所述目标对象的多维度行为特征、所述多个评估结果和所述目标图片进行组合,得到风险评估报告,其中,所述目标图片为包含所述目标可视化界面的图片;
    按照预置方式将所述风险评估报告发送到目标终端,其中,所述预置方式包括邮件方式和消息推送方式。
  16. 一种计算机可读存储介质,其中,所述计算机可读存储介质存储有计算机程序,所述计算机程序包括程序指令,所述程序指令被处理器执行时,用于实现以下步骤:
    从预设埋点机制的终端中采集目标对象的第一行为数据和标识数据,其中,所述标识数据包括埋点标识和目标对象标识;
    根据所述埋点标识、所述目标对象标识和预置特征类型对所述目标对象的第一行为数据进行数据挖掘,得到目标对象的多维度行为特征;
    通过预置风险识别引擎对所述目标对象的多维度行为特征进行风险评估,得到多个评估结果;
    当检测到至少一个评估结果存在风险时,获取目标对象的第二行为数据,通过预置二维地图应用对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行处理,生成并展示目标可视化界面,其中,所述第二行为数据生成于所述第一行为数据生成之前,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
  17. 根据权利要求16所述的计算机可读存储介质,其中,所述程序指令被处理器执行时,还用于实现以下步骤:
    接收预设埋点机制的终端发送的埋点数据,所述预设埋点机制的终端包括手机、车载自动诊断装置和考勤设备;
    对所述埋点数据按照第一预设数据格式进行数据解析,得到目标对象的第一行为数据和标识数据。
  18. 根据权利要求16所述的计算机可读存储介质,其中,所述程序指令被处理器执行时,还用于实现以下步骤:
    基于所述埋点标识判断所述目标对象的第一行为数据是否属于目标定位***数据;
    若所述目标对象的第一行为数据不属于目标定位***数据,则按照所述预置特征类型对所述目标对象的第一行为数据进行业务特征挖掘,得到第一行为特征;
    若所述目标对象的第一行为数据属于目标定位***数据,则对所述目标对象的第一行为数据进行特征提取,并将提取到的特征存储到预置数据库中,得到第二行为特征,所述第二行为特征包括目标定位***密度特征、轨迹合成特征和滞留地点特征;
    将所述第一行为特征和所述第二行为特征进行合并,并将合并后的特征数据与所述目标对象标识进行关系映射,得到目标对象的多维度行为特征。
  19. 根据权利要求16所述的计算机可读存储介质,其中,所述程序指令被处理器执行时,还用于实现以下步骤:
    通过预置风险识别引擎对所述目标对象的多维度行为特征进行匹配识别,得到风险规则数据集,其中,所述风险规则数据集包括多个风险规则;
    按照所述多个风险规则分别调用多个预置风险评估模型,以使得所述多个预置风险评估模型对所述目标对象的多维度行为特征进行风险评估,得到多个风险分数或多个风险等级,并将所述多个风险分数或所述多个风险等级设置为多个评估结果。
  20. 根据权利要求16所述的计算机可读存储介质,其中,所述程序指令被处理器执行时,还用于实现以下步骤:
    当检测到所述至少一个评估结果存在风险时,按照所述目标对象标识获取目标对象的第二行为数据,其中,所述第二行为数据生成于所述第一行为数据生成之前;
    按照第二预设数据格式对所述目标对象的第一行为数据和所述目标对象的第二行为数据进行数据封装,得到已封装的数据;
    调用预置二维地图应用对所述已封装的数据进行图层渲染并展示,得到目标可视化界面,其中,所述目标可视化界面用于按照二维地图方式显示所述第一行为数据和/或所述第二行为数据表征的风险行为。
PCT/CN2020/098880 2020-05-27 2020-06-29 行为风险识别的可视化方法、装置、设备及存储介质 WO2021114615A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010461003.0 2020-05-27
CN202010461003.0A CN111738558B (zh) 2020-05-27 2020-05-27 行为风险识别的可视化方法、装置、设备及存储介质

Publications (1)

Publication Number Publication Date
WO2021114615A1 true WO2021114615A1 (zh) 2021-06-17

Family

ID=72647846

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/098880 WO2021114615A1 (zh) 2020-05-27 2020-06-29 行为风险识别的可视化方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN111738558B (zh)
WO (1) WO2021114615A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112613729A (zh) * 2020-12-19 2021-04-06 前海飞算科技(深圳)有限公司 违约风险大数据可视化方法、装置及存储介质
CN114445088A (zh) * 2022-01-13 2022-05-06 内蒙古蒙商消费金融股份有限公司 一种欺诈行为的判定方法、装置、电子设备和存储介质
CN114969169A (zh) * 2022-05-25 2022-08-30 上海东普信息科技有限公司 签收数据监控方法、装置、设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150235321A1 (en) * 2014-02-18 2015-08-20 Mastercard International Incorporated Insurance risk modeling method and apparatus
CN107871279A (zh) * 2017-09-30 2018-04-03 上海壹账通金融科技有限公司 用户身份验证方法及应用服务器
US20180351991A1 (en) * 2015-10-12 2018-12-06 Palantir Technologies Inc. Systems for computer network security risk assessment including user compromise analysis associated with a network of devices
CN109327439A (zh) * 2018-09-29 2019-02-12 武汉极意网络科技有限公司 业务请求数据的风险识别方法、装置、存储介质及设备
CN110570097A (zh) * 2019-08-15 2019-12-13 中国平安人寿保险股份有限公司 基于大数据的业务人员风险识别方法、装置及存储介质
CN111199474A (zh) * 2019-12-16 2020-05-26 北京淇瑀信息科技有限公司 一种基于双方网络图数据的风险预测方法、装置和电子设备

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011063269A1 (en) * 2009-11-20 2011-05-26 Alert Enterprise, Inc. Method and apparatus for risk visualization and remediation
ES2755780T3 (es) * 2011-09-16 2020-04-23 Veracode Inc Análisis estático y de comportamiento automatizado mediante la utilización de un espacio aislado instrumentado y clasificación de aprendizaje automático para seguridad móvil
CN103793774A (zh) * 2012-11-02 2014-05-14 王亚利 对碳排放或碳减排行为实现管理的账户***及方法
US11250364B2 (en) * 2017-12-06 2022-02-15 Hartford Fire Insurance Company System and method for evaluating images to support multiple risk applications
CN108876600B (zh) * 2018-08-20 2023-09-05 平安科技(深圳)有限公司 预警信息推送方法、装置、计算机设备和介质
CN110619568A (zh) * 2019-09-17 2019-12-27 王文斌 风险评估报告的生成方法、装置、设备及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150235321A1 (en) * 2014-02-18 2015-08-20 Mastercard International Incorporated Insurance risk modeling method and apparatus
US20180351991A1 (en) * 2015-10-12 2018-12-06 Palantir Technologies Inc. Systems for computer network security risk assessment including user compromise analysis associated with a network of devices
CN107871279A (zh) * 2017-09-30 2018-04-03 上海壹账通金融科技有限公司 用户身份验证方法及应用服务器
CN109327439A (zh) * 2018-09-29 2019-02-12 武汉极意网络科技有限公司 业务请求数据的风险识别方法、装置、存储介质及设备
CN110570097A (zh) * 2019-08-15 2019-12-13 中国平安人寿保险股份有限公司 基于大数据的业务人员风险识别方法、装置及存储介质
CN111199474A (zh) * 2019-12-16 2020-05-26 北京淇瑀信息科技有限公司 一种基于双方网络图数据的风险预测方法、装置和电子设备

Also Published As

Publication number Publication date
CN111738558B (zh) 2024-03-08
CN111738558A (zh) 2020-10-02

Similar Documents

Publication Publication Date Title
US20180349685A1 (en) Identity verification via validated facial recognition and graph database
CN106960006B (zh) 一种不同轨迹间相似度度量***及其度量方法
EP3241370B1 (en) Analyzing semantic places and related data from a plurality of location data reports
WO2021114615A1 (zh) 行为风险识别的可视化方法、装置、设备及存储介质
CN110688495A (zh) 一种事件信息的知识图谱模型构建方法、装置、存储介质
US9313233B2 (en) Systems and methods for detecting associated devices
CN110874362A (zh) 一种数据关联分析方法及装置
WO2017215523A1 (zh) 识别用户所在地理位置的类别的方法及装置
CN104090970A (zh) 兴趣点的展现方法及装置
CN112861972B (zh) 一种展业区域的选址方法、装置、计算机设备和介质
CN111930868A (zh) 一种基于多维数据采集的大数据行为轨迹分析方法
CN110598204B (zh) 基于知识图谱的实体识别数据增强方法及***
US8843480B2 (en) Server, information-management method, information-management program, and computer-readable recording medium with said program recorded thereon, for managing information input by a user
CA2760624C (en) Server, dictionary creation method, dictionary creation program, and computer-readable recording medium recording the program
CN102833320A (zh) Gps车辆监控***的网页客户端***
CN114070760B (zh) 一种网络空间资产的测绘方法及相关装置
CN111782980B (zh) 地图兴趣点的挖掘方法、装置、设备及存储介质
CN112364176A (zh) 一种人员行动轨迹的构建方法、设备及***
CN111259167B (zh) 用户请求风险识别方法及装置
CN112925899B (zh) 排序模型建立方法、案件线索推荐方法、装置及介质
CN114530038A (zh) 一种基于时空数据聚类的出行兴趣区域提取方法及***
WO2019187107A1 (ja) 情報処理装置、制御方法、及びプログラム
CN110300371B (zh) 一种基于物联网设备行为的计算机设备管理方法
CN111611337B (zh) 终端数据处理***
CN114218453A (zh) 数据展示方法、装置、设备及可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20897710

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205N DATED 24/01/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20897710

Country of ref document: EP

Kind code of ref document: A1