Data processing method, device, equipment and storage medium based on artificial intelligence
By obtaining judge review data and using machine learning models to estimate the review time, intelligently assisting judge review, solving the problem of low efficiency in online case review and improving judges' work efficiency and case handling speed.
Patent Information
- Application Number
- CN202111152130.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-09-29
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2041-09-29
AI Technical Summary
When handling online cases, judges face cumbersome review operations and lack of intelligent auxiliary means, which leads to a backlog of cases and affects work efficiency.
By obtaining the behavioral data of the judge's review of legal documents, generating sample data and storing it in a distributed file system, using the machine learning model to estimate the review time, determining the target review time period and outputting reminder information, intelligently assisting the judge in review.
It improves the convenience and efficiency of judges' review of legal documents, reduces case backlog, and optimizes judicial workflow.
Smart Images

Figure CN113887933B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence, and in particular to a data processing method, device, equipment and storage medium based on artificial intelligence. Background Art
[0002] With the rapid development of Internet technology, more and more companies or government agencies rely on Internet technology to realize business or government affairs. In the judicial field, the People's Court has opened online litigation service channels based on Internet technology, which facilitates users to submit case filing application materials, evidence materials, application materials, sign for document delivery, communicate with judges, etc. through mobile phones, computers and other terminal devices. But at the same time, with the blowout of online cases, judges are also facing problems such as changes in working methods and increased pressure on business processing. Due to the cumbersome online review operations, the lack of intelligent auxiliary means, and the judges' overly busy work tasks, it is often easy to cause a backlog of cases, resulting in cases exceeding the processing time limit, which is not convenient for judges to review. Summary of the invention
[0003] The embodiments of the present application provide a data processing method, apparatus, device and storage medium based on artificial intelligence, aiming to improve the convenience of judges in reviewing legal documents.
[0004] In a first aspect, an embodiment of the present application provides a data processing method, including:
[0005] Obtaining behavioral data of judges reviewing legal documents, and generating sample data based on the behavioral data;
[0006] Obtaining a judge's identity from the behavior data, and storing the sample data in a storage area corresponding to the judge's identity in a distributed file system;
[0007] When the number of the sample data is greater than or equal to a preset number threshold, iteratively training the preset machine learning model based on all the sample data in the storage area to obtain a document review time estimation model corresponding to the judge identity identifier;
[0008] Determine an estimated time value for reviewing the legal document to be reviewed based on the document review time estimation model and the document feature information of the legal document to be reviewed corresponding to the judge's identity identifier;
[0009] Obtaining a work schedule corresponding to the judge's identity identifier, and determining a target review time period for the legal document to be reviewed based on the work schedule and the review time estimate;
[0010] The audit reminder time of the to-be-audited legal document is determined according to the target audit time period, and the audit reminder information is output when the current system time reaches the audit reminder time.
[0011] In a second aspect, an embodiment of the present application further provides a data processing device, the data processing device comprising:
[0012] An acquisition module is used to obtain the behavior data of judges reviewing legal documents;
[0013] A sample generation module, used to generate sample data according to the behavior data;
[0014] A sending module, used for obtaining the judge's identity from the behavior data, and storing the sample data in a storage area corresponding to the judge's identity in a distributed file system;
[0015] A model training module, used for iteratively training a preset machine learning model based on all sample data in the storage area when the number of sample data is greater than or equal to a preset number threshold, to obtain a document review time estimation model corresponding to the judge identity;
[0016] A determination module, used to determine an estimated time for reviewing the legal document to be reviewed based on the document review time estimation model and the document feature information of the legal document to be reviewed corresponding to the judge's identity identifier;
[0017] The determination module is further used to obtain a work schedule corresponding to the judge's identity identifier, and determine a target review time period for the legal document to be reviewed based on the work schedule and the review time estimate;
[0018] The determination module is further used to determine the review reminder time of the to-be-reviewed legal document according to the target review time period;
[0019] The output module is used to output the audit reminder information when the current system time reaches the audit reminder time.
[0020] In a third aspect, an embodiment of the present application further provides a server, the server comprising a processor and a memory;
[0021] The memory is used to store computer programs;
[0022] The processor is used to execute the computer program and implement the steps of the artificial intelligence-based data processing method as described above when executing the computer program.
[0023] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the artificial intelligence-based data processing method as described above are implemented.
[0024] The embodiment of the present application provides a data processing method, device, equipment and storage medium based on artificial intelligence. The method obtains behavioral data of a judge reviewing a legal document, generates sample data based on the behavioral data, and stores the sample data in a storage area corresponding to the judge's identity in a distributed file system. Then, when the number of sample data in the storage area reaches a preset number threshold, iteratively trains a preset machine learning model based on all sample data in the storage area to obtain a file review time estimation model corresponding to the judge's identity, and determines the estimated review time of the legal document to be reviewed based on the file review time estimation model and file feature information of the legal document to be reviewed corresponding to the judge's identity. Finally, according to a work schedule corresponding to the judge's identity and the estimated review time, a target review time period of the legal document to be reviewed is determined, and a review reminder time of the legal document to be reviewed is determined based on the target review time period. When the current system time reaches the review reminder time, the review reminder information is output, so as to intelligently assist the judge in reviewing the legal document and improve the convenience and efficiency of the judge in reviewing the legal document. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0026] Figure 1 This is a schematic diagram of a scenario for implementing the artificial intelligence-based data processing method provided in an embodiment of the present application;
[0027] Figure 2 It is a flowchart of a data processing method based on artificial intelligence provided in an embodiment of the present application;
[0028] Figure 3 yes Figure 2 A schematic diagram of the sub-step flow chart of the data processing method in FIG.
[0029] Figure 4 is a schematic block diagram of a data processing device provided in an embodiment of the present application;
[0030] Figure 5 yes Figure 4 A schematic block diagram of submodules of a data processing device in;
[0031] Figure 6 It is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application.
[0032] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0033] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0034] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.
[0035] With the rapid development of Internet technology, more and more companies or government agencies rely on Internet technology to realize business or government affairs. In the judicial field, the People's Court has opened online litigation service channels based on Internet technology, which facilitates users to submit case filing application materials, evidence materials, application materials, sign for document delivery, communicate with judges, etc. through mobile phones, computers and other terminal devices. But at the same time, with the blowout of online cases, judges are also facing problems such as changes in working methods and increased pressure on business processing. Due to the cumbersome online review operations, the lack of intelligent auxiliary means, and the judges' overly busy work tasks, it is often easy to cause a backlog of cases, resulting in cases exceeding the processing time limit, which is not convenient for judges to review.
[0036] To solve the above problems, the embodiment of the present application provides a data processing method, device, equipment and storage medium based on artificial intelligence. The method obtains the behavior data of the judge's review of legal documents, generates sample data based on the behavior data, and stores the sample data in the storage area corresponding to the judge's identity in the distributed file system. Then, when the number of sample data in the storage area reaches the preset number threshold, the preset machine learning model is iteratively trained according to all the sample data in the storage area to obtain the file review time estimation model corresponding to the judge's identity, and the estimated review time of the legal document to be reviewed is determined according to the file review time estimation model and the file feature information of the legal document to be reviewed corresponding to the judge's identity. Finally, according to the work schedule corresponding to the judge's identity and the estimated review time, the target review time period of the legal document to be reviewed is determined, and the review reminder time of the legal document to be reviewed is determined according to the target review time period, and when the current system time reaches the review reminder time, the review reminder information is output, which can intelligently assist the judge in reviewing the legal document and improve the convenience and efficiency of the judge in reviewing the legal document.
[0037] Among them, the embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.
[0038] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.
[0039] The embodiments of the present application provide an artificial intelligence-based data processing method that can be applied to a terminal device or a server. The terminal device can be a mobile phone, a tablet computer, a laptop computer, a personal computer, etc. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), as well as big data and artificial intelligence platforms.
[0040] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.
[0041] See also Figure 1 , Figure 1 Schematic diagram of a scenario for implementing the artificial intelligence-based data processing method provided in the embodiment of the present application. Figure 1 As shown, the scenario includes a terminal device 100, a server 200 and a distributed file system 300. The terminal device 100 is connected to the server 200 in communication, and the server 200 is connected to the distributed file system 300 in communication. The terminal device 100 may include a mobile phone, a tablet computer, a laptop computer, a desktop computer, etc., and the distributed file system 300 may include: GlusterFS file system, CephFS file system, HDFS (Hadoop Distributed File System), MooseFS file system.
[0042] Among them, the terminal device 100 collects the behavior data of the judge reviewing the legal documents and sends the behavior data to the server 200; the server 200 obtains the behavior data of the judge reviewing the legal documents sent by the terminal device 100, and generates sample data based on the behavior data, obtains the judge's identity from the behavior data, and sends the sample data to the distributed file system 300; the distributed file system 300 stores the sample data in the storage area corresponding to the judge's identity.
[0043] When the number of sample data is greater than or equal to a preset number threshold, the server 200 iteratively trains the preset machine learning model based on all the sample data in the storage area to obtain a file review time estimation model corresponding to the judge's identity identification; the server 200 determines the estimated review time value of the legal document to be reviewed based on the file review time estimation model and the file feature information of the legal document to be reviewed corresponding to the judge's identity identification; the server 200 obtains the work schedule corresponding to the judge's identity identification, and determines the target review time period for the legal document to be reviewed based on the work schedule and the estimated review time value; determines the review reminder time of the legal document to be reviewed based on the target review time period, and outputs the review reminder information when the current system time reaches the review reminder time.
[0044] See also Figure 2 , Figure 2 It is a flowchart of an artificial intelligence-based data processing method provided in an embodiment of the present application.
[0045] like Figure 2 As shown, the data processing method includes steps S101 to S106.
[0046] Step S101: Obtain behavioral data of judges reviewing legal documents, and generate sample data based on the behavioral data.
[0047] For example, legal documents may include indictment, counterclaim, reply, appeal, application for compulsory execution, application for property preservation, application for preliminary execution, application for recusation, application for declaration of disappearance, application for declaration of death, application for payment order, appeal for retrial, application for administrative reconsideration, administrative litigation indictment, administrative litigation reply, application for administrative compulsory execution, administrative appeal, administrative retrial appeal, divorce agreement, complaint, criminal private prosecution, application for bail pending trial, evidence documents, etc. Behavior data includes the file type of legal documents, review time (the length of time the judge takes to review the legal documents), and file feature information.
[0048] Exemplarily, the method of generating sample data based on behavior data may be: obtaining the review time and file feature information of the reviewed legal document from the behavior data; and using the review time and file feature information of the reviewed legal document as a piece of sample data. The file feature information includes the case type, case cause information, file type, number of attachments, and file size of the legal document.
[0049] Step S102: Obtain the judge's identity from the behavior data, and store the sample data in a storage area corresponding to the judge's identity in the distributed file system.
[0050] Among them, when a judge registers an account, the server will allocate a storage area to each judge account in the distributed file system, and bind the allocated storage area to the judge's identity identification. The judge's identity identification can be the judge's ID number, the judge's mobile phone number, or a randomly generated multi-digit string including numbers and / or letters. This embodiment does not make specific limitations on this.
[0051] For example, in a distributed file system, the storage areas allocated to Judge A and Judge B are storage area A and storage area B, respectively. When the behavior data of Judge A reviewing legal documents is obtained, the review time and file feature information of the reviewed legal documents of Judge A are obtained from the behavior data, and the review time and file feature information are stored as a sample data in storage area A. When the behavior data of Judge B reviewing legal documents is obtained, the review time and file feature information of the reviewed legal documents of Judge B are obtained from the behavior data, and the review time and file feature information are stored as a sample data in storage area B.
[0052] Step S103: When the number of sample data is greater than or equal to the preset number threshold, the preset machine learning model is iteratively trained based on all the sample data in the storage area to obtain a document review time estimation model corresponding to the judge's identity.
[0053] For example, when the number of sample data in storage area A reaches 1000, the preset machine learning model is iteratively trained based on all the sample data in storage area A to obtain a model for estimating the time required for document review by Judge A. For another example, when the number of sample data in storage area B reaches 1000, the preset machine learning model is iteratively trained based on all the sample data in storage area B to obtain a model for estimating the time required for document review by Judge B. The preset machine learning model may include a random forest tree model, a decision tree model, a neural network model, etc.
[0054] In one embodiment, when the number of sample data in multiple storage areas is greater than or equal to a preset number threshold, the preset neural network models can be iteratively trained in parallel based on all the sample data in each storage area to obtain the corresponding file review time estimation model. For example, when the number of sample data in storage area A and storage area B reaches 1000, the preset neural network model is iteratively trained based on all the sample data in storage area A to obtain the file review time estimation model of Judge A, and at the same time, the preset neural network model is iteratively trained based on all the sample data in storage area B to obtain the file review time estimation model of Judge B.
[0055] Exemplarily, when the number of sample data is greater than or equal to a preset number threshold, a piece of sample data is selected from the storage area as the target sample data; the file feature information in the target sample data is input into the preset neural network model to obtain the predicted audit time; the preset loss function is used to calculate the loss value of the predicted audit time and the actual audit time in the target sample data to obtain the model loss value; when the model loss value is greater than or equal to the preset loss value, the parameters of the neural network model are updated; and the step of selecting a piece of training sample data from the storage area as the target sample data is executed until the model loss value is less than the preset loss value. Among them, the preset number threshold, the preset loss function, and the preset loss value can be set based on the actual situation, and this embodiment does not make specific limitations on this. The preset loss function can be a mean square error function, and the preset number threshold is 1000.
[0056] Step S104: Determine an estimated value of the review time for the legal document to be reviewed based on the document review time estimation model and the document feature information of the legal document to be reviewed corresponding to the judge's identity identifier.
[0057] Exemplarily, the legal documents to be reviewed corresponding to the judge's identity are obtained at preset intervals, and the file feature information of the legal documents to be reviewed is obtained. The preset time can be set based on actual conditions, and this embodiment does not specifically limit this. For example, the preset time is 5 minutes.
[0058] Exemplarily, the review mark of the legal document corresponding to the judge's identity mark is obtained, wherein the review mark includes a first mark and / or a second mark, the first mark indicates that the legal document is in an unreviewed state, and the second mark indicates that the legal document is in a reviewed state; the legal document with the review mark of the first mark is determined as the legal document to be reviewed corresponding to the judge's identity mark. Exemplarily, the remaining processing time of the legal document corresponding to the judge's identity mark is obtained; the legal document with the shortest remaining processing time is determined as the legal document to be reviewed. The remaining processing time refers to the time between the processing deadline of the legal document and the current system time.
[0059] The file characteristic information of the legal document to be reviewed includes the case type, case information, file type, number of attachments and file size of the legal document to be reviewed. The file characteristic information of the legal document to be reviewed is input into the review time estimation model to obtain the review time estimation value of the legal document to be reviewed.
[0060] Step S105: Obtain a work schedule corresponding to the judge's identity identifier, and determine a target review time period for the legal document to be reviewed based on the work schedule and the estimated review time.
[0061] The server stores the judge's work schedule, which records the judge's work arrangement information, including the processing time period of each pending matter, such as the time period for court sessions, meeting times, and deliberation times.
[0062] In one embodiment, see Figure 3 , step S105 includes: sub-step S1051 to sub-step S1052.
[0063] Sub-step S1051, determining at least one idle time period according to the work schedule corresponding to the judge's identity.
[0064] For example, the work schedule records information including court hearing from 9:00-11:00, meeting from 11:20-12:00, lunch from 12:00-12:40, meeting from 13:30-15:50, meeting from 16:10-18:20, and leaving get off work at 19:00. Therefore, the free time periods include 12:40-13:30, 15:50-16:10, and 18:20-18:50.
[0065] Sub-step S1052: determining a target review time period for the legal document to be reviewed from at least one idle time period according to the estimated review time period for the legal document to be reviewed.
[0066] Exemplarily, the duration of each idle time period is determined, and the idle time period whose duration is greater than or equal to the estimated review time is determined as the candidate time period; the file type and case type of the legal document to be reviewed are determined; according to the file type and case type, the historical review time period of the legal document to be reviewed is determined; and the candidate time period that matches the historical review time period is determined as the target review time period of the legal document to be reviewed. When the candidate review time period partially overlaps with the historical review time period, the candidate review time period is determined to match the historical review time period.
[0067] Exemplarily, the method of determining the historical review time period of the legal document to be reviewed based on the file type and case type can be: according to the judge's identity, obtain a mapping relationship table between the file type, case type and review time period; select the review time period corresponding to the file type and case type from the mapping relationship table as the historical review time period of the legal document to be reviewed. The mapping relationship table between the file type, case type and review time period is generated based on the behavior data corresponding to the judge's identity.
[0068] For example, the estimated time for reviewing the legal documents to be reviewed is 10 minutes, and the judges' idle time periods of 12:40-13:30, 15:50-16:05 and 18:30-18:50 are 50 minutes, 15 minutes and 20 minutes respectively. Therefore, the idle time periods of 12:40-13:30, 15:50-16:05 or 18:30-18:50 can be determined as the target review time periods for the legal documents to be reviewed.
[0069] For another example, the estimated time for reviewing a legal document to be reviewed is 10 minutes, and the idle time periods of 12:40-13:30, 15:50-16:05 or 18:30-18:50 are determined as candidate review time periods for the legal document to be reviewed, and the historical review time period for the legal document to be reviewed is 15:55-16:05. Through comparison, it is found that the candidate review time period 15:50-16:05 matches the historical review time period 15:55-16:05. Therefore, the idle time period 15:50-16:05 is determined as the target review time period for the legal document to be reviewed.
[0070] Step S106: Determine the audit reminder time of the legal document to be audited according to the target audit time period, and output the audit reminder information when the current system time reaches the audit reminder time.
[0071] Exemplarily, the starting endpoint moment of the target review time period is obtained, and the starting endpoint moment is determined as the review reminder moment of the legal document to be reviewed. Alternatively, the starting endpoint moment of the target review time period is obtained, and the preset duration is subtracted from the starting endpoint moment to obtain the review reminder moment of the legal document to be reviewed. Alternatively, the starting endpoint moment of the target review time period is obtained, and the preset duration is added to the starting endpoint moment to obtain the review reminder moment of the legal document to be reviewed. For example, if the target review time period is 15:50-16:05, the starting endpoint moment 15:50 can be determined as the review reminder moment of the legal document to be reviewed. For another example, if the target review time period is 15:50-16:05 and the preset duration is 1 minute, the moment 15:49 or 15:51 can be determined as the review reminder moment of the legal document to be reviewed.
[0072] Exemplary methods for outputting review reminder information may be: determining the judge account to which the legal document to be reviewed belongs, and when the current system time reaches the review reminder time, pushing the review reminder information of the legal document to be reviewed to the judge account, or sending the review reminder information of the legal document to be reviewed to the mobile phone number bound to the judge account via SMS, or initiating a call request to the mobile phone number bound to the judge account, and sending the review reminder information after the call is established.
[0073] The data processing method provided in the above embodiment obtains the behavioral data of the judge's review of legal documents, generates sample data based on the behavioral data, and stores the sample data in the storage area corresponding to the judge's identity in the distributed file system. Then, when the number of sample data in the storage area reaches a preset number threshold, the preset machine learning model is iteratively trained according to all the sample data in the storage area to obtain a file review time estimation model corresponding to the judge's identity, and the estimated review time of the legal document to be reviewed is determined according to the file review time estimation model and the file feature information of the legal document to be reviewed corresponding to the judge's identity. Finally, according to the work schedule corresponding to the judge's identity and the estimated review time, the target review time period of the legal document to be reviewed is determined, and the review reminder time of the legal document to be reviewed is determined according to the target review time period, and when the current system time reaches the review reminder time, the review reminder information is output, which can intelligently assist the judge in reviewing the legal documents and improve the convenience and efficiency of the judge in reviewing the legal documents.
[0074] See also Figure 4 , Figure 4 It is a schematic block diagram of a data processing device provided in an embodiment of the present application.
[0075] like Figure 4 As shown, the data processing device 400 includes:
[0076] An acquisition module 410 is used to acquire the behavior data of the judge reviewing the legal document;
[0077] A sample generating module 420, configured to generate sample data according to the behavior data;
[0078] A sending module 430 is used to obtain the judge's identity from the behavior data, and store the sample data in a storage area corresponding to the judge's identity in a distributed file system;
[0079] The model training module 440 is used to iteratively train the preset machine learning model according to all the sample data in the storage area when the number of the sample data is greater than or equal to the preset number threshold, so as to obtain a document review time estimation model corresponding to the judge identity;
[0080] A determination module 450 is used to determine an estimated time for reviewing the legal document to be reviewed based on the document review time estimation model and the document feature information of the legal document to be reviewed corresponding to the judge's identity identifier;
[0081] The determination module 450 is further used to obtain a work schedule corresponding to the judge's identity identifier, and determine a target review time period for the legal document to be reviewed based on the work schedule and the review time estimate;
[0082] The determination module 450 is further used to determine the review reminder time of the to-be-reviewed legal document according to the target review time period;
[0083] The output module 460 is used to output the audit reminder information when the current system time reaches the audit reminder time.
[0084] In one embodiment, the sample generation module 420 is further configured to:
[0085] Obtaining review time and document feature information of the reviewed legal documents from the behavioral data;
[0086] The review time and document feature information of the reviewed legal document are taken as a piece of sample data.
[0087] In one embodiment, the model training module 440 is further used to:
[0088] Selecting a piece of sample data from the storage area as target sample data;
[0089] Inputting the file feature information in the target sample data into a preset machine learning model to obtain a predicted review time;
[0090] Using a preset loss function, a loss value is calculated for the predicted audit time and the actual audit time in the target sample data to obtain a model loss value;
[0091] When the model loss value is greater than or equal to a preset loss value, updating the parameters of the neural network model;
[0092] The step of selecting a piece of training sample data from the storage area as target sample data is executed until the model loss value is less than a preset loss value.
[0093] In one embodiment, if Figure 5 As shown, the determination module 450 includes:
[0094] A first determining submodule 451 is used to determine at least one idle time period according to the work schedule corresponding to the judge's identity;
[0095] The second determining submodule 452 is configured to determine a target review time period for the legal document to be reviewed from at least one of the idle time periods according to the estimated review time for the legal document to be reviewed.
[0096] In one embodiment, the second determining submodule 452 is further configured to:
[0097] Determine the duration of each idle time period, and determine the idle time period whose duration is greater than or equal to the estimated audit time as a candidate time period;
[0098] Determine the document type and case type of the legal document to be reviewed;
[0099] Determine the historical review period of the legal document to be reviewed based on the document type and case type;
[0100] The candidate time period that matches the historical review time period is determined as the target review time period for the legal document to be reviewed.
[0101] In one embodiment, the second determining submodule 452 is further configured to:
[0102] According to the judge's identity identifier, a mapping relationship table between the file type, the case type and the review time period is obtained;
[0103] A review time period corresponding to the file type and the case type is selected from the mapping relationship table as the historical review time period of the legal document to be reviewed.
[0104] In one embodiment, the determining module 450 is further configured to:
[0105] Obtain the starting endpoint time of the target audit time period;
[0106] Subtract the preset time from the starting endpoint time to obtain the review reminder time of the legal document to be reviewed.
[0107] It should be noted that technicians in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described devices and modules and units can refer to the corresponding processes in the aforementioned data processing method embodiments, and will not be repeated here.
[0108] The apparatus provided in the above embodiment may be implemented in the form of a computer program. The computer program may be Figure 6 Run on the server shown.
[0109] See also Figure 6 , Figure 6 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server or a terminal device.
[0110] like Figure 6 As shown, the computer device includes a processor, a memory, and a network interface connected via a system bus, wherein the memory may include a storage medium and an internal memory.
[0111] The storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any data processing method based on artificial intelligence.
[0112] The processor is used to provide computing and control capabilities to support the operation of the entire server.
[0113] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0114] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0115] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:
[0116] Obtaining behavioral data of judges reviewing legal documents, and generating sample data based on the behavioral data;
[0117] Obtaining a judge's identity from the behavior data, and storing the sample data in a storage area corresponding to the judge's identity in a distributed file system;
[0118] When the number of the sample data is greater than or equal to a preset number threshold, iteratively training the preset machine learning model based on all the sample data in the storage area to obtain a document review time estimation model corresponding to the judge identity identifier;
[0119] Determine an estimated time value for reviewing the legal document to be reviewed based on the document review time estimation model and the document feature information of the legal document to be reviewed corresponding to the judge's identity identifier;
[0120] Obtaining a work schedule corresponding to the judge's identity identifier, and determining a target review time period for the legal document to be reviewed based on the work schedule and the review time estimate;
[0121] The audit reminder time of the to-be-audited legal document is determined according to the target audit time period, and the audit reminder information is output when the current system time reaches the audit reminder time.
[0122] In one embodiment, when generating sample data according to the behavior data, the processor is used to implement:
[0123] Obtaining review time and document feature information of the reviewed legal documents from the behavioral data;
[0124] The review time and document feature information of the reviewed legal document are taken as a piece of sample data.
[0125] In one embodiment, when the processor implements iterative training of a preset machine learning model based on all sample data in the storage area, it is used to implement:
[0126] Selecting a piece of sample data from the storage area as target sample data;
[0127] Inputting the file feature information in the target sample data into a preset machine learning model to obtain a predicted review time;
[0128] Using a preset loss function, a loss value is calculated for the predicted audit time and the actual audit time in the target sample data to obtain a model loss value;
[0129] When the model loss value is greater than or equal to a preset loss value, updating the parameters of the neural network model;
[0130] The step of selecting a piece of training sample data from the storage area as target sample data is executed until the model loss value is less than a preset loss value.
[0131] In one embodiment, when determining the target review time period of the to-be-reviewed legal document according to the work schedule and the estimated review time, the processor is configured to implement:
[0132] Determine at least one idle time period according to the work schedule corresponding to the judge's identity identifier;
[0133] According to the estimated time required for reviewing the legal document to be reviewed, a target review time period for the legal document to be reviewed is determined from at least one of the idle time periods.
[0134] In one embodiment, when the processor determines the target review time period of the legal document to be reviewed from at least one of the idle time periods according to the estimated review time of the legal document to be reviewed, it is configured to:
[0135] Determine the duration of each idle time period, and determine the idle time period whose duration is greater than or equal to the estimated audit time as a candidate time period;
[0136] Determine the document type and case type of the legal document to be reviewed;
[0137] Determine the historical review period of the legal document to be reviewed based on the document type and case type;
[0138] The candidate time period that matches the historical review time period is determined as the target review time period for the legal document to be reviewed.
[0139] In one embodiment, when determining the historical review time period of the to-be-reviewed legal document according to the document type and the case type, the processor is configured to implement:
[0140] According to the judge's identity identifier, a mapping relationship table between the file type, the case type and the review time period is obtained;
[0141] A review time period corresponding to the file type and the case type is selected from the mapping relationship table as the historical review time period of the legal document to be reviewed.
[0142] In one embodiment, when determining the review reminder time of the to-be-reviewed legal document according to the target review time period, the processor is used to implement:
[0143] Obtain the starting endpoint time of the target audit time period;
[0144] Subtract the preset time from the starting endpoint time to obtain the review reminder time of the legal document to be reviewed.
[0145] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the computer device described above can refer to the corresponding process in the aforementioned data processing method embodiment, and will not be repeated here.
[0146] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.
[0147] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. The computer program includes program instructions. The method implemented when the program instructions are executed can refer to the various embodiments of the data processing method of the present application.
[0148] The computer-readable storage medium may be volatile or non-volatile. The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.
[0149] Furthermore, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc.
[0150] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.
[0151] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.
[0152] It should also be understood that the term "and / or" used in the specification of this application and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.
[0153] The serial numbers of the embodiments of the present application are for description only and do not represent the advantages and disadvantages of the embodiments. The above description is only a specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.
[0154] It should be noted that in each specific implementation method of the present application, when it comes to the need to process data related to the user's identity or characteristics, such as the judge's identity identification, the judge's behavioral data on reviewing legal documents, legal documents, legal documents to be reviewed, and reviewed legal documents, the user's permission or consent will be obtained first, and the collection, use, and processing of such data will comply with relevant laws, regulations, and standards.
Claims
1. A data processing method based on artificial intelligence, characterized in that: include: Obtaining behavioral data of a judge reviewing a legal document, and generating sample data based on the behavioral data, wherein the sample data includes review time and document feature information; Obtaining a judge's identity from the behavior data, and storing the sample data in a storage area corresponding to the judge's identity in a distributed file system; When the number of the sample data is greater than or equal to a preset number threshold, iteratively training the preset machine learning model based on all the sample data in the storage area to obtain a document review time estimation model corresponding to the judge identity identifier; Determine an estimated time value for reviewing the legal document to be reviewed based on the document review time estimation model and the document feature information of the legal document to be reviewed corresponding to the judge's identity identifier; Obtaining a work schedule corresponding to the judge's identity identifier, and determining a target review time period for the legal document to be reviewed based on the work schedule and the review time estimate; Determine the audit reminder time of the to-be-audited legal document according to the target audit time period, and output the audit reminder information when the current system time reaches the audit reminder time; The iterative training of the preset machine learning model according to all the sample data in the storage area includes: Selecting a piece of sample data from the storage area as target sample data; Inputting the file feature information in the target sample data into a preset machine learning model to obtain a predicted review time; Using a preset loss function, a loss value is calculated for the predicted audit time and the actual audit time in the target sample data to obtain a model loss value; When the model loss value is greater than or equal to a preset loss value, updating the parameters of the machine learning model; Execute the step of selecting a piece of training sample data from the storage area as target sample data until the model loss value is less than a preset loss value; Determining the target review time period for the legal document to be reviewed based on the work schedule and the estimated review time includes: Determine at least one idle time period according to the work schedule corresponding to the judge's identity identifier; Determine the duration of each idle time period, and determine the idle time period whose duration is greater than or equal to the estimated audit time as a candidate time period; Determine the document type and case type of the legal document to be reviewed; Determine the historical review period of the legal document to be reviewed based on the document type and case type; The candidate time period that matches the historical review time period is determined as the target review time period for the legal document to be reviewed.
2. The data processing method according to claim 1, characterized in that: The generating sample data according to the behavior data comprises: Obtaining review time and document feature information of the reviewed legal documents from the behavioral data; The review time and document feature information of the reviewed legal document are taken as a piece of sample data.
3. The data processing method according to claim 1, characterized in that: Determining the historical review time period of the legal document to be reviewed based on the document type and case type includes: According to the judge's identity identifier, a mapping relationship table between the file type, the case type and the review time period is obtained; A review time period corresponding to the file type and the case type is selected from the mapping relationship table as the historical review time period of the legal document to be reviewed.
4. The data processing method according to any one of claims 1 to 3, characterized in that: The step of determining the audit reminder time of the to-be-audited legal document according to the target audit time period includes: Obtain the starting endpoint time of the target audit time period; Subtract the preset time from the starting endpoint time to obtain the review reminder time of the legal document to be reviewed.
5. A data processing device, characterized in that: The data processing device comprises: An acquisition module is used to obtain the behavior data of judges reviewing legal documents; A sample generation module, used to generate sample data according to the behavior data, wherein the sample data includes audit time and file feature information; A sending module, used for obtaining the judge's identity from the behavior data, and storing the sample data in a storage area corresponding to the judge's identity in a distributed file system; A model training module, used for iteratively training a preset machine learning model based on all sample data in the storage area when the number of sample data is greater than or equal to a preset number threshold, to obtain a document review time estimation model corresponding to the judge identity; A determination module, used to determine an estimated time for reviewing the legal document to be reviewed based on the document review time estimation model and the document feature information of the legal document to be reviewed corresponding to the judge's identity identifier; The determination module is further used to obtain a work schedule corresponding to the judge's identity identifier, and determine a target review time period for the legal document to be reviewed based on the work schedule and the review time estimate; The determination module is further used to determine the review reminder time of the to-be-reviewed legal document according to the target review time period; An output module, used for outputting audit reminder information when the current system time reaches the audit reminder time; The model training module is further used to select a piece of sample data from the storage area as target sample data; input the file feature information in the target sample data into a preset machine learning model to obtain a predicted audit time; use a preset loss function to calculate the loss value of the predicted audit time and the actual audit time in the target sample data to obtain a model loss value; when the model loss value is greater than or equal to the preset loss value, update the parameters of the machine learning model; and execute the step of selecting a piece of training sample data from the storage area as the target sample data until the model loss value is less than the preset loss value; The determination module includes: a first determination submodule, used to determine at least one idle time period according to the work schedule corresponding to the judge's identity identification; a second determination submodule, used to determine the duration of each idle time period, and determine the idle time period whose duration is greater than or equal to the estimated review time as a candidate time period; determine the file type and case type of the legal document to be reviewed; determine the historical review time period of the legal document to be reviewed based on the file type and case type; and determine the candidate time period matching the historical review time period as the target review time period for the legal document to be reviewed.
6. A computer device, characterized in that: The computer device includes a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program and implement the steps of the artificial intelligence-based data processing method as described in any one of claims 1 to 4 when executing the computer program.
7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the artificial intelligence-based data processing method according to any one of claims 1 to 4 are implemented.
Citation Information
Patent Citations
Electronic trial data processing method and device based on element formula
CN110175605A
Intelligent examination method and device for punishment-related case data
CN110489524A
Interpretable legal judgment prediction method and device, electronic equipment and storage medium
CN112269878A
AEO qualification intelligent rating management system and method based on intelligent analysis and recognition and storage medium
CN112418813A
Computer-guided Corporate Governance with Document Generation and Execution
US20150220649A1
