Case processing method and device, equipment, medium and product
Through the automated case handling method, the case classification model and workflow examples are used to solve the problem of inefficiency of manual handling methods in the insurance industry’s claims business, and more efficient and economical case handling is achieved.
Patent Information
- Application Number
- CN202411615522.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-12
- Publication Date
- 2025-02-18
AI Technical Summary
In the prior art, the case handling of claims business in the insurance industry relies on manual methods, resulting in high cost, time-consuming and low efficiency.
By obtaining case information, we judge whether the case is automatically processed, and input the case information to the converging case classification model, determine the processing category, execute preset workflow examples, and generate processing results.
It realizes automated handling of cases, reduces manual intervention, improves processing efficiency, and reduces costs.
Smart Images
Figure CN119477211A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of financial technology, and in particular to a case handling method, device, equipment, medium and product. Background Art
[0002] In the financial field, such as the insurance industry, the claims business it often handles involves a large number of Internet-related claims businesses with clear responsibilities, low amounts, dispersed risks and dispersed claims. The number of cases in this type of business is large and the payment time requirements are high.
[0003] At present, the above-mentioned business is generally handled manually. Manually handling each case that needs to be handled not only requires a lot of manpower costs, but also takes a long time and has low efficiency.
[0004] Therefore, the handling method of the above business needs to be further optimized. Summary of the invention
[0005] The present application provides a case handling method, device, equipment, medium and product to solve the problem that current case handling is based on manual processing and the processing method needs to be further optimized.
[0006] The first aspect of the present application provides a case handling method, comprising:
[0007] Obtaining case information of a target case to be processed; the target case is a case related to claims settlement;
[0008] Determining whether the target case is an automatically processed case according to the case information;
[0009] If it is determined that the target case is an automatically processed case, the case information is input into a case classification model that has been trained to convergence, so as to use the case classification model to determine the case processing category corresponding to the target case;
[0010] According to the case processing category and the case information, a corresponding preset workflow instance is determined and executed to generate a processing result for the target case.
[0011] Furthermore, in the above method, obtaining case information of the target case to be processed includes:
[0012] Receiving an import request sent by a user device or other claims-related electronic device; the import request includes case information of a target case;
[0013] Determine the case information of the target case in the import request; the case information includes at least one of the following: basic customer information, transaction flow, product type, product contract information, insurance information, and compensation information.
[0014] Further, in the method as described above, the case information at least includes: product type or compensation information;
[0015] The determining, based on the case information, whether the target case is an automatically processed case includes:
[0016] Determine whether the compensation amount in the compensation information is less than the preset compensation amount upper limit;
[0017] If it is determined that the compensation amount is less than the preset compensation amount upper limit, the target case is determined to be an automatically processed case;
[0018] or
[0019] Determine whether the product type meets the preset type range;
[0020] If it is determined that the product type meets the preset type range, the target case is determined to be an automatically processed case.
[0021] Furthermore, in the above method, determining and executing a corresponding preset workflow instance according to the case processing category and the case information to generate a processing result for the target case includes:
[0022] If the case handling category is a high-risk category, the workflow engine will be used to add the target case to the preset priority processing queue;
[0023] Determine a preset workflow instance corresponding to the case information, and use a workflow engine to execute the preset workflow instance to generate a processing result for the target case; the preset workflow instance includes a plurality of corresponding case processing flows;
[0024] If the case processing category is a low risk level category, a preset workflow instance corresponding to the case information is determined, and after the case processing in the preset priority processing queue is completed, the preset workflow instance is executed using the workflow engine to generate a processing result for the target case.
[0025] Furthermore, the method as described above further comprises:
[0026] If the case handling category is a high risk level, the target case is added to a preset message queue, and the target case is pushed to a corresponding user device based on the preset message queue so that the user can handle the target case.
[0027] Furthermore, the method as described above, before inputting the case information into the case classification model trained to convergence to determine the case handling category corresponding to the target case using the case classification model, further includes:
[0028] Acquire training samples; the training samples include historical case information of multiple historical cases and case handling categories corresponding to the historical cases;
[0029] Inputting the training samples into a preset case classification model to train the preset case classification model; the preset case classification model is constructed based on the ID3 algorithm in the decision tree model;
[0030] Determine whether the preset case classification model has converged;
[0031] If it is determined that the preset case classification model converges, the preset case classification model is determined as the case classification model trained to convergence.
[0032] Furthermore, the method as described above, before determining whether the target case is an automatically processed case according to the case information, further includes:
[0033] Perform data verification on the target case and generate data verification results;
[0034] If the data verification result is verification passed, the case information of the target case is saved in a preset case record database, and a unique identifier corresponding to the target case is generated.
[0035] A second aspect of the present application provides a case handling device, comprising:
[0036] An acquisition module, used to acquire case information of a target case to be processed; the target case is a case related to claims settlement;
[0037] A judgment module, used for judging whether the target case is an automatically processed case according to the case information;
[0038] A determination module, for inputting the case information into a case classification model trained to convergence if the target case is determined to be an automatically processed case, so as to determine the case processing category corresponding to the target case using the case classification model;
[0039] A generation module is used to determine and execute a corresponding preset workflow instance according to the case processing category and the case information, and generate a processing result for the target case.
[0040] Furthermore, in the above device, the acquisition module is specifically used for:
[0041] Receive an import request sent by a user device or other claims-related electronic device; the import request includes case information of a target case; determine the case information of the target case in the import request; the case information includes at least one of the following: basic customer information, transaction flow, product type, product contract information, risk information, and compensation information.
[0042] Further, in the above device, the case information at least includes: product type or compensation information;
[0043] The judgment module is specifically used for:
[0044] Determine whether the compensation amount in the compensation information is less than the preset compensation amount upper limit; If it is determined that the compensation amount is less than the preset compensation amount upper limit, determine the target case as an automatically processed case;
[0045] or
[0046] Determine whether the product type meets the preset type range; if it is determined that the product type meets the preset type range, determine the target case as an automatically processed case.
[0047] Furthermore, in the above device, the generating module is specifically used for:
[0048] If the case handling category is a high-risk level category, the workflow engine is used to add the target case to a preset priority processing queue; a preset workflow instance corresponding to the case information is determined, and the workflow engine is used to execute the preset workflow instance to generate a processing result for the target case; the preset workflow instance includes multiple corresponding case handling processes; if the case handling category is a low-risk level category, the preset workflow instance corresponding to the case information is determined, and after the case processing in the preset priority processing queue is completed, the workflow engine is used to execute the preset workflow instance to generate a processing result for the target case.
[0049] Furthermore, the device as described above further comprises:
[0050] The push module is used to add the target case to a preset message queue if the case handling category is a high risk level, and push the target case to a corresponding user device based on the preset message queue so that the user can handle the target case.
[0051] Furthermore, the device as described above further comprises:
[0052] A training module is used to obtain training samples; the training samples include historical case information of multiple historical cases and case handling categories corresponding to the historical cases; the training samples are input into a preset case classification model to train the preset case classification model; the preset case classification model is constructed based on the ID3 algorithm in the decision tree model; it is determined whether the preset case classification model converges; if it is determined that the preset case classification model converges, the preset case classification model is determined as a case classification model trained to convergence.
[0053] Furthermore, the device as described above further comprises:
[0054] The verification module is used to perform data verification processing on the target case and generate a data verification result; if the data verification result is verification passed, the case information of the target case is saved in a preset case record database, and a unique identifier corresponding to the target case is generated.
[0055] A third aspect of the present application provides an electronic device, including: a memory and a processor;
[0056] The memory stores computer-executable instructions;
[0057] The processor executes the computer-executable instructions stored in the memory to implement the case handling method as described in any one of the first aspects.
[0058] The fourth aspect of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the case handling method described in any one of the first aspects.
[0059] The fifth aspect of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the case handling method described in any one of the first aspects.
[0060] The present application provides a case processing method, device, equipment, medium and product, the method comprising: obtaining case information of a target case to be processed; the target case is a case related to claims; judging whether the target case is an automatically processed case according to the case information; if the target case is determined to be an automatically processed case, the case information is input into a case classification model trained to convergence, so as to use the case classification model to determine the case processing category corresponding to the target case; according to the case processing category and the case information, the corresponding preset workflow instance is determined and executed to generate the processing result of the target case. The case processing method of the present application judges whether the target case is an automatically processed case according to the case information, and if the target case is determined to be an automatically processed case, the case information is input into a case classification model trained to convergence to determine the corresponding case processing category. At the same time, the corresponding preset workflow instance is determined and executed according to the case processing category and the case information to generate the processing result of the target case. Thus, cases that can be automatically processed do not need to be processed manually, which is less time-consuming and more efficient. BRIEF DESCRIPTION OF THE DRAWINGS
[0061] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0062] Figure 1A scenario diagram for implementing the case handling method of the embodiment of the present application;
[0063] Figure 2 A flowchart of a case handling method provided in one embodiment of the present application;
[0064] Figure 3 A flowchart of a case handling method provided in another embodiment of the present application;
[0065] Figure 4 A schematic diagram of the structure of the case processing device provided for this application;
[0066] Figure 5 A schematic diagram of the structure of the electronic device provided in this application.
[0067] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION
[0068] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0069] In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of financial data or user data and other information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals. It should be noted that in the embodiments of this application, some existing solutions in the industry such as certain software, components, models, etc. may be mentioned, and they should be considered as exemplary. Their purpose is only to illustrate the feasibility of the implementation of the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.
[0070] It should be noted that the case handling method, device, equipment, medium and product disclosed herein can be used in the field of financial technology. It can also be used in any field other than the field of financial technology. The application field of the case handling method, device, equipment, medium and product disclosed herein is not limited.
[0071] The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.
[0072] In order to clearly understand the technical solution of the present application, the inventive concept of the present application is first introduced in detail. At present, in the financial field, such as the insurance industry, non-auto insurance claims business involves a large number of Internet-based claims businesses with clear responsibilities, low amounts, dispersed risks, and dispersed claims. The current processing flow is the traditional general manual processing flow, which distinguishes different processing links such as property damage investigation, case filing, liability determination, and recovery. At the same time, different positions are set up according to these links. The current manual processing method is time-consuming and is more suitable for cases with unclear responsibilities, large amounts, and complex cases.
[0073] Therefore, the handling method of the above business needs to be further optimized.
[0074] Therefore, in order to solve the problem that the business processing method in the prior art needs to be further optimized, the inventors have found in their research that for cases suitable for automatic processing, automatic processing of cases can be achieved by classifying and presetting corresponding workflow instances, thereby improving the efficiency of case processing.
[0075] Specifically, the case information of the target case to be processed is obtained. The target case is a case related to claims. According to the case information, it is determined whether the target case is an automatically processed case. If it is determined that the target case is an automatically processed case, the case information is input into the case classification model trained to convergence, so as to use the case classification model to determine the case processing category corresponding to the target case. According to the case processing category and the case information, the corresponding preset workflow instance is determined and executed to generate the processing result of the target case.
[0076] The case processing method of the present application determines whether the target case is an automatically processed case based on the case information. If the target case is determined to be an automatically processed case, the case information is input into the case classification model trained to convergence to determine the corresponding case processing category. At the same time, the corresponding preset workflow instance is determined and executed based on the case processing category and case information to generate the processing result of the target case. As a result, cases that can be automatically processed do not need to be processed manually, which is time-saving and more efficient.
[0077] Based on the above creative findings, the inventor proposed the technical solution of the present application.
[0078] The following is an introduction to the application scenarios of the case handling method provided in the embodiment of the present application. Figure 1As shown, 1 is a first electronic device, and 2 is a second electronic device. The first electronic device 1 can be a user terminal, an electronic device related to a claims settlement system, etc. The second electronic device 2 can be a user terminal, an Internet platform device, an insurance system device, etc.
[0079] Exemplarily, when the first electronic device 1 processes a case, it is specifically as follows:
[0080] ① The second electronic device 2 sends a case import request to the first electronic device 1.
[0081] The first electronic device 1 performs the following process:
[0082] ② Obtain case information of the target case to be processed.
[0083] ③ Determine whether the target case is an automatically processed case based on the case information.
[0084] ④ If the target case is determined to be an automatically processed case, the case information is input into the case classification model that has been trained to convergence, so as to use the case classification model to determine the case processing category corresponding to the target case.
[0085] ⑤ Determine and execute the corresponding preset workflow instance based on the case processing category and case information, and generate the processing results for the target case.
[0086] Optionally, the processing results may also be displayed to the user so that the user is aware of the case processing progress and corresponding results.
[0087] The embodiments of the present application are introduced below in conjunction with the drawings in the specification.
[0088] Figure 2 A flowchart of a case handling method provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the execution subject in this embodiment is a case processing device, which can be integrated into an electronic device, such as an electronic device related to a claims settlement system. The case processing method provided in this embodiment includes the following steps:
[0089] Step S101, obtaining case information of a target case to be processed. The target case is a case related to claims settlement.
[0090] In this embodiment, the target case is a claim-related case, such as non-auto insurance claim, auto insurance claim, etc. Case information includes but is not limited to: basic customer information, transaction flow, product type, product contract details, accident information, compensation information, etc. The data format of the case information can be JSON (JavaScript Object Notation), XML (full name: Extensible Markup Language, Chinese: Extensible Markup Language), etc.
[0091] In this embodiment, the acquisition method can be obtained from other relevant electronic devices, or it can be obtained based on a case import request sent by a user terminal or a claims-related electronic device.
[0092] Step S102: determine whether the target case is an automatically processed case based on the case information.
[0093] In this embodiment, the scope of automatically processed cases can be pre-set, so as to determine whether the case is to be automatically processed based on case information such as product type, compensation information, etc.
[0094] Optionally, in this embodiment, cases corresponding to clear responsibilities, low amounts, and dispersed risks can be preset as the scope of automatically processed cases.
[0095] Optionally, if it is determined that the case is not to be processed automatically, the target case can be transmitted to the staff's electronic device for manual processing.
[0096] Specifically, when it is determined that a task is to be assigned to a relevant person, the task can be assigned to the relevant person through a user interface or other means. After the relevant person completes the task, the result is fed back, and then the case status and related information are updated according to the feedback result.
[0097] Among them, the notification can be sent via email, text message, instant message, etc.
[0098] Optionally, a user interface is provided to allow the user to view the case processing progress and case status of the case in real time.
[0099] Step S103: If it is determined that the target case is an automatically processed case, the case information is input into a case classification model that has been trained to convergence, so as to use the case classification model to determine the case processing category corresponding to the target case.
[0100] In this embodiment, the case classification model can be generated in advance by training a decision tree model. The case processing category can be a priority processing category, a risk level category, a product category, etc., or a combination of the priority processing category, the risk level category, the product category, etc., which is not limited in this embodiment.
[0101] Step S104, determining and executing the corresponding preset workflow instance according to the case processing category and case information, and generating a processing result for the target case.
[0102] In this embodiment, a workflow instance is a complete process execution process composed of multiple link processes, which are transferred and executed through predefined sequences and logical relationships. The workflow instance is responsible for managing and controlling the data and information transmission between each link process, as well as the execution process and status of the entire workflow.
[0103] Optionally, the full-process processing corresponding to the preset workflow instance may include the following links: general process - (reporting - scheduling - case processing - claims verification - case closing), and case processing is a major node (investigation - filing - settlement).
[0104] Optionally, the workflow instance is executed by a preset workflow engine.
[0105] The case processing method of this embodiment obtains case information of a target case to be processed. The target case is a case related to claims. It is determined whether the target case is an automatically processed case based on the case information. If it is determined that the target case is an automatically processed case, the case information is input into a case classification model trained to convergence, so as to use the case classification model to determine the case processing category corresponding to the target case. According to the case processing category and the case information, the corresponding preset workflow instance is determined and executed to generate the processing result of the target case.
[0106] The case processing method of the present application determines whether the target case is an automatically processed case based on the case information. If the target case is determined to be an automatically processed case, the case information is input into the case classification model trained to convergence to determine the corresponding case processing category. At the same time, the corresponding preset workflow instance is determined and executed based on the case processing category and case information to generate the processing result of the target case. As a result, cases that can be automatically processed do not need to be processed manually, which is time-saving and more efficient.
[0107] Figure 3 A flowchart of a case handling method provided in another embodiment of the present application is shown in FIG. Figure 3 As shown, the case handling method provided in this embodiment is a further refinement of the case handling method provided in the previous embodiment of the present application. The case handling method provided in this embodiment includes the following steps.
[0108] Step S201, receiving an import request sent by a user device or other electronic device related to claims, wherein the import request includes case information of a target case.
[0109] In this embodiment, other claims-related electronic devices include financial-related equipment in the claims system, internal underwriting systems, other Internet platform equipment, etc.
[0110] Step S202, determine the case information of the target case in the import request. The case information includes at least one of the following: basic customer information, transaction flow, product type, product contract information, insurance information, and compensation information.
[0111] In this embodiment, the basic customer information may include basic information such as the customer's name and gender, the transaction flow is the transaction flow data of the case, the product type is the type of product corresponding to the target case, such as accident insurance, etc., the product contract information is the data of the pre-signed contract, the accident information is the relevant information at the time of the accident, such as the time of the accident, the accident record, etc., and the compensation information is the relevant information of the compensation, such as the amount of compensation, the time of compensation, etc.
[0112] Step S203: determine whether the target case is an automatically processed case based on the case information.
[0113] Optionally, in this embodiment, the case information includes at least: product type or compensation information.
[0114] S203 may be specifically as follows:
[0115] Determine whether the compensation amount in the compensation information is less than the preset compensation settlement amount upper limit.
[0116] If the determined compensation amount is less than the preset compensation settlement amount upper limit, the target case will be determined as an automatically processed case.
[0117] or
[0118] Determine whether the product type fits within a preset type range.
[0119] If it is determined that the product type meets the preset type range, the target case is determined to be an automatically processed case.
[0120] In this embodiment, the preset upper limit of the compensation amount can be set according to the actual application. When the compensation amount is less than the preset upper limit of the compensation amount, it means that the compensation amount of the target case is small and no manual processing is required.
[0121] The preset type range can be set according to actual application, and this embodiment does not limit this.
[0122] Optionally, in this embodiment, before S203, the following process is also included:
[0123] Perform data verification processing on the target case and generate data verification results.
[0124] If the data verification result is verification passed, the case information of the target case is saved in the preset case record database, and a unique identifier corresponding to the target case is generated.
[0125] In this embodiment, the validity of the target case can be determined by performing data verification on the target case, so as to avoid invalid cases from being processed in the subsequent process. If the verification fails, the target case can be rejected or directly not processed. If the data verification result is that the verification passes, the case information of the target case can be saved in the preset case record database to facilitate subsequent inquiries about the relevant information of the target case, and the corresponding unique identifier can be generated for the target case to facilitate subsequent inquiries about the processing progress of the target case.
[0126] Step S204: If the target case is determined to be an automatically processed case, the case information is input into a case classification model that has been trained to convergence, so as to use the case classification model to determine the case processing category corresponding to the target case.
[0127] Optionally, in this embodiment, before S204, a training process may also be included, which is as follows:
[0128] Obtain training samples. The training samples include historical case information of multiple historical cases and case handling categories corresponding to the historical cases.
[0129] The training samples are input into the preset case classification model to train the preset case classification model. The preset case classification model is constructed based on the ID3 algorithm in the decision tree model.
[0130] Determine whether the preset case classification model converges.
[0131] If it is determined that the preset case classification model converges, the preset case classification model is determined as the case classification model trained to convergence.
[0132] In this embodiment, the method for judging whether convergence occurs may be whether the classification accuracy of the preset case classification model can meet the preset accuracy requirement. If the preset accuracy requirement can be met, it can be determined that the preset case classification model converges.
[0133] When the ID3 algorithm in the decision tree model constructs a preset case classification model, it can be based on the following historical case information: the case's insurance policy, estimated loss amount, clause risk code, liability, compensation amount, compensation ratio, loss category, case status, payee, customer information and other related data. At the same time, after the construction and training are completed, it can be further iterated and optimized multiple times according to the actual application situation to optimize the decision-making ability of the model. Specifically, the machine learning model can use multi-dimensional data such as the characteristics of the claims product, transaction patterns, and customer behavior to conduct a preliminary risk assessment and liability prediction of the case to distinguish between high-risk and low-risk cases, and then give priority to high-risk cases in the subsequent process.
[0134] Step S205: If the case handling category is a high risk level category, the workflow engine is used to add the target case to a preset priority processing queue.
[0135] In this embodiment, the high-risk level category in the case handling category may be a high-priority case, and the low-risk case may be a low-priority case.
[0136] If the case handling category is a high-risk category, it can be processed first. Therefore, the workflow engine is used to add the target case to the preset priority processing queue. Cases in the preset priority processing queue are generally processed first.
[0137] Optionally, in this embodiment, cases with high risk levels may also be handed over to manual processing, as follows:
[0138] If the case handling category is a high risk level, the target case is added to a preset message queue, and the target case is pushed to a corresponding user device based on the preset message queue so that the user can handle the target case.
[0139] In this embodiment, by using a message queue to asynchronously process these tasks, the response speed and throughput of the system can be improved.
[0140] Step S206, determine the preset workflow instance corresponding to the case information, and use the workflow engine to execute the preset workflow instance to generate a processing result for the target case. The preset workflow instance includes multiple corresponding case processing processes.
[0141] Step S207, if the case processing category is a low risk level category, determine the preset workflow instance corresponding to the case information, and after the case processing in the preset priority processing queue is completed, use the workflow engine to execute the preset workflow instance to generate a processing result for the target case.
[0142] In order to further illustrate the case handling method of this embodiment in detail, it will be further explained below in combination with specific applications.
[0143] 1. Intelligent interface and case information standardization module:
[0144] Design a set of standardized interfaces that meet the requirements of claim characteristics and elements, which can directly connect to the client, customer complaint system, case record database, external third-party Internet platform of external regulatory agencies, internal underwriting system and financial system. Based on the information imported from the external system, automatically connect to the underwriting and financial systems to complete relevant information verification and information processing. Among them, the interface processing and case record database can be equipped with idempotent mechanisms and database locks to ensure data consistency and integrity.
[0145] This module is used to perform the following steps:
[0146] Step 11: Receive a request to import case information.
[0147] The case information includes but is not limited to basic customer information, transaction flow, product type, product contract details, insurance information, compensation information, etc. The data format of the case information can be JSON, XML, etc.
[0148] Step 12: Verify the validity of the case information.
[0149] It should be understood that the interface has corresponding data validation and error handling mechanisms. For example, when a user or an external system sends a request to import case information through the application, the backend service can receive and validate the case information.
[0150] When the case information is verified, execute step 13. Otherwise, all case information of the case is rejected according to the preset error handling mechanism.
[0151] Step 13: Save the case information into the case record database and assign it a unique case ID.
[0152] Step 14: Automatically import and standardize case information.
[0153] 2. Intelligent classification and risk assessment engine for claims cases:
[0154] Step 21: Determine whether the product is to go through a micro-process (i.e., the product that can be automatically processed in the above embodiment) based on the configured rules.
[0155] Among them, the configurable rules include: claim type, upper limit of claim amount, products and institutions allowed to be imported, premium payment mark, etc. The micro-process is a fully automated process (different from the traditional cumbersome offline manual processing process, it is a fast processing channel for low-payment, homogeneous cases).
[0156] If the product is to be processed through a small or micro process, perform the following step 22.
[0157] Step 22: Use machine learning algorithms to automatically classify products that are to go through the micro-process to obtain classification results. Use the ID3 algorithm in the decision tree model to build a model based on the case's insurance policy, estimated loss amount, clause and insurance code, liability, compensation amount, compensation ratio, loss category, case status, payee, customer information and other relevant data, and continue to train, adjust and optimize the model, and collect results and feed them back to the model to optimize the model's decision-making ability. This can achieve the classification of micro-products, distinguish risks and priorities, and then give priority to high-risk and high-priority cases in subsequent processes or transfer high-risk and high-priority cases to manual intervention. Low-risk and low-priority cases can be automatically reviewed and approved.
[0158] Specifically, the machine learning model can use multi-dimensional data such as the characteristics of the claims product, transaction patterns, and customer behavior to conduct a preliminary risk assessment and liability prediction for the case, so as to distinguish between high-risk and low-risk cases, and then give priority to high-risk cases in subsequent processes.
[0159] High-risk cases are high-priority cases, and low-risk cases are low-priority cases.
[0160] 3. Use workflow engine to process the entire case process:
[0161] Step 31: Introduce a workflow engine of a specified type, and when conditions are met, create and execute a workflow instance corresponding to the case.
[0162] The condition may be: receiving the classification result of the case to be processed - the case processing category.
[0163] The full-process processing corresponding to a certain type of workflow instance may include the following links: general process - (reporting - dispatching - case handling - claims settlement - case closing), and case handling is a major node (investigation - filing - settlement).
[0164] There is a corresponding link process for each link, which may involve: processing logic, conditions, tasks, etc.
[0165] Therefore, a workflow instance is a complete process execution process consisting of multiple link processes, which are transferred and executed through predefined sequences and logical relationships. The workflow instance is responsible for managing and controlling the data and information transmission between each link process, as well as the execution process and status of the entire workflow. At the same time, the workflow instance is also created based on the workflow model and executed according to the definition in the workflow model.
[0166] Specifically, for each designated link, this embodiment performs the following process design: when a case enters a designated link, the workflow engine automatically triggers the corresponding task according to the conditions corresponding to the designated link. Tasks may include: fully automated tasks and tasks that require human participation. The following two tasks are analyzed as follows:
[0167] (1) When the task is fully automated, the workflow engine executes the corresponding logic of the specified link and updates the case status and related information.
[0168] Optionally, (1.1) different types of workflow instances are established for cases with different classification results. Or, (1.2) different types of workflow instances are established for cases with different product types (i.e., there is a mapping relationship between product types and workflow instance types).
[0169] Optionally, different types of workflow instances may correspond to different full-process processing procedures, including: differences in any one or more of the link processes, etc.
[0170] For example, workflow instance 1 corresponding to cases of product type 1 and workflow instance 2 corresponding to cases of product type 2 both involve (case reporting - dispatching - case processing - claims settlement - case closing), but the case processing link corresponding to workflow instance 1 and the case processing link corresponding to workflow instance 2 have different configuration information in at least one of the processing logic, conditions, and tasks.
[0171] Specifically, the C4.5 algorithm in the decision tree model is used, and its information gain ratio is utilized to support the processing of continuous values. A workflow model is constructed based on the case's insurance policy, case status, case processing time, compensation amount, clause insurance type, cause of accident, reporting time, compensation approval time, case closing time, special agreements, investigation, data collection, and intelligent classification results. The workflow model determines which node the case should go to: reporting-dispatching-case processing-compensation-case closing, and prompts which tasks must be handled manually and which tasks can be automatically generated and processed according to the corresponding nodes.
[0172] Optionally, this embodiment may use a message queue to asynchronously process these tasks to improve the response speed and throughput of the system.
[0173] For tasks that require human participation, assign tasks to relevant personnel and send notifications.
[0174] Specifically, when it is determined to assign a task to a relevant person, the present embodiment can assign the task to the relevant person through a user interface or other means. After the relevant person completes the task, he / she will provide feedback on the result, and then the workflow engine will update the case status and related information according to the feedback result.
[0175] Among them, the notification can be sent via email, text message, instant message, etc.
[0176] Optionally, when the case status changes, the workflow engine can also send notifications to relevant personnel.
[0177] Optionally, after a workflow instance is established, a case ID can be associated with the workflow instance so that staff can track and manage the case during the entire process.
[0178] Optionally, a user interface is provided to allow the user to view the case processing progress and case status of the case in real time.
[0179] Specifically, the user can query case information from the case record database through the case identification or other conditions on the user interface and obtain detailed processing history records.
[0180] Optionally, the method of this embodiment may further include the following content:
[0181] The introduction of blockchain technology ensures the non-tamperability of case information, processing history records, etc. The system automatically collects data and integrates it into the corresponding chain of evidence, and automatically verifies the validity and relevance of the chain of evidence through smart contracts, thereby improving the accuracy and efficiency of case handling.
[0182] Establish a highly automated case processing process engine to automatically dispatch relevant departments to participate according to the nature of the case, such as compliance, customer service, etc., while providing real-time tracking and notification functions for case processing progress to ensure that each step of the case can be quickly advanced.
[0183] This embodiment can also perform data protection and privacy protection. For example, this embodiment can perform the following processing on case information, processing history records, etc. before storing them: such as data encryption, access control, audit logs, etc., to protect the security and integrity of case information.
[0184] Perform performance tests on the system to ensure stable operation under high concurrency and large data volumes. Use caching, load balancing and other technical means to improve system performance. Consider scalability when designing the system architecture so that new features can be easily added or processing capabilities can be expanded when needed.
[0185] The method of this embodiment creates a model through machine learning and continuously improves the decision-making ability of the model, combines with the workflow engine to realize automatic decision-making and automatic execution of case processing procedures, realizes stable high concurrency capabilities through multi-threading technology, locking mechanism and current limiting algorithm, and improves the overall processing capability of the system through asynchronous processing of time-consuming operations and timed retransmission mechanism.
[0186] At the same time, the method of this embodiment greatly improves the efficiency of case handling:
[0187] Through the automated processing technology of this embodiment, the processing time of claims cases can be reduced by more than 50% compared with traditional methods, greatly improving the case processing speed of products that require small and micro processes.
[0188] Enhance the accuracy and objectivity of decision-making:
[0189] Intelligent classification and risk assessment by artificial intelligence reduces the subjectivity of human judgment and ensures that each case is handled fairly and reasonably.
[0190] Optimize resource allocation:
[0191] Automated processes reduce the need for manual intervention, allowing staff to focus more on other aspects of case processing.
[0192] Figure 4A schematic diagram of the structure of the case processing device provided for this application, such as Figure 4 As shown, in this embodiment, the case handling device 300 includes:
[0193] The acquisition module 301 is used to acquire case information of a target case to be processed. The target case is a case related to claim settlement.
[0194] The judgment module 302 is used to judge whether the target case is an automatically processed case according to the case information.
[0195] The determination module 303 is used to input the case information into the case classification model that has been trained to convergence if the target case is determined to be an automatically processed case, so as to use the case classification model to determine the case processing category corresponding to the target case.
[0196] The generation module 304 is used to determine and execute the corresponding preset workflow instance according to the case processing category and case information, and generate the processing result of the target case.
[0197] The case processing device provided in this embodiment can execute Figure 2 The technical solution of the method embodiment shown in the figure has the same implementation principle and technical effect as Figure 2 The method embodiments shown are similar and will not be described in detail here.
[0198] The case handling device provided in the present application is further refined on the basis of the case handling device provided in the previous embodiment, and the case handling device 300 includes:
[0199] Optionally, in this embodiment, the acquisition module 301 is specifically used for:
[0200] Receive an import request sent by a user device or other electronic device related to claims. The import request includes case information of a target case. Determine the case information of the target case in the import request. The case information includes at least one of the following: basic customer information, transaction flow, product type, product contract information, insurance information, and compensation information.
[0201] Optionally, in this embodiment, the case information includes at least: product type or compensation information.
[0202] The determination module 302 is specifically used for:
[0203] Determine whether the compensation amount in the compensation information is less than the preset compensation amount upper limit. If it is determined that the compensation amount is less than the preset compensation amount upper limit, the target case is determined to be an automatically processed case.
[0204] or
[0205] Determine whether the product type meets the preset type range. If it is determined that the product type meets the preset type range, the target case is determined to be an automatically processed case.
[0206] Optionally, in this embodiment, the generating module 304 is specifically used for:
[0207] If the case handling category is a high-risk category, the workflow engine is used to add the target case to the preset priority processing queue. The preset workflow instance corresponding to the case information is determined, and the workflow engine is used to execute the preset workflow instance to generate the processing result of the target case. The preset workflow instance includes multiple corresponding case processing processes. If the case handling category is a low-risk category, the preset workflow instance corresponding to the case information is determined, and after the case processing in the preset priority processing queue is completed, the workflow engine is used to execute the preset workflow instance to generate the processing result of the target case.
[0208] Optionally, in this embodiment, the case handling device 300 further includes:
[0209] The push module is used to add the target case to the preset message queue if the case handling category is a high risk level, and push the target case to the corresponding user device based on the preset message queue so that the user can handle the target case.
[0210] Furthermore, in the above device, the case handling device 300 further includes:
[0211] A training module is used to obtain training samples. The training samples include historical case information of multiple historical cases and case handling categories corresponding to the historical cases. The training samples are input into a preset case classification model to train the preset case classification model. The preset case classification model is constructed based on the ID3 algorithm in the decision tree model. It is determined whether the preset case classification model converges. If it is determined that the preset case classification model converges, the preset case classification model is determined as a case classification model trained to convergence.
[0212] Furthermore, in the above device, the case handling device 300 further includes:
[0213] The verification module is used to perform data verification processing on the target case and generate a data verification result. If the data verification result is verification passed, the case information of the target case is saved in a preset case record database, and a unique identifier corresponding to the target case is generated.
[0214] The case processing device provided in this embodiment can execute Figure 2-Figure 3 The technical solution of the method embodiment shown in the figure has the same implementation principle and technical effect as Figure 2-Figure 3 The method embodiments shown are similar and will not be described in detail here.
[0215] According to an embodiment of the present application, the present application also provides an electronic device, a computer-readable storage medium and a computer program product.
[0216] like Figure 5 As shown, Figure 5 is a schematic diagram of the structure of the electronic device provided by the present application. The electronic device is intended for various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.
[0217] like Figure 5 As shown, the electronic device includes: a processor 401 and a memory 402. The various components are connected to each other using different buses and can be installed on a common mainboard or in other ways as needed. The processor can process instructions executed in the electronic device. The electronic device can also include a transceiver, which is used to send and receive data.
[0218] The memory 402 is a non-transitory computer-readable storage medium provided in the present application. The memory stores instructions executable by at least one processor to enable at least one processor to execute the case processing method provided in the present application. The non-transitory computer-readable storage medium of the present application stores computer instructions, which are used to enable a computer to execute the case processing method provided in the present application.
[0219] The memory 402 is a non-transient computer-readable storage medium that can be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the case handling method in the embodiment of the present application (for example, the attached Figure 4 The processor 401 executes various functional applications and data processing of the electronic device by running the non-transient software programs, instructions and modules stored in the memory 402, that is, implements the case processing method in the above method embodiment.
[0220] This embodiment also provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute the case handling method of the above embodiment.
[0221] At the same time, this embodiment also provides a computer product. When the instructions in the computer product are executed by the processor of an electronic device, the electronic device can execute the case handling method of the above embodiment.
[0222] Those skilled in the art will readily come up with other implementations of the embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modifications, uses or adaptations of the embodiments of the present application, which follow the general principles of the embodiments of the present application and include common knowledge or customary technical means in the art that are not disclosed in the embodiments of the present application.
[0223] It should be understood that the embodiments of the present application are not limited to the precise structures described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the embodiments of the present application is limited only by the appended claims.
Claims
1. A case handling method, characterized in that: include: Obtain case information of a target case to be processed; The target cases are claims-related cases; Determining whether the target case is an automatically processed case according to the case information; If it is determined that the target case is an automatically processed case, the case information is input into a case classification model that has been trained to convergence, so as to use the case classification model to determine the case processing category corresponding to the target case; According to the case processing category and the case information, a corresponding preset workflow instance is determined and executed to generate a processing result for the target case.
2. The method according to claim 1, characterized in that The step of obtaining case information of a target case to be processed includes: Receiving an import request sent by a user device or other claims-related electronic device; the import request includes case information of a target case; Determine the case information of the target case in the import request; the case information includes at least one of the following: basic customer information, transaction flow, product type, product contract information, insurance information, and compensation information.
3. The method according to claim 1, characterized in that: The case information at least includes: product type or compensation information; The determining, based on the case information, whether the target case is an automatically processed case includes: Determine whether the compensation amount in the compensation information is less than the preset compensation amount upper limit; If it is determined that the compensation amount is less than the preset compensation amount upper limit, the target case is determined to be an automatically processed case; or Determine whether the product type meets the preset type range; If it is determined that the product type meets the preset type range, the target case is determined to be an automatically processed case.
4. The method according to claim 1, characterized in that The determining and executing a corresponding preset workflow instance according to the case processing category and the case information to generate a processing result for the target case includes: If the case handling category is a high-risk category, the workflow engine will be used to add the target case to the preset priority processing queue; Determine a preset workflow instance corresponding to the case information, and use a workflow engine to execute the preset workflow instance to generate a processing result for the target case; the preset workflow instance includes a plurality of corresponding case processing flows; If the case processing category is a low risk level category, a preset workflow instance corresponding to the case information is determined, and after the case processing in the preset priority processing queue is completed, the preset workflow instance is executed using the workflow engine to generate a processing result for the target case.
5. The method according to claim 4, characterized in that The method further comprises: If the case handling category is a high risk level, the target case is added to a preset message queue, and the target case is pushed to a corresponding user device based on the preset message queue so that the user can handle the target case.
6. The method according to claim 1, characterized in that Before inputting the case information into a case classification model trained to convergence to determine the case handling category corresponding to the target case using the case classification model, the method further includes: Acquire training samples; the training samples include historical case information of multiple historical cases and case handling categories corresponding to the historical cases; Inputting the training samples into a preset case classification model to train the preset case classification model; the preset case classification model is constructed based on the ID3 algorithm in the decision tree model; Determine whether the preset case classification model has converged; If it is determined that the preset case classification model converges, the preset case classification model is determined as the case classification model trained to convergence.
7. The method according to any one of claims 1 to 6, characterized in that: Before judging whether the target case is an automatically processed case according to the case information, the method further includes: Perform data verification on the target case and generate data verification results; If the data verification result is verification passed, the case information of the target case is saved in a preset case record database, and a unique identifier corresponding to the target case is generated.
8. A case handling device, characterized in that: include: An acquisition module, used to acquire case information of a target case to be processed; The target cases are claims-related cases; A judgment module, used for judging whether the target case is an automatically processed case according to the case information; A determination module, for inputting the case information into a case classification model trained to convergence if the target case is determined to be an automatically processed case, so as to determine the case processing category corresponding to the target case using the case classification model; A generation module is used to determine and execute a corresponding preset workflow instance according to the case processing category and the case information, and generate a processing result for the target case.
9. An electronic device, characterized in that: include: Memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the case handling method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the case handling method according to any one of claims 1 to 7.
11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the case processing method as described in any one of claims 1 to 7 is implemented.