Claim settlement progress pushing method and device, computer equipment and storage medium

Abstract

The invention belongs to the technical field of artificial intelligence, and relates to a claim settlement progress pushing method, which comprises the steps of obtaining a processing progress node, progress state data and business knowledge data of a claim settlement case; determining whether the processing progress node is a key progress node or not according to the progress state data and the business knowledge data by adopting the trained claim settlement processing model; if the progress node is the key progress node, generating progress push content according to the processing progress node, the progress state data and the business knowledge data; and pushing the progress push content to a target customer of the claim settlement case. The invention further provides a claim settlement progress pushing device, computer equipment and a storage medium. The method can be applied to business management program systems of financial insurance and the like, can solve the problem of low business processing efficiency of an existing progress feedback mechanism of insurance claim settlement, and improves the customer experience.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology and is applied to online financial insurance processing business scenarios, and in particular to a method, device, computer equipment and storage medium for pushing claims progress. Background Art

[0002] In the insurance business, the progress feedback mechanism for insurance claims is plagued by numerous issues. During the claims process, customers generally lack a clear understanding of the progress of their claims, often needing to proactively contact the insurance company to inquire about the status of their claims. This passive method of obtaining information creates an inconvenient experience. Due to the highly specialized nature of insurance claims, involving complex rules and procedures, it can be difficult for customers without a professional insurance background to understand the details of the progress. This is especially true in claims involving multiple sources of evidence. The complexity of the scenario leads to delayed information transmission and a lack of transparency, further exacerbating customer anxiety.

[0003] Existing methods for providing feedback on claims progress typically rely on passive inquiries through human customer service or online platforms. Limited by staff numbers and workload, human customer service cannot respond to every customer's needs promptly and accurately. Online inquiry tools offered by many insurance companies often only display simple progress information, lacking detailed explanations of specific progress and next steps. This leads to customers lacking understanding of claim details, leading to unnecessary back-and-forth communication, which not only increases the burden on customers but also puts pressure on insurance company customer service.

[0004] Therefore, an innovative technical solution is urgently needed to address the shortcomings of the progress feedback mechanism for insurance claims and improve customer experience and business processing efficiency. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to propose a method, apparatus, computer equipment and storage medium for pushing claims progress, aiming to solve the problem of low business processing efficiency in the existing insurance claims progress feedback mechanism.

[0006] First, a method for pushing claims progress is provided, which adopts the following technical solution:

[0007] Obtain the processing progress nodes, progress status data and business knowledge data of claims cases;

[0008] Using the trained claims processing model, determining whether the processing progress node is a key progress node based on the progress status data and the business knowledge data;

[0009] If it is a key progress node, generating progress push content based on the key progress node, the progress status data and the business knowledge data;

[0010] The progress push content is pushed to the target customers of the claim case.

[0011] In the second aspect, a claim progress push device is provided, which adopts the following technical solution:

[0012] Data acquisition module, used to obtain the processing progress nodes, progress status data and business knowledge data of claims cases;

[0013] a node determination module, configured to use the trained claims processing model to determine whether the processing progress node is a key progress node based on the progress status data and the business knowledge data;

[0014] A content generation module is configured to generate progress push content based on the key progress node, the progress status data, and the business knowledge data if the progress is a key progress node;

[0015] The content push module is used to push the progress push content to the target customers of the claim case.

[0016] According to a third aspect, a computer device is provided, comprising a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the above-mentioned claim progress push method when executing the computer-readable instructions.

[0017] In a fourth aspect, a computer-readable storage medium is provided, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the steps of the above-mentioned claim progress push method are implemented.

[0018] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0019] The embodiment of the present application accurately locates the stage and related business processes of the claim case during the processing by obtaining the processing progress nodes, progress status data and business knowledge data of the claim case; by adopting the trained claim processing model, it determines whether the processing progress node is a key progress node based on the progress status data and business knowledge data, and can dynamically identify the key progress nodes that have a greater impact on customer experience and business efficiency; by generating progress push content based on key progress nodes, progress status data and business knowledge data, it can perform personalized push customization based on the specific circumstances of each case. The progress push content not only contains the current progress information of the claim case, but also explains and illustrates it in combination with business knowledge, making it easier for customers to understand the claim progress and reducing anxiety and misunderstandings caused by information opacity. By timely pushing the progress push content to the target customers of the claim case, customers can keep abreast of the latest developments of the claim case and enhance customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the solutions in this application, a brief introduction will be given below to the drawings required for use in the description of the embodiments of this application. Obviously, the drawings described below are some embodiments of this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0022] Figure 2 This is a flowchart of an embodiment of the method for pushing the claims progress of the present application;

[0023] Figure 3 This is a flowchart of an embodiment of step S202 of the present application;

[0024] Figure 4 This is a flowchart of an embodiment of step S203 of the present application;

[0025] Figure 5 This is a structural diagram of an embodiment of the claim progress pushing device of the present application;

[0026] Figure 6 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which this application belongs. The terms used in the specification of the application are for the purpose of describing specific embodiments only and are not intended to limit this application. The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0028] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0029] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0030] like Figure 1 As shown, system architecture 100 may include a terminal device 101, a network 102, and a server 103. Terminal device 101 may be a laptop computer 1011, a tablet computer 1012, or a mobile phone 1013. Network 102 is a medium for providing a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0031] The user can use the terminal device 101 to interact with the server 103 via the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0032] The terminal device 101 can be various electronic devices with a display screen and supporting web browsing. In addition to the laptop computer 1011, tablet computer 1012 or mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop computer and a desktop computer, etc.

[0033] The server 103 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal device 101 .

[0034] It should be noted that the claim progress push method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the claim progress push device is generally set in the server / terminal device.

[0035] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0036] Continue to refer Figure 2 , shows a flow chart of an embodiment of a method for pushing claims progress according to the present application. The method for pushing claims progress includes the following steps:

[0037] Step S201: Obtain the processing progress node, progress status data and business knowledge data of the claim case;

[0038] In this embodiment, the claim progress push method is executed on the electronic device (eg Figure 1 The server / terminal device shown in the figure can obtain the processing progress nodes, progress status data, and business knowledge data of the claim case through a wired connection or a wireless connection. It should be noted that the above-mentioned wireless connection methods may include but are not limited to 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other wireless connection methods currently known or to be developed in the future.

[0039] In this embodiment, a claim case's processing progress nodes refer to the stage division points in the insurance claims process, such as the material receipt node, review node, processing node, and payment node. Each node represents a key step in the claims process. Progress status data reflects the current status and updates of the claim case at each processing progress node, such as whether a node has been completed, the completion time, and the estimated completion time. Business knowledge data covers insurance product information, business process data, business rules data, business interpretation data, relevant laws and policies, and other knowledge content related to the claim case, providing a foundation for claims processing and progress notification.

[0040] Specifically, the processing progress nodes, progress status data, and business knowledge data of the claim case are obtained, wherein the processing progress nodes and progress status data of the claim case are extracted in real time from the claims management system, and the progress status data at least includes the case status and processing time; and the business knowledge data related to the claim case is obtained from the pre-built business knowledge base. The business knowledge data may include business process data, business rule data, and business explanation data, wherein the business process data records the complete process steps of the insurance claim from the beginning to the end, such as the sequence of material receipt, review, processing, payment, etc. and the connection relationship between each link; the business rule data contains various rules that the claim case needs to follow in the claims process, such as material requirements, review standards, payment conditions, time limits, etc. The business explanation data provides easy-to-understand explanations of complex business processes and rules to help customers understand the significance and requirements of each link in the claims process.

[0041] For example, in an insurance claims scenario, for a car insurance claim case, the business knowledge data related to the car insurance claim case is obtained, as well as real-time information on whether the repair invoice, accident photos, hospital certificates and other materials submitted by the customer have been received during the material receiving stage (i.e., the processing progress node), as well as the specific time of receipt (i.e., progress status data); information such as whether the review has been completed during the review stage, the result of whether the review has been passed or failed, the materials required to be supplemented, and the review time.

[0042] Step S202: using the trained claims processing model, determining whether the processing progress node is a key progress node based on the progress status data and the business knowledge data;

[0043] In this embodiment, a large amount of historical claims case data is used in advance to train a claims processing model. The model has the ability to analyze and process claims progress data and business knowledge data, and can determine key progress nodes.

[0044] Specifically, the acquired progress status data and business knowledge data are input into the trained claims processing model. The model then determines, based on the business knowledge data, whether the current progress status data contains critical status information, such as abnormal conditions, the need for supplementary materials, or a node with a high attention weight. If the progress status data contains critical status information, the processing progress node is identified as a critical progress node.

[0045] For example, at a material review node, if the review time exceeds the normal range specified in the business knowledge data, or if the review result is failure and the customer needs to provide a large amount of additional materials, then the node is determined to be a critical progress node.

[0046] Step S203: If it is a key progress node, generating progress push content according to the key progress node, the progress status data and the business knowledge data;

[0047] In this embodiment, when a processing progress node is determined to be a critical progress node, personalized progress push content is generated based on the critical progress node, progress status data, and business knowledge data. For example, business explanation data for the critical progress node is extracted from the business knowledge data, and the progress push content is generated by combining the business explanation data and progress status data. The push content can be generated using natural language generation technology to convert complex claims information into concise and easy-to-understand language.

[0048] For example, if the key progress node is the materials review stage, and the progress status data indicates that the materials review failed and that supplementary material A is required, business explanation data for this key progress node is extracted from the business knowledge data, such as "The materials review stage is the second stage in the auto insurance claims process, and the materials that need to be reviewed include A, B, and C." The progress push content generated by combining the business explanation data and progress status data may be: "Dear customer, your auto insurance claim case failed the materials review in the second stage - materials review due to the lack of material A. Please supplement the relevant materials as soon as possible so that we can continue to process your claim." In addition, the push content can also be combined with business explanation data to answer questions that customers may be concerned about, such as the submission method and review time for supplementary materials.

[0049] Step S204: Push the progress push content to the target customer of the claim case.

[0050] In this embodiment, the generated progress push content is pushed to the target customer of the claim case. For example, an appropriate push method is selected based on the target customer's preference information to convey the progress push content to the target customer. The push method may include a push template, a push channel, and a push time.

[0051] For example, if the customer's preference information includes that they like to receive email notifications through the APP between 10-12 o'clock, the generated progress push content will be pushed to the target customer of the claim case in the form of an email through the APP between 10-12 o'clock.

[0052] In this embodiment, by obtaining the processing progress nodes, progress status data and business knowledge data of the claims case, the stage and related business processes of the claims case in the processing process can be accurately located; by using the trained claims processing model, based on the progress status data and business knowledge data, it is determined whether the processing progress node is a key progress node, and the key progress nodes that have a greater impact on customer experience and business efficiency can be dynamically identified; by generating progress push content based on key progress nodes, progress status data and business knowledge data, personalized push customization can be performed based on the specific circumstances of each case. The progress push content not only contains the current progress information of the claims case, but also provides explanations and descriptions in combination with business knowledge, making it easier for customers to understand the claims progress and reducing anxiety and misunderstandings caused by information opacity. By timely pushing the progress push content to the target customers of the claims case, customers can keep abreast of the latest developments of the claims case and enhance customer satisfaction.

[0053] In some optional implementations of this embodiment, refer to Figure 3The above step S202, i.e., using the trained claims processing model to determine whether the processing progress node is a key progress node based on the progress status data and the business knowledge data, may include the following steps:

[0054] Step S2021: using the trained claims processing model, extracting business rule data related to the processing progress node from the business knowledge data;

[0055] In this embodiment, a claims processing model is pre-trained using a large amount of historical claims case data. This historical claims case data includes claims progress data and business knowledge data related to multiple historical claims processes. The claims progress data includes each processing progress node involved in the historical claims process and the corresponding progress status data; the business knowledge data includes at least the business process data, business rules data, and business interpretation data involved in the historical claims cases. The trained claims processing model is capable of analyzing and processing claims progress data and business knowledge data, and can identify key progress nodes.

[0056] Specifically, the acquired processing progress nodes, progress status data and business knowledge data are input into the trained claims processing model. The model uses the learned features and rules to extract node features from the processing progress nodes and business rule features from the business knowledge data. The extracted node features and business rule features are matched, and the matching business rule features are screened out as the business rule data related to the processing progress nodes extracted from the business knowledge data.

[0057] For example, when a car insurance claim reaches the "Document Review" stage, the claims processing model extracts rule features related to the auto insurance document review from the business knowledge data. These include the required types of review documents (such as repair invoices, accident liability determinations, and copies of driver's licenses and vehicle registrations), review time limits (required to complete the review within a few business days of receipt), and criteria for determining the authenticity and completeness of the documents. This extracted business rule data provides a basis for subsequent determination of anomalies at the processing progress node.

[0058] Step S2022: determining whether the processing progress node is in an abnormal state based on the progress status data and the business rule data;

[0059] Specifically, the claims processing model is used to compare and analyze the acquired progress status data with the extracted business rule data. The progress status data reflects the actual progress of the claim case at each processing progress node, such as whether the review has started at the material review node, whether the review is completed, and the review results. The business rule data contains the various rules that need to be followed in the claims process, especially at each processing progress node, such as the material type requirements, review time limits, and material authenticity and completeness judgment criteria stipulated at the material review node. The progress status data is compared with the business rule data one by one to determine whether there is an abnormal state at the processing progress node. An abnormal state refers to a situation where the progress status data does not match the business rule data.

[0060] For example, if the progress status data indicates that the review has exceeded the time limit specified in the business rules, but the review results are still pending, the model will determine that the node is in an abnormal state. Alternatively, if the progress status data indicates that the review materials are missing necessary information, such as an unclear amount on a repair invoice, preventing the review from proceeding, the model will determine that the node is in an abnormal state and requires further processing. Through this comparative analysis, the model can accurately identify whether there are problems with the processing progress node.

[0061] Step S2023: If an abnormal state exists, the processing progress node is determined as a key progress node.

[0062] Specifically, when the model identifies an abnormal state at a processing progress node, it designates that node as a critical progress node. For example, in a car insurance claim, an abnormality at the document review node can delay the claim settlement, leading the customer to repeatedly inquire about the review progress. Therefore, once the document review node is designated as a critical progress node, relevant progress information can be pushed to the customer, promptly informing them of the reason for the delay and the subsequent handling plan, keeping them informed and alleviating their anxiety.

[0063] In some optional implementations of this embodiment, refer to Figure 3 After the above step of determining whether the processing progress node is in an abnormal state, the following steps may also be included:

[0064] Step S2024: If there is no abnormal state, obtain historical claim case data;

[0065] Specifically, when the claims processing model determines that the current processing progress node is not in an abnormal state, it obtains pre-collected historical claims case data. This historical claims case data contains historical user attention points for various types of historical claims at different claim progress nodes. This historical attention point may include historical user attention information, consultation questions, and claims experience feedback related to the claim progress node.

[0066] Step S2025: determining whether there is a user focus point at the processing progress node based on the progress status data and the historical claims case data;

[0067] Specifically, the current claim's progress status data is compared with historical claim data, and historical progress nodes similar to the current progress status data are extracted. By analyzing the customer's behavior and feedback at these similar progress nodes in the historical claim data, possible historical customer concerns are discovered. If a historical customer concern exists, it is determined that the processing progress node has a user concern.

[0068] For example, consider a health insurance claim currently being processed, currently at the "Waiting for Payment" stage, with the current progress node status displayed as "Documents reviewed and approved, awaiting payment." The system searches historical claims data for all health insurance claims at the "Waiting for Payment" stage, including customer concerns, inquiries, complaints, and subsequent feedback on the claims experience. This historical data indicates that many customers, at this stage, are concerned about key points such as the accuracy of the claim amount, the expected payment timeframe, and the convenience of the payment method. This means that similar historical progress nodes in historical claims data share historical customer concerns. Based on this, it is determined that the current health insurance claim is experiencing user concerns at the "Waiting for Payment" stage.

[0069] Step S2026: If there is a user focus point, the processing progress node is determined as a key progress node.

[0070] Specifically, when it is determined that a processing progress node has a user focus, the processing progress node is determined as a key progress node.

[0071] In this embodiment, the user's focus reflects that even if the current processing progress node does not have an abnormal state, the customer still has the need and expectation to understand the processing progress node. Timely pushing relevant information of the processing progress node where the user focuses to the customer can enhance the customer's trust and satisfaction.

[0072] In some optional implementations of this embodiment, refer to Figure 4The above step S203, i.e., generating progress push content according to the key progress nodes, the progress status data and the business knowledge data, may include the following steps:

[0073] Using the trained language model, extracting first business explanation data related to the key progress node from the business knowledge data;

[0074] Specifically, a language model is pre-trained using a large amount of historical claims case data. This allows the model to extract each claim progress node and status data from the historical claims case data, as well as the corresponding business analysis data. The model then learns the relationship between each claim progress node and status data and the business explanation data, resulting in a trained language model. The language model can utilize pre-trained models such as BERT and GPT. The trained language model can extract relevant first-level business explanation data from business knowledge data based on the characteristics of key progress nodes, and generate progress push content in combination with progress status data.

[0075] Specifically, key progress nodes, progress status data and business knowledge data are input into the trained language model, key node features are extracted from the key progress nodes, and business explanation features are extracted from the business knowledge data. According to the relationship between the claims progress nodes and status data and the business explanation data learned by the model, the key node features are matched with the business explanation features, and the matching business explanation features are screened out as the first business explanation data.

[0076] According to the first business explanation data, the progress status data is converted into progress explanation content as progress push content.

[0077] Specifically, once the first service explanation data is extracted, it is integrated with the progress status data using a language model to generate easily understandable progress explanation content, which serves as the progress push content. For example, the progress status data is parsed using the encoding layer of the speech model to generate a state feature vector. Semantic analysis is performed on the first service explanation data to extract a semantic feature vector. Fusion strategies such as concatenation, weighted summation, and attention mechanisms are used to fuse the state feature vector and the semantic feature vector. Based on the fused feature vector, the progress push content is generated. This progress push content is a natural language description of the progress status and service explanation of the current node.

[0078] For example, in a car insurance claim, if the key progress node is "damage assessment," the language model will extract primary business explanation data related to damage assessment from the business knowledge data, such as the definition of damage assessment (the process of determining the extent of vehicle damage and repair costs), the basis for damage assessment (such as vehicle damage conditions and market repair prices), and the role of damage assessment results (as a key reference for calculating compensation amounts). This primary business explanation data provides a professional knowledge foundation for the subsequent conversion of progress status data into progress explanation content.

[0079] Suppose the progress status data indicates that the damage assessment was completed on [specific date] and the assessed amount was [X] yuan. Based on the first business interpretation data, the language model will convert this progress status data into progress notification content: "Dear customer, the damage assessment for your auto insurance claim has been completed. The assessment was conducted on [specific date]. Based on the damage to your vehicle and market repair prices, the assessed amount was [X] yuan. This amount will serve as an important basis for calculating the subsequent compensation amount."

[0080] In this embodiment, a trained language model is used to accurately extract first-level business explanation data related to key progress milestones from business knowledge data, ensuring the pertinence and accuracy of the information. By converting progress status data into progress explanation content based on the first-level business explanation data, the pushed information is easier to understand. This not only improves the efficiency of information transmission but also enhances the user experience, allowing users to clearly understand the business logic behind the progress status, helping to improve their understanding and satisfaction with the claims process.

[0081] In some optional implementations of this embodiment, to help customers better understand the operational requirements and timelines for future progress, the progress push content pushed to customers may also include a preview of future progress. Therefore, before the above-mentioned step of generating the progress push content, the following steps may also be included:

[0082] Using the trained language model, extracting business process information of the claim case from the business knowledge data;

[0083] Specifically, a language model is pre-trained using a large amount of historical claims case data to enable the model to extract the claims process characteristics of various types of claims from the historical claims case data, resulting in a trained language model. This trained language model can determine future progress nodes following the current progress node based on the claims process characteristics. Based on the characteristics of the future progress nodes, it extracts relevant second business explanation data from the business knowledge data, generates preview explanation content based on the second business explanation data, and further combines this generated progress explanation content to generate the final progress push content.

[0084] Specifically, the trained language model is used to determine the case type of the claim case, and the business process features are extracted from the business knowledge data through the model. According to the claims process features of each type of claims case learned by the model, the target process features corresponding to the case type are determined, and the target process features are matched with the extracted business process features, and the matching business process features are screened out as the business process information of the claim case.

[0085] For example, for a car insurance claim, the language model will extract complete business process information: after the customer reports the case, the insurance company arranges surveyors to go to the scene for inspection; after the inspection is completed, the damage assessment process begins to determine the extent of vehicle damage and repair costs; after the damage assessment is completed, the customer submits the claim materials, and the insurance company reviews them; after the review is passed, the compensation is processed.

[0086] Determining future progress nodes based on the key progress nodes and the business process information;

[0087] Specifically, after extracting the business process information, the language model is combined with the current key progress nodes to determine the subsequent future progress nodes according to the order of the business process information.

[0088] Taking car insurance claims as an example, if the current key progress node is "inspection completed", based on business process information, the language model will determine that the next future progress nodes are "damage assessment" and "material review and payment".

[0089] extracting second business interpretation data related to the future progress node from the business knowledge data;

[0090] Specifically, after determining the future progress node, the language model is used to extract future node features from the future progress node and business explanation features from the business knowledge data. Based on the relationship between the claims progress node and status data learned by the model and the business explanation data, the future node features are matched with the business explanation features, and matching business explanation features are selected as the second business explanation data. This second business explanation data may include information such as the specific operation content, required time, possible problems, and solutions for the future progress node.

[0091] generating a preview explanation content according to the second business explanation data;

[0092] Specifically, for the extracted second business explanation data, easy-to-understand preview explanation content is generated based on the second business explanation data using a language model.

[0093] Furthermore, the above step of generating progress push content specifically includes:

[0094] The preview explanation content and the progress explanation content are integrated to obtain progress push content.

[0095] Specifically, the preview explanation content and the progress explanation content are integrated to generate the progress push content. For example, the second service explanation data is semantically analyzed using the encoding layer of the speech model to extract semantic feature vectors, and the preview explanation content is generated based on the extracted semantic feature vectors. Subsequently, the progress explanation content generated based on the first service explanation data is obtained, and the progress explanation content and the preview explanation content are spliced ​​together to generate the final progress push content.

[0096] Taking a car insurance claim as an example, let's assume the current progress explanation is: "Dear customer, the investigation of your car insurance claim has been completed. The investigators have taken detailed notes and photographs of the on-site situation." Combined with the preview explanation, the progress push content will be: "Dear customer, the investigation of your car insurance claim has been completed. The investigators have taken detailed notes and photographs of the on-site situation. Next, we will enter the damage assessment phase. Our professional assessors will determine the damage amount based on the vehicle damage and market repair price standards. The assessment is expected to be completed within [X] business days."

[0097] In this embodiment, a language model is used to extract business process information from business knowledge data, and future progress nodes are determined based on key progress nodes, accurately positioning subsequent processes. Secondary business explanation data related to future progress nodes is extracted to generate preview explanation content, which is then integrated with the progress explanation content. This makes the progress push more comprehensive, not only informing customers of the current progress status but also clearly presenting future progress operation requirements and time nodes, helping them plan ahead and enhancing their understanding of the entire claims process.

[0098] In some optional implementations of this embodiment, the above step S204, i.e., pushing the progress push content to the target customer of the claim case, may include the following steps:

[0099] Obtaining preference information of target customers for the claim case, and matching push content templates and push timings based on the preference information;

[0100] Specifically, information on the preferences of target customers for claims cases is collected in advance through various channels. For example, when customers purchase insurance products, questionnaires are used to understand their preferences for push notification methods, such as whether they prefer to receive notifications via SMS, app, or email; their preferences for push notification times, such as whether they prefer to receive notifications in the morning, afternoon, or evening on weekdays; and their preferences for the level of detail in the notification content. Some customers prefer concise and clear content, while others desire more detailed information. Based on the target customers' preferences, the notification model, such as the push channel, content template, and timing, is determined.

[0101] Generate personalized push text according to the push content template and the progress push content;

[0102] Specifically, after determining the push content template and progress push content, the progress push content is filled into the corresponding push content template to generate personalized push text. The push content template reserves multiple variable positions, such as progress node name, progress status, and estimated completion time. These variable positions are replaced with specific information based on the actual progress push content.

[0103] For example, for a target auto insurance customer, let's assume the current progress notification reads, "Your auto insurance claim has been assessed for [X] yuan. The expected payout will be received on [specific date]." The system will populate this information into the simplified template, generating a personalized notification message like, "Dear customer, your auto insurance claim has been assessed for [50,000] yuan. The expected payout will be received within [three business days]. If you have any questions, please feel free to contact us."

[0104] The personalized push text is sent to the target customer according to the push time.

[0105] Specifically, the generated personalized push text is sent to the target customer according to the pre-set push schedule. When the push schedule arrives, the corresponding interface is called to send the text based on the customer's preferred push method. If the push is an app push, the text is sent to the customer's linked app; if the push is an SMS push, the text is sent to the customer's mobile phone via the SMS platform; if the push is an email push, the text is sent as the body of the email to the customer's email address.

[0106] In this embodiment, by determining personalized push content and push method according to the target customer's preference information, the timeliness and effectiveness of the progress push content are ensured.

[0107] Further references Figure 5 , as a response to the above Figure 2 In order to realize the method shown in the figure, the present application provides an embodiment of a claim settlement progress pushing device. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0108] like Figure 5 As shown, the claim progress pushing device 400 described in this embodiment includes: a data acquisition module 401, a node determination module 402, a content generation module 403 and a content pushing module 404.

[0109] in:

[0110] Data acquisition module 401, used to obtain the processing progress nodes, progress status data and business knowledge data of the claim case;

[0111] A node determination module 402 is configured to use the trained claims processing model to determine whether the processing progress node is a key progress node based on the progress status data and the business knowledge data;

[0112] The content generation module 403 is configured to generate progress push content based on the key progress node, the progress status data, and the business knowledge data if the progress node is a key progress node;

[0113] The content push module 404 is used to push the progress push content to the target customer of the claim case.

[0114] In this embodiment, a claim case's processing progress nodes refer to the stage division points in the insurance claims process, such as the material receipt node, review node, processing node, and payment node. Each node represents a key step in the claims process. Progress status data reflects the current status and updates of the claim case at each processing progress node, such as whether a node has been completed, the completion time, and the estimated completion time. Business knowledge data covers insurance product information, business process data, business rules data, business interpretation data, relevant laws and policies, and other knowledge content related to the claim case, providing a foundation for claims processing and progress notification.

[0115] Specifically, the processing progress nodes, progress status data, and business knowledge data of the claim case are obtained, wherein the processing progress nodes and progress status data of the claim case are extracted in real time from the claims management system, and the progress status data at least includes the case status and processing time; and the business knowledge data related to the claim case is obtained from the pre-built business knowledge base. The business knowledge data may include business process data, business rule data, and business explanation data, wherein the business process data records the complete process steps of the insurance claim from the beginning to the end, such as the sequence of material receipt, review, processing, payment, etc. and the connection relationship between each link; the business rule data contains various rules that the claim case needs to follow in the claims process, such as material requirements, review standards, payment conditions, time limits, etc. The business explanation data provides easy-to-understand explanations of complex business processes and rules to help customers understand the significance and requirements of each link in the claims process.

[0116] For example, in an insurance claims scenario, for a car insurance claim case, the business knowledge data related to the car insurance claim case is obtained, as well as real-time information on whether the repair invoice, accident photos, hospital certificates and other materials submitted by the customer have been received during the material receiving stage (i.e., the processing progress node), as well as the specific time of receipt (i.e., progress status data); information such as whether the review has been completed during the review stage, the result of whether the review has been passed or failed, the materials required to be supplemented, and the review time.

[0117] In this embodiment, a large amount of historical claims case data is used in advance to train a claims processing model. The model has the ability to analyze and process claims progress data and business knowledge data, and can determine key progress nodes.

[0118] Specifically, the acquired progress status data and business knowledge data are input into the trained claims processing model. The model then determines, based on the business knowledge data, whether the current progress status data contains critical status information, such as abnormal conditions, the need for supplementary materials, or a node with a high attention weight. If the progress status data contains critical status information, the processing progress node is identified as a critical progress node.

[0119] For example, at a material review node, if the review time exceeds the normal range specified in the business knowledge data, or if the review result is failure and the customer needs to provide a large amount of additional materials, then the node is determined to be a critical progress node.

[0120] In this embodiment, when a processing progress node is determined to be a critical progress node, personalized progress push content is generated based on the critical progress node, progress status data, and business knowledge data. For example, business explanation data for the critical progress node is extracted from the business knowledge data, and the progress push content is generated by combining the business explanation data and progress status data. The push content can be generated using natural language generation technology to convert complex claims information into concise and easy-to-understand language.

[0121] For example, if the key progress node is the materials review stage, and the progress status data indicates that the materials review failed and that supplementary material A is required, business explanation data for this key progress node is extracted from the business knowledge data, such as "The materials review stage is the second stage in the auto insurance claims process, and the materials that need to be reviewed include A, B, and C." The progress push content generated by combining the business explanation data and progress status data may be: "Dear customer, your auto insurance claim case failed the materials review in the second stage - materials review due to the lack of material A. Please supplement the relevant materials as soon as possible so that we can continue to process your claim." In addition, the push content can also be combined with business explanation data to answer questions that customers may be concerned about, such as the submission method and review time for supplementary materials.

[0122] In this embodiment, the generated progress push content is pushed to the target customer of the claim case. For example, an appropriate push method is selected based on the target customer's preference information to convey the progress push content to the target customer. The push method may include a push template, a push channel, and a push time.

[0123] For example, if the customer's preference information includes that they like to receive email notifications through the APP between 10-12 o'clock, the generated progress push content will be pushed to the target customer of the claim case in the form of an email through the APP between 10-12 o'clock.

[0124] The claim progress push device 400 of the present application accurately locates the stage and related business processes of the claim case during the processing by obtaining the processing progress nodes, progress status data and business knowledge data of the claim case; by adopting the trained claim processing model, it determines whether the processing progress node is a key progress node based on the progress status data and business knowledge data, and can dynamically identify the key progress nodes that have a greater impact on customer experience and business efficiency; by generating progress push content based on key progress nodes, progress status data and business knowledge data, it can be personalized and customized according to the specific circumstances of each case. The progress push content not only includes the current progress information of the claim case, but also can be explained and illustrated in combination with business knowledge, making it easier for customers to understand the claim progress and reducing anxiety and misunderstandings caused by information opacity. By timely pushing the progress push content to the target customers of the claim case, customers can keep abreast of the latest developments of the claim case and enhance customer satisfaction.

[0125] In some optional implementations of this embodiment, the node determination module 402 includes a rule extraction submodule, a state determination submodule, and a node determination submodule, wherein:

[0126] A rule extraction submodule is used to extract business rule data related to the processing progress node from the business knowledge data using the trained claims processing model;

[0127] A status determination submodule, configured to determine whether the processing progress node is in an abnormal state based on the progress status data and the business rule data;

[0128] The node determination submodule is used to determine the processing progress node as a key progress node if an abnormal state exists.

[0129] In this embodiment, a claims processing model is pre-trained using a large amount of historical claims case data. This historical claims case data includes claims progress data and business knowledge data related to multiple historical claims processes. The claims progress data includes each processing progress node involved in the historical claims process and the corresponding progress status data; the business knowledge data includes at least the business process data, business rules data, and business interpretation data involved in the historical claims cases. The trained claims processing model is capable of analyzing and processing claims progress data and business knowledge data, and can identify key progress nodes.

[0130] Specifically, the acquired processing progress nodes, progress status data and business knowledge data are input into the trained claims processing model. The model uses the learned features and rules to extract node features from the processing progress nodes and business rule features from the business knowledge data. The extracted node features and business rule features are matched, and the matching business rule features are screened out as the business rule data related to the processing progress nodes extracted from the business knowledge data.

[0131] For example, when a car insurance claim reaches the "Document Review" stage, the claims processing model extracts rule features related to the auto insurance document review from the business knowledge data. These include the required types of review documents (such as repair invoices, accident liability determinations, and copies of driver's licenses and vehicle registrations), review time limits (required to complete the review within a few business days of receipt), and criteria for determining the authenticity and completeness of the documents. This extracted business rule data provides a basis for subsequent determination of anomalies at the processing progress node.

[0132] Specifically, the claims processing model is used to compare and analyze the acquired progress status data with the extracted business rule data. The progress status data reflects the actual progress of the claim case at each processing progress node, such as whether the review has started at the material review node, whether the review is completed, and the review results. The business rule data contains the various rules that need to be followed in the claims process, especially at each processing progress node, such as the material type requirements, review time limits, and material authenticity and completeness judgment criteria stipulated at the material review node. The progress status data is compared with the business rule data one by one to determine whether there is an abnormal state at the processing progress node. An abnormal state refers to a situation where the progress status data does not match the business rule data.

[0133] For example, if the progress status data indicates that the review has exceeded the time limit specified in the business rules, but the review results are still pending, the model will determine that the node is in an abnormal state. Alternatively, if the progress status data indicates that the review materials are missing necessary information, such as an unclear amount on a repair invoice, preventing the review from proceeding, the model will determine that the node is in an abnormal state and requires further processing. Through this comparative analysis, the model can accurately identify whether there are problems with the processing progress node.

[0134] Specifically, when the model identifies an abnormal state at a processing progress node, it designates that node as a critical progress node. For example, in a car insurance claim, an abnormality at the document review node can delay the claim settlement, leading the customer to repeatedly inquire about the review progress. Therefore, once the document review node is designated as a critical progress node, relevant progress information can be pushed to the customer, promptly informing them of the reason for the delay and the subsequent handling plan, keeping them informed and alleviating their anxiety.

[0135] In some optional implementations of this embodiment, the node determination module 402 further includes a case acquisition submodule, a focus determination submodule, and a progress determination submodule, wherein:

[0136] The case acquisition submodule is used to obtain historical claims case data if there is no abnormal state;

[0137] A focus determination submodule, configured to determine whether a user focus point exists at the processing progress node based on the progress status data and the historical claims case data;

[0138] The progress determination submodule is configured to determine the processing progress node as a key progress node if there is a user focus point.

[0139] Specifically, when the claims processing model determines that the current processing progress node is not in an abnormal state, it obtains pre-collected historical claims case data. This historical claims case data contains historical user attention points for various types of historical claims at different claim progress nodes. This historical attention point may include historical user attention information, consultation questions, and claims experience feedback related to the claim progress node.

[0140] Specifically, the current claim's progress status data is compared with historical claim data, and historical progress nodes similar to the current progress status data are extracted. By analyzing the customer's behavior and feedback at these similar progress nodes in the historical claim data, possible historical customer concerns are discovered. If a historical customer concern exists, it is determined that the processing progress node has a user concern.

[0141] For example, consider a health insurance claim currently being processed, currently at the "Waiting for Payment" stage, with the current progress node status displayed as "Documents reviewed and approved, awaiting payment." The system searches historical claims data for all health insurance claims at the "Waiting for Payment" stage, including customer concerns, inquiries, complaints, and subsequent feedback on the claims experience. This historical data indicates that many customers, at this stage, are concerned about key points such as the accuracy of the claim amount, the expected payment timeframe, and the convenience of the payment method. This means that similar historical progress nodes in historical claims data share historical customer concerns. Based on this, it is determined that the current health insurance claim is experiencing user concerns at the "Waiting for Payment" stage.

[0142] Specifically, if a processing progress node is identified as having user concerns, it is designated as a key progress node. These user concerns reflect that even if the current processing progress node is not experiencing an abnormal state, the customer still has a need and expectation to understand the processing progress node. Promptly delivering relevant information about the processing progress node to the customer can enhance customer trust and satisfaction.

[0143] In some optional implementations of this embodiment, the content generation module 403 includes a first data extraction submodule and a first content generation submodule, wherein:

[0144] A first data extraction submodule is configured to extract first business explanation data related to the key progress node from the business knowledge data using the trained language model;

[0145] The first content generation submodule is configured to convert the progress status data into progress explanation content as progress push content according to the first business explanation data.

[0146] Specifically, a language model is pre-trained using a large amount of historical claims case data. This allows the model to extract each claim progress node and status data from the historical claims case data, as well as the corresponding business analysis data. The model then learns the relationship between each claim progress node and status data and the business explanation data, resulting in a trained language model. The language model can utilize pre-trained models such as BERT and GPT. The trained language model can extract relevant first-level business explanation data from business knowledge data based on the characteristics of key progress nodes, and generate progress push content in combination with progress status data.

[0147] Specifically, key progress nodes, progress status data and business knowledge data are input into the trained language model, key node features are extracted from the key progress nodes, and business explanation features are extracted from the business knowledge data. According to the relationship between the claims progress nodes and status data and the business explanation data learned by the model, the key node features are matched with the business explanation features, and the matching business explanation features are screened out as the first business explanation data.

[0148] Specifically, once the first service explanation data is extracted, it is integrated with the progress status data using a language model to generate easily understandable progress explanation content, which serves as the progress push content. For example, the progress status data is parsed using the encoding layer of the speech model to generate a state feature vector. Semantic analysis is performed on the first service explanation data to extract a semantic feature vector. Fusion strategies such as concatenation, weighted summation, and attention mechanisms are used to fuse the state feature vector and the semantic feature vector. Based on the fused feature vector, the progress push content is generated. This progress push content is a natural language description of the progress status and service explanation of the current node.

[0149] For example, in a car insurance claim, if the key progress node is "damage assessment," the language model will extract primary business explanation data related to damage assessment from the business knowledge data, such as the definition of damage assessment (the process of determining the extent of vehicle damage and repair costs), the basis for damage assessment (such as vehicle damage conditions and market repair prices), and the role of damage assessment results (as a key reference for calculating compensation amounts). This primary business explanation data provides a professional knowledge foundation for the subsequent conversion of progress status data into progress explanation content.

[0150] Suppose the progress status data indicates that the damage assessment was completed on [specific date] and the assessed amount was [X] yuan. Based on the first business interpretation data, the language model will convert this progress status data into progress notification content: "Dear customer, the damage assessment for your auto insurance claim has been completed. The assessment was conducted on [specific date]. Based on the damage to your vehicle and market repair prices, the assessed amount was [X] yuan. This amount will serve as an important basis for calculating the subsequent compensation amount."

[0151] In some optional implementations of this embodiment, the content generation module 403 further includes a process extraction submodule, a future node determination submodule, a second data extraction submodule, and a second content generation submodule, wherein:

[0152] A process extraction submodule is used to extract the business process information of the claim case from the business knowledge data using the trained language model;

[0153] A future node determination submodule, configured to determine a future progress node based on the key progress node and the business process information;

[0154] A second data extraction submodule is configured to extract second business interpretation data related to the future progress node from the business knowledge data;

[0155] The second content generation submodule is configured to generate preview explanation content according to the second business explanation data, and integrate the preview explanation content with the progress explanation content to obtain progress push content.

[0156] In this embodiment, a language model is pre-trained using a large amount of historical claims case data to enable the model to extract the claims process characteristics of various types of claims from the historical claims case data, resulting in a trained language model. This trained language model can determine future progress nodes following the current progress node based on the claims process characteristics. Based on the characteristics of the future progress nodes, it extracts relevant second business explanation data from the business knowledge data, generates preview explanation content based on the second business explanation data, and further combines this generated progress explanation content to generate the final progress push content.

[0157] Specifically, the trained language model is used to determine the case type of the claim case, and the business process features are extracted from the business knowledge data through the model. According to the claims process features of each type of claims case learned by the model, the target process features corresponding to the case type are determined, and the target process features are matched with the extracted business process features, and the matching business process features are screened out as the business process information of the claim case.

[0158] For example, for a car insurance claim, the language model will extract complete business process information: after the customer reports the case, the insurance company arranges surveyors to go to the scene for inspection; after the inspection is completed, the damage assessment process begins to determine the extent of vehicle damage and repair costs; after the damage assessment is completed, the customer submits the claim materials, and the insurance company reviews them; after the review is passed, the compensation is processed.

[0159] Specifically, after extracting the business process information, the language model is combined with the current key progress nodes to determine the subsequent future progress nodes according to the order of the business process information.

[0160] Taking car insurance claims as an example, if the current key progress node is "inspection completed", based on business process information, the language model will determine that the next future progress nodes are "damage assessment" and "material review and payment".

[0161] Specifically, after determining the future progress node, the language model is used to extract future node features from the future progress node and business explanation features from the business knowledge data. Based on the relationship between the claims progress node and status data learned by the model and the business explanation data, the future node features are matched with the business explanation features, and matching business explanation features are selected as the second business explanation data. This second business explanation data may include information such as the specific operation content, required time, possible problems, and solutions for the future progress node.

[0162] Specifically, upon extracting the second service explanation data, a language model is used to generate easily understandable preview explanation content based on the second service explanation data. This preview explanation content is then integrated with the progress explanation content to generate the progress push content. For example, the second service explanation data is semantically analyzed using the encoding layer of the speech model to extract semantic feature vectors. The preview explanation content is then generated based on the extracted semantic feature vectors. Subsequently, the progress explanation content generated based on the first service explanation data is obtained, and the progress explanation content and the preview explanation content are concatenated to generate the final progress push content.

[0163] Taking a car insurance claim as an example, let's assume the current progress explanation is: "Dear customer, the investigation of your car insurance claim has been completed. The investigators have taken detailed notes and photographs of the on-site situation." Combined with the preview explanation, the progress push content will be: "Dear customer, the investigation of your car insurance claim has been completed. The investigators have taken detailed notes and photographs of the on-site situation. Next, we will enter the damage assessment phase. Our professional assessors will determine the damage amount based on the vehicle damage and market repair price standards. The assessment is expected to be completed within [X] business days."

[0164] In some optional implementations of this embodiment, the content push module 404 includes a preference acquisition submodule, a personalized generation submodule, and a content push submodule, wherein:

[0165] A preference acquisition submodule is used to obtain the preference information of the target customer of the claim case and match the push content template and push time according to the preference information;

[0166] A personalized generation submodule, configured to generate a personalized push text based on the push content template and the progress push content;

[0167] The content push submodule is used to send the personalized push text to the target customer according to the push time.

[0168] Specifically, information on the preferences of target customers for claims cases is collected in advance through various channels. For example, when customers purchase insurance products, questionnaires are used to understand their preferences for push notification methods, such as whether they prefer to receive notifications via SMS, app, or email; their preferences for push notification times, such as whether they prefer to receive notifications in the morning, afternoon, or evening on weekdays; and their preferences for the level of detail in the notification content. Some customers prefer concise and clear content, while others desire more detailed information. Based on the target customers' preferences, the notification model, such as the push channel, content template, and timing, is determined.

[0169] Specifically, after determining the push content template and progress push content, the progress push content is filled into the corresponding push content template to generate personalized push text. The push content template reserves multiple variable positions, such as progress node name, progress status, and estimated completion time. These variable positions are replaced with specific information based on the actual progress push content.

[0170] For example, for a target auto insurance customer, let's assume the current progress notification reads, "Your auto insurance claim has been assessed for [X] yuan. The expected payout will be received on [specific date]." The system will populate this information into the simplified template, generating a personalized notification message like, "Dear customer, your auto insurance claim has been assessed for [50,000] yuan. The expected payout will be received within [three business days]. If you have any questions, please feel free to contact us."

[0171] Specifically, the generated personalized push text is sent to the target customer according to the pre-set push schedule. When the push schedule arrives, the corresponding interface is called to send the text based on the customer's preferred push method. If the push is an app push, the text is sent to the customer's linked app; if the push is an SMS push, the text is sent to the customer's mobile phone via the SMS platform; if the push is an email push, the text is sent as the body of the email to the customer's email address.

[0172] To solve the above technical problems, the present application also provides a computer device. Figure 6 , Figure 6 This is a basic structural block diagram of the computer device in this embodiment.

[0173] The computer device 6 includes a memory 61, a processor 62, and a network interface 63 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 6 with a memory 61, a processor 62, and a network interface 63, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to a microprocessor, an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), an embedded device, etc.

[0174] The computer device may be a desktop computer, notebook computer, PDA, cloud server, etc. The computer device may interact with the user via a keyboard, mouse, remote control, touchpad, or voice control device.

[0175] The memory 61 includes at least one type of readable storage medium, including flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic storage, a magnetic disk, an optical disk, etc. In some embodiments, the memory 61 may be an internal storage unit of the computer device 6, such as the hard disk or memory of the computer device 6. In other embodiments, the memory 61 may also be an external storage device of the computer device 6, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash memory card, etc. Of course, the memory 61 may also include both the internal storage unit of the computer device 6 and its external storage device. In this embodiment, the memory 61 is generally used to store the operating system and various application software installed on the computer device 6, such as computer-readable instructions for the claims progress push method. In addition, the memory 61 can also be used to temporarily store various types of data that have been output or are to be output.

[0176] In some embodiments, the processor 62 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 62 is generally used to control the overall operation of the computer device 6. In this embodiment, the processor 62 is used to execute computer-readable instructions stored in the memory 61 or process data, such as computer-readable instructions for executing the method for pushing the claim progress.

[0177] The network interface 63 may include a wireless network interface or a wired network interface. The network interface 63 is generally used to establish a communication connection between the computer device 6 and other electronic devices.

[0178] The present application also provides another embodiment, namely, providing a computer-readable storage medium, which stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the claim progress push method as described above.

[0179] The embodiment of the present application provides a computer device and a computer-readable storage medium, which obtains the processing progress nodes, progress status data and business knowledge data of the claims case through the processor, and accurately locates the stage and related business processes of the claims case during the processing; by adopting the trained claims processing model, it determines whether the processing progress node is a key progress node based on the progress status data and business knowledge data, and can dynamically identify the key progress nodes that have a greater impact on customer experience and business efficiency; by generating progress push content based on the key progress nodes, progress status data and business knowledge data, it can be personalized and customized according to the specific circumstances of each case. The progress push content not only contains the current progress information of the claims case, but also explains and illustrates it in combination with business knowledge, making it easier for customers to understand the claims progress and reducing anxiety and misunderstandings caused by information opacity. By pushing the progress push content to the target customers of the claims case in a timely manner, customers can keep abreast of the latest developments of the claims case and enhance customer satisfaction.

[0180] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0181] Obviously, the embodiments described above are only some of the embodiments of the present application, rather than all of the embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present application specification and the accompanying drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present application.

[0182] The non-Company software tools or components appearing in the embodiments of this application are merely examples and do not represent actual use.

Claims

1. A method for pushing claim progress, characterized in that: The steps include: Obtain the processing progress nodes, progress status data and business knowledge data of claims cases; Using the trained claims processing model, determining whether the processing progress node is a key progress node based on the progress status data and the business knowledge data; If it is a key progress node, generating progress push content based on the key progress node, the progress status data and the business knowledge data; The progress push content is pushed to the target customers of the claim case.

2. The method for pushing claim progress according to claim 1, characterized in that: The step of using the trained claims processing model to determine whether the processing progress node is a key progress node based on the progress status data and the business knowledge data includes: Using the trained claims processing model, extracting business rule data related to the processing progress node from the business knowledge data; Determining whether the processing progress node is in an abnormal state according to the progress status data and the business rule data; If an abnormal state exists, the processing progress node is determined as a key progress node.

3. The method for pushing claim progress according to claim 2, characterized in that: After the step of determining whether the processing progress node is in an abnormal state, the method further includes: If there is no abnormal state, obtain historical claims case data; Determining whether a user's attention point exists at the processing progress node based on the progress status data and the historical claims case data; If there is a user focus point, the processing progress node is determined as a key progress node.

4. The method for pushing claim progress according to claim 1, characterized in that: The step of generating progress push content according to the key progress nodes, the progress status data, and the business knowledge data includes: Using the trained language model, extracting first business explanation data related to the key progress node from the business knowledge data; According to the first business explanation data, the progress status data is converted into progress explanation content as progress push content.

5. The method for pushing claim progress according to claim 4, characterized in that: The progress push content also includes preview explanation content. Before the step of generating the progress push content, the step further includes: Using the trained language model, extracting business process information of the claim case from the business knowledge data; Determining future progress nodes based on the key progress nodes and the business process information; extracting second business interpretation data related to the future progress node from the business knowledge data; generating a preview explanation content according to the second business explanation data; The step of generating progress push content specifically includes: The preview explanation content and the progress explanation content are integrated to obtain progress push content.

6. The method for pushing claim progress according to any one of claims 1 to 5, characterized in that: The step of pushing the progress push content to the target customer of the claim case includes: Obtaining preference information of target customers for the claim case, and matching push content templates and push timings based on the preference information; Generate personalized push text according to the push content template and the progress push content; The personalized push text is sent to the target customer according to the push time.

7. A claim progress push device, characterized in that: include: Data acquisition module, used to obtain the processing progress nodes, progress status data and business knowledge data of claims cases; a node determination module, configured to use the trained claims processing model to determine whether the processing progress node is a key progress node based on the progress status data and the business knowledge data; A content generation module is configured to generate progress push content based on the key progress node, the progress status data, and the business knowledge data if the progress is a key progress node; The content push module is used to push the progress push content to the target customers of the claim case.

8. The device according to claim 7, characterized in that The node determination module includes: A rule extraction submodule is used to extract business rule data related to the processing progress node from the business knowledge data using the trained claims processing model; A status determination submodule, configured to determine whether the processing progress node is in an abnormal state based on the progress status data and the business rule data; The node determination submodule is used to determine the processing progress node as a key progress node if an abnormal state exists.

9. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, it implements the steps of the claim progress pushing method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the claim progress pushing method according to any one of claims 1 to 6.

Images