Method and system for automatic medical insurance settlement based on rule engine and model fusion

Abstract

This invention relates to the fields of data processing and insurance claims technology, specifically disclosing an automatic claims calculation method and system for accident and health insurance based on the fusion of a rule engine and a model. This invention constructs an automatic claims calculation architecture for accident and health insurance that integrates a rule engine and a model. It receives claims application data, performs structured processing and validity verification, matches pre-set claims calculation rule sets based on case characteristics, inputs core calculation parameters into a business rule model, and finally automatically calculates the compensation amount. This solves the technical problems of low efficiency, high human error rate, poor rule adaptability, high resource consumption, and cumbersome review processes associated with manual claims calculation. It improves the efficiency and accuracy of accident and health insurance claims processing, reduces the labor and system maintenance costs for insurance companies, shortens the claims cycle, and improves the customer claims experience.

Description

Technical Field

[0001] This invention relates to the fields of data processing and insurance claims technology, and in particular to an automatic claims calculation method and system for accident and health insurance based on rule engine and model fusion. Background Technology

[0002] As a core part of the insurance claims process, the efficiency and accuracy of accident and health insurance claims processing directly affect customer satisfaction and the operating costs of insurance companies. Currently, the industry primarily uses a manual claims processing model. Claims personnel must receive each customer's claim application and related documents, manually verify policy information, the authenticity of the accident, and the reasonableness of the expenses, and then calculate and confirm the compensation amount based on insurance terms, reimbursement ratios, deductibles, and other rules. While some companies have introduced information technology-assisted systems, these only achieve electronic storage of claims documents and basic data entry functions. A few systems, although capable of calculating fixed amounts, cannot adapt to the complexity of accident and health insurance product terms and the diversity of claims scenarios.

[0003] The aforementioned traditional technical solutions have significant drawbacks. First, they are extremely inefficient: manual verification and calculation on a case-by-case basis lead to lengthy processing times per case, easily resulting in backlogs when dealing with large volumes of claims, and a poor customer experience. Second, they suffer from high human error rates: claims personnel, due to differences in proficiency, fatigue, and misunderstandings of the terms, are prone to calculation errors in areas such as insured amount verification, ratio calculation, and deductible deduction, leading to discrepancies in payout amounts. Third, they have poor rule adaptability: accident and health insurance product terms cover complex rules such as diverse payout ratios, deductible conditions, exclusions of liability, and medical expense reimbursement scope. Existing simple systems cannot flexibly configure and adapt to various rules; adding or modifying terms requires redeveloping code, resulting in high maintenance costs. Fourth, they consume significant resources: insurance companies need to invest a large number of professional claims personnel in claims processing, leading to high labor costs and long training cycles, making it difficult to quickly adapt to business expansion. Fifth, the review process is cumbersome: manual calculation results require multi-level review, further lengthening the claims cycle, and the reliance on manual judgment in the review process makes standardized control of the entire process impossible.

[0004] Therefore, there is an urgent need to provide a technical solution to address the above problems. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides an automatic claims settlement method and system for health and accident insurance based on the fusion of rule engines and models.

[0006] Firstly, this invention provides an automatic claims settlement method for health and accident insurance based on the fusion of rule engines and models. The technical solution of this method is as follows: Receive accident and health insurance claim application data, and perform structured processing on the accident and health insurance claim application data to obtain structured claim data; Obtain the policy information corresponding to the structured claims data, and perform validity verification on the structured claims data based on the policy information; Extract case features from the validated structured claims data, input the case features into the rule engine, and use the rule engine to match the case features in a pre-set rule model library to obtain a set of claims calculation rules corresponding to the case features; The structured claims data and the claims calculation rule set are input into a pre-trained business rule model, and the business rule model outputs the core calculation parameters required to determine the compensation amount. Based on the set of claims rules and the core calculation parameters, the compensation amount is calculated on the structured claims data to obtain the final compensation amount.

[0007] The beneficial effects of the automatic claims settlement method for health and accident insurance based on the fusion of rule engine and model of the present invention are as follows: The method of this invention constructs an automatic claims settlement architecture for accident and health insurance that integrates a rule engine and a model. It receives claims application data, performs structured processing and validity verification, matches a pre-set set of claims settlement rules based on case characteristics, inputs the core calculation parameters into the business rule model, and finally completes the automatic calculation of the compensation amount. This solves the technical problems of low efficiency, high human error rate, poor rule adaptability, high resource consumption, and cumbersome review process in manual claims settlement. It improves the efficiency and accuracy of claims processing in accident and health insurance, reduces the labor costs and system maintenance costs of insurance companies, shortens the claims settlement cycle, and improves the customer claims experience.

[0008] Based on the above scheme, the automatic claims settlement method for health and accident insurance based on the fusion of rule engine and model of the present invention can be further improved as follows.

[0009] In one alternative approach, the step of structuring the accident and health insurance claim data to obtain structured claim data includes: Receive health insurance claim application data from different data sources, including structured data, semi-structured data and unstructured data; For the unstructured data, text information is extracted using optical character recognition technology; For the semi-structured data, key fields are extracted using a preset parsing template; The text information, key fields, and structured data are integrated in a unified storage format to generate the structured claims data.

[0010] The beneficial effects of adopting the above optional methods are as follows: further extracting unstructured data text information through optical character recognition technology, and extracting key fields of semi-structured data through preset parsing templates, integrating multi-source heterogeneous data into structured claims data with a unified storage format, solving the technical problem that claims data with diverse sources and different formats are difficult to process directly, and improving the degree of data standardization and subsequent automated processing efficiency.

[0011] In one optional approach, the step of obtaining policy information corresponding to the structured claims data and performing validity verification on the structured claims data based on the policy information includes: Retrieve the policy information associated with the policy identifier in the structured claims data from the core business system; Verify whether the date of the incident in the structured claims data is within the coverage period based on the coverage period in the policy information; Verify whether the identity information in the structured claims data is consistent with the insured information in the policy information; Verify whether the accident description in the structured claims data falls under the insurance liability based on the scope of liability in the policy information; Verify whether the application date in the structured claims data is within the claim statute of limitations according to the claim statute of limitations provisions in the policy information; Verify whether the medical expense details in the structured claims data fall within the reimbursable scope according to the expense type provisions in the policy information; When all verifications pass, the structured claims data is deemed to have passed the validity verification.

[0012] The beneficial effects of adopting the above optional methods are as follows: further retrieve policy information from the core business system, and verify the date of the incident, the coverage period, the consistency of identity information, the attribution of liability for the accident, the compliance of the claim time limit, and the scope of medical expense reimbursement in turn. This solves the technical problem of incomplete verification of the validity of claims data and improves the ability to control risks in advance and the accuracy of claims settlement.

[0013] In one alternative approach, the step of extracting case features from validated structured claims data includes: The insurance type identifier is parsed from the structured claims data that has passed verification; Identify accident description information from the verified structured claims data, and extract accident type tags from the accident description information according to preset accident type classification rules; Traverse the medical expense details from the verified structured claims data, and extract the medical expense type label for each item according to the preset expense category mapping relationship; Read the insured's date of birth from the verified structured claims data, and calculate the insured's age based on the current date and the date of birth; The case features are formed by combining the insurance type identifier, the accident type label, the medical expense type label, and the insured's age.

[0014] The beneficial effects of adopting the above optional methods are as follows: further analyzing the insurance type identifier, extracting the accident type label and medical expense type label from the verified claims data, and calculating the insured's age, combining multi-dimensional information into case features, solving the technical problem of single-dimensional case feature extraction, and improving the accuracy and personalized adaptation capabilities of the rule engine matching.

[0015] In one alternative approach, the step of inputting the case features into a rule engine and using the rule engine to match and obtain a set of claims settlement rules corresponding to the case features from a pre-set rule model library includes: The case features are input into the rule engine, which parses the insurance type identifier, accident type label, medical expense type label, and insured age from the case features. Based on the insurance type identifier, the rule group corresponding to the insurance type is located in the rule model library. The rule model library pre-stores the following rule types: reimbursement ratio rule model, deductible rule model, deductible rate rule model, medical expense reimbursement scope rule model, liability exemption rule model, and superimposed reimbursement rule model. Based on the accident type label, the medical expense type label, and the insured's age, at least one rule model matching the case characteristics is traversed and filtered in the rule group. Extract corresponding rule parameters from the at least one rule model and combine them to generate the set of reasoning rules corresponding to the case characteristics.

[0016] The advantages of adopting the above optional method are as follows: further locate the rule group according to the insurance type identifier, and combine and generate the claim settlement rule set based on the rule model of accident type label, medical expense type label and insured age through screening and matching. This solves the technical problem of low rule retrieval efficiency under complex terms and improves the flexibility and convenience of claim settlement rule configuration.

[0017] In one alternative approach, the steps of inputting the structured claims data and the claims calculation rule set into a pre-trained business rule model, and outputting the core calculation parameters required to determine the compensation amount through the business rule model, include: The structured claims data and the claims calculation rule set are used as input features to the business rule model, which is a machine learning model or rule reasoning model trained based on historical accident and health insurance claims data. The business rule model is used to infer and calculate the input features to generate the cost assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient corresponding to the structured claims data. The reasonableness of the value range of the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient shall be verified respectively. When the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient are all within a preset reasonable range, the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient are combined into the core calculation parameter; when at least one of the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient is not within the preset reasonable range, a verification failure flag is generated, and the structured claims data and the claims calculation rule set are marked as pending manual review.

[0018] The beneficial effects of adopting the above-mentioned optional methods are as follows: By further inputting structured claims data and claims calculation rule sets into the business rule model, the cost assessment coefficient, liability determination coefficient and special situation adjustment coefficient are generated and their reasonableness is verified. This solves the technical problem that pure rule calculation is difficult to cover special situations and improves the level of intelligence in claims assessment and the ability to identify anomalies.

[0019] In one optional approach, the step of calculating the compensation amount based on the claim calculation rule set and the core calculation parameters to obtain the final compensation amount includes: Based on the medical expense reimbursement scope rule model in the claim settlement rule set, the details of reimbursable medical expenses are extracted from the structured claims data, and the total amount of reimbursable medical expenses is obtained by summarizing them. Based on the liability exemption rule model in the claim settlement rule set, expenses that fall within the scope of liability exemption are removed from the total amount of reimbursable medical expenses to obtain the verified total amount of medical expenses; Based on the deductible rule model or deductible rate rule model in the settlement rule set, the corresponding deductible amount is deducted from the total amount of verified medical expenses to obtain the base for reimbursement. The base amount to be compensated is calculated by proportional conversion based on the compensation ratio rule model in the compensation rule set. The basic compensation amount is weighted and adjusted based on the cost assessment coefficient, liability determination coefficient and special circumstances adjustment coefficient in the core calculation parameters to obtain the individual liability compensation amount. Identify the number of coverage items contained in the structured claims data; when the structured claims data contains a single coverage item, determine the single coverage item's compensation amount as the final compensation amount. When the structured claims data contains multiple coverage liabilities, the following steps are performed for each coverage liability: extracting details of reimbursable medical expenses from the structured claims data according to the medical expense reimbursement scope rule model in the claims calculation rule set; and then weighting and adjusting the basic compensation amount according to the expense assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient in the core calculation parameters to obtain the individual liability compensation amount corresponding to each coverage liability. Finally, the individual liability compensation amounts corresponding to each coverage liability are merged according to the superimposed compensation rule model in the claims calculation rule set to generate the final compensation amount.

[0020] The beneficial effects of adopting the above optional methods are as follows: further extracting reimbursable expenses based on the scope of medical expense reimbursement, eliminating exclusion expenses according to the liability exemption rules, calculating the base amount to be reimbursed based on the deductible conditions, converting the basic reimbursement amount according to the reimbursement ratio, and solving the technical problem of chaotic calculation logic in multi-insurance liability scenarios through weighted adjustment of core calculation parameters and merging of superimposed reimbursement rules, thereby improving the standardization and transparency of the final reimbursement amount calculation.

[0021] Secondly, this invention provides an automatic claims settlement system for health and accident insurance based on the fusion of rule engines and models. The technical solution of this system is as follows: The processing module is used to receive accident and health insurance claim application data and perform structured processing on the accident and health insurance claim application data to obtain structured claim data. The verification module is used to obtain policy information corresponding to the structured claims data and to verify the validity of the structured claims data based on the policy information. The matching module is used to extract case features from the verified structured claims data, input the case features into the rule engine, and use the rule engine to match the case features in a pre-set rule model library to obtain a set of claims calculation rules corresponding to the case features. The output module is used to input the structured claims data and the claims calculation rule set into a pre-trained business rule model, and output the core calculation parameters required to determine the compensation amount through the business rule model; The calculation module is used to calculate the compensation amount based on the set of compensation rules and the core calculation parameters, and obtain the final compensation amount.

[0022] The beneficial effects of the automatic claims system for health and accident insurance based on rule engine and model fusion of the present invention are as follows: The system of this invention constructs an automatic claims settlement architecture for accident and health insurance that integrates a rule engine and a model. It receives claims application data, performs structured processing and validity verification, matches pre-set claims settlement rule sets based on case characteristics, inputs business rule models, and outputs core calculation parameters. Finally, it automatically calculates the compensation amount, solving the technical problems of low efficiency, high human error rate, poor rule adaptability, high resource consumption, and cumbersome review process in manual claims settlement. It improves the efficiency and accuracy of accident and health insurance claims processing, reduces the labor costs and system maintenance costs of insurance companies, shortens the claims settlement cycle, and improves the customer claims experience.

[0023] Thirdly, the technical solution of an electronic device according to the present invention is as follows: It includes a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps of the automatic claims settlement method for health and accident insurance based on rule engine and model fusion of the present invention.

[0024] Fourthly, the technical solution of a computer-readable storage medium provided by the present invention is as follows: The computer-readable storage medium stores instructions that, when read, cause the computer-readable storage medium to perform the steps of the automatic claims processing method for health and accident insurance based on rule engine and model fusion of the present invention.

[0025] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0026] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a flowchart illustrating an embodiment of an automatic claims settlement method for health and accident insurance based on the fusion of rule engine and model according to the present invention. Figure 2 This is a schematic diagram of an embodiment of the automatic claims settlement system for health and accident insurance based on the fusion of rule engine and model according to the present invention. Figure 3 This is a schematic diagram of an embodiment of an electronic device according to the present invention. Detailed Implementation

[0027] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0028] Figure 1 This diagram illustrates a flowchart of an embodiment of an automatic claims settlement method for health and accident insurance based on rule engine and model fusion provided by the present invention. This method can be executed by electronic devices such as terminal devices or servers. The terminal device can be any fixed or mobile terminal, such as user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, or wearable device. The server can be a single server or a server cluster consisting of multiple servers. Any electronic device can implement the automatic claims settlement method for health and accident insurance based on rule engine and model fusion by having its processor call computer-readable instructions stored in its memory. Figure 1 As shown, it includes the following steps: S1. Receive accident and health insurance claim application data, and perform structured processing on the accident and health insurance claim application data to obtain structured claim data.

[0029] Among them, accident and health insurance claim application data refers to all original materials and information submitted by the policyholder or insured to the insurance company for claim application after an insured event occurs; for example, Mr. A, after being injured in a traffic accident, submitted a claim application form, ID photo, scanned copy of the traffic accident determination letter, hospital outpatient medical records, medical expense invoices and expense lists, etc., through the insurance company's APP. Structured claim data refers to standardized data that is organized according to a predetermined data model after the original claim application data has been cleaned, parsed and formatted, making it easy for computers to read and process directly; for example, the name and ID number fields extracted from Mr. A's ID photo using optical character recognition technology are integrated with the information such as the time of the accident and the location of the accident filled in online and stored in a database as a single record.

[0030] S2. Obtain the policy information corresponding to the structured claims data, and verify the validity of the structured claims data based on the policy information.

[0031] Among them, policy information refers to all the agreed contents related to a specific policy as stated in the insurance contract, including data such as the policyholder, the insured, the insured object, the sum insured, the insurance period, the scope of liability, the deductible, and the reimbursement ratio; for example, the claims settlement system retrieves Mr. A's accidental injury medical insurance policy information from the core business system, including policy number P123456, insured Mr. A, coverage period from January 1, 2026 to December 31, 2026, accidental medical insurance amount of RMB 100,000, deductible of RMB 100 per accident, and reimbursement ratio of 90%, etc.

[0032] Among them, validity verification refers to a multi-dimensional review of claim application data based on policy information and business rules to determine whether the claim application is eligible to enter the settlement process. For example, the settlement system checks whether Mr. A's accident date of January 1, 2026 is within the coverage period, whether the ID information is consistent with the insured person on the policy, whether the accident description "traffic accident injury" falls within the scope of accidental injury liability, whether the application date of January 5, 2026 is within the claim period, and whether the drug expenses in the medical expense details are reimbursable items. If all of these checks pass, the verification is deemed valid.

[0033] S3. Extract case features from the verified structured claims data, input the case features into the rule engine, and use the rule engine to match the set of claims settlement rules corresponding to the case features in the preset rule model library.

[0034] Case features refer to the set of elements extracted from structured claims data that characterize the key attributes of a claims case, used for subsequent rule matching and model reasoning. For example, the claims system extracts information from Mr. A's claims data such as "accidental medical insurance," "traffic accident," "medical expenses" including "examination fees," "medication fees," and "treatment fees," and the insured's age of 45, combining this information into case features. A rule engine is a software component based on predefined business rules, capable of dynamically matching and executing corresponding rule logic based on input factual data. For example, after inputting Mr. A's case features into the rule engine, the engine locates the accidental medical insurance rule group based on the insurance type, and then filters applicable reimbursement ratio rules, deductible rules, etc., based on the accident type, expense type, and age.

[0035] The rule model library refers to a collection that stores and manages various business rule models. These models are defined in a structured format and support visual configuration and dynamic loading. For example, the rule model library contains rule models for accidental medical insurance, such as reimbursement ratio rules, deductible rules, medical expense reimbursement scope rules, liability exclusion rules, and cumulative reimbursement rules. Each model includes rule conditions and parameters. The claims settlement rule set refers to the set of all specific rule parameters matched from the rule model library for a specific claims case. For example, the claims settlement rule set generated by the rule engine for Mr. A's case includes: a reimbursement ratio of 90%, a deductible of 100 yuan, medical expense reimbursement scope limited to expenses within the medical insurance catalog, liability exclusion clauses excluding traffic accidents, and no cumulative reimbursement rules.

[0036] S4. Input the structured claims data and the claims calculation rule set into the pre-trained business rule model, and output the core calculation parameters required to determine the compensation amount through the business rule model.

[0037] Among them, the business rule model refers to a machine learning model or an expert-defined rule reasoning model trained based on historical claims data, which can output the core adjustment parameters required for calculating the compensation amount based on the input structured claims data and the set of claims calculation rules. For example, the gradient boosting tree model, after inputting Mr. A's structured claims data and the set of claims calculation rules, can infer the cost assessment coefficient of 1.0, the liability determination coefficient of 1.0, and the special circumstances adjustment coefficient of 1.0.

[0038] Among them, the core calculation parameters refer to the set of coefficients output by the business rule model, which are used to make fine adjustments based on the basic compensation amount; for example, the cost assessment coefficient 1.0, liability determination coefficient 1.0, and special circumstances adjustment coefficient 1.0 output by the model together constitute the core calculation parameters, which are used for the weighted calculation of subsequent compensation amounts.

[0039] S5. Based on the set of claims rules and the core calculation parameters, calculate the compensation amount for the structured claims data to obtain the final compensation amount.

[0040] The final compensation amount refers to the amount of compensation that the insurance company should pay to the insured or beneficiary after a complete settlement process. For example, Mr. A's final compensation amount was calculated to be 4,365 yuan, which is the final claim result.

[0041] The technical solution in this embodiment constructs an automatic claims settlement architecture for accident and health insurance that integrates a rule engine and a model. It receives claims application data, performs structured processing and validity verification, matches pre-set claims settlement rule sets based on case characteristics, inputs business rule models, and outputs core calculation parameters. Finally, it automatically calculates the compensation amount, solving the technical problems of low efficiency, high human error rate, poor rule adaptability, high resource consumption, and cumbersome review process in manual claims settlement. This improves the efficiency and accuracy of accident and health insurance claims processing, reduces the labor costs and system maintenance costs of insurance companies, shortens the claims settlement cycle, and improves the customer claims experience.

[0042] In one alternative approach, the step of structuring the accident and health insurance claim data to obtain structured claim data includes: The system receives health insurance claim data from different data sources, including structured data, semi-structured data, and unstructured data.

[0043] Structured data refers to data that follows a fixed format and data type and can be directly stored and processed in a two-dimensional table format. For example, the insured's name, ID number, and accident date entered by Mr. A in an online claim form are structured data. Semi-structured data refers to data that has some structure but lacks strict pattern constraints, typically containing labels or tags to separate data elements. For example, a PDF file of an electronic invoice issued by a hospital contains tagged information such as item names and amounts, but is not in a tabular format, thus it is semi-structured data. Unstructured data refers to text, images, audio, etc., that lack a predefined data model and cannot be directly stored in a row-column format. For example, a scanned copy of a paper medical certificate uploaded by Mr. A, containing handwritten text from a doctor, is unstructured data.

[0044] For the unstructured data, text information is extracted using optical character recognition technology.

[0045] Optical character recognition (OCR) technology refers to the technology of converting text information in an image into editable and searchable text data. For example, the Alipay system uses OCR to process a scanned copy of Mr. A's ID card, extracting text information such as "Name: Mr. A" and "ID Number: 110101198001011234". Text information refers to the text content recognized from an image using OCR technology; for example, recognizing text such as "Preliminary Diagnosis: Right Tibial Fracture" from a scanned copy of Mr. A's medical certificate.

[0046] For the semi-structured data, key fields are extracted using a preset parsing template.

[0047] Here, the parsing template refers to predefined field extraction rules for semi-structured data in a specific format, used to extract key fields from the data. For example, the data processing system has a pre-defined parsing template for hospital electronic invoices (PDF), defining the extraction location and method for fields such as invoice number, invoice date, item name, and amount. Key fields refer to core data items with business meaning extracted from semi-structured data; for example, the parsing template can be used to extract item names and amounts such as "examination fee 500 yuan" and "medication fee 400 yuan" from Mr. A's electronic invoice.

[0048] The text information, key fields, and structured data are integrated in a unified storage format to generate the structured claims data.

[0049] Among them, a unified storage format refers to a standard format formed by organizing data from different sources and of different types according to consistent field definitions, data types, and storage structures. For example, the Lisuan system converts Mr. A's online form data, ID card information extracted by optical character recognition, and expense details extracted from electronic invoices into a unified JSON format containing "field name-field value" pairs and stores it in the database.

[0050] Among the above-mentioned optional methods, unstructured data text information is further extracted through optical character recognition technology, and key fields of semi-structured data are extracted through preset parsing templates. Multi-source heterogeneous data are integrated into structured claims data with a unified storage format, which solves the technical problem that claims data with diverse sources and different formats are difficult to process directly, and improves the degree of data standardization and subsequent automated processing efficiency.

[0051] In one alternative approach, S2 specifically includes: Retrieve the policy information associated with the policy identifier in the structured claims data from the core business system.

[0052] The core business system refers to the information system used by the insurance company to manage core business processes such as policies, underwriting, and claims settlement. For example, the claims settlement system uses an application programming interface (API) to query and retrieve detailed information about Mr. A's policy P123456 from the insurance company's core business system in real time. The policy identifier is a unique identifier for each policy, typically the policy number. For example, the claims settlement system extracts "policy number P123456" from structured claims data as the policy identifier to retrieve the corresponding policy information.

[0053] Verify whether the date of the incident in the structured claims data is within the coverage period based on the coverage period in the policy information.

[0054] The coverage period refers to the time interval between the commencement and termination of insurance liability as stipulated in the insurance contract; for example, Mr. A's policy coverage period is from 00:00 on January 1, 2026 to 24:00 on December 31, 2026. The incident date refers to the specific date on which the insured event occurred; for example, Mr. A's traffic accident incident date is January 1, 2026.

[0055] Verify the consistency of the identity information in the structured claims data based on the insured information in the policy information.

[0056] Among them, the insured's information refers to the insured's identity information stated in the insurance contract, including name, ID number, etc.; for example, the claims settlement system compares the ID number in the claims application data with the information of Mr. A, the insured in the policy, and confirms that they match.

[0057] Verify whether the accident description in the structured claims data falls under the insurance liability based on the scope of liability in the policy information.

[0058] The scope of liability refers to the types of insured events and loss items for which the insurance company is liable for compensation, as stipulated in the insurance contract. For example, Mr. A's accidental medical insurance covers reasonable and necessary medical expenses caused by accidental injury, but does not cover medical treatment for diseases.

[0059] Verify whether the application date in the structured claims data is within the claim statute of limitations based on the claim statute of limitations provisions in the policy information.

[0060] The statute of limitations for claims refers to the effective period stipulated by insurance law or insurance contract for the insured or beneficiary to submit a claim application to the insurance company; for example, Mr. A's insurance contract stipulates that the statute of limitations for claims is within 2 years from the date of the insured event.

[0061] Verify whether the medical expense details in the structured claims data fall within the reimbursable scope based on the expense type provisions in the policy information.

[0062] Among them, the expense type refers to the nature classification of each expense in the medical expense list, such as registration fee, examination fee, drug fee, treatment fee, etc. For example, Mr. A's medical expense details include "examination fee of 500 yuan" and "drug fee of 400 yuan". The settlement system determines which ones are within the reimbursable scope based on the expense type.

[0063] When all verifications pass, the structured claims data is deemed to have passed the validity verification.

[0064] Among the above-mentioned optional methods, policy information is further retrieved from the core business system to verify the date of the incident, the coverage period, the consistency of identity information, the attribution of liability for the accident, the compliance of the claim time limit, and the scope of medical expense reimbursement. This solves the technical problem of incomplete verification of the validity of claims data and improves the ability to control risks in advance and the accuracy of claims settlement.

[0065] In one alternative approach, the step of extracting case features from validated structured claims data includes: The insurance type identifier is parsed from the structured claims data that has passed verification.

[0066] Among them, the insurance type identifier refers to the code or name used to distinguish different types of insurance products; for example, the claims settlement system identifies the insurance type as "accidental medical" from structured claims data, and the corresponding code is "YIWAI01".

[0067] Accident description information is identified from the verified structured claims data, and accident type tags are extracted from the accident description information according to preset accident type classification rules.

[0068] The accident description information refers to the written account of the insured event in the claim application; for example, Mr. A's claim statement, "On January 1, 2026, I was hit by a car while riding a bicycle, and my right leg was injured," constitutes the accident description information. The accident type classification rules refer to a set of predefined rules for extracting accident categories from the accident description information based on keywords or natural language processing technology; for example, the rule defines that when words such as "collision" or "traffic accident" appear in the description, the accident type is classified as "traffic accident." The accident type label refers to the standardized category used to characterize the nature of the accident after classification based on the accident description information; for example, after classifying Mr. A's accident description, the claims system labels it "traffic accident."

[0069] The medical expense details are traversed from the verified structured claims data, and the medical expense type label of each item is extracted according to the preset expense category mapping relationship.

[0070] Among them, the detailed medical expense items refer to the chargeable items listed in the medical expense list, including information such as item name, amount, and quantity; for example, Mr. A's expense list includes items such as "registration fee 100 yuan", "examination fee 500 yuan", "medication fee 400 yuan" and "treatment fee 3000 yuan".

[0071] The cost category mapping relationship refers to the corresponding rules that match the names of detailed medical expense items with standard cost types; for example, the settlement system pre-maps "registration fee" to "outpatient fee", "examination fee" to "examination fee", and "Western medicine fee" to "drug fee", etc. The medical expense type label refers to the standardized cost category name assigned to each detailed medical expense item; for example, the settlement system labels Mr. A's "examination fee of 500 yuan" as "examination fee".

[0072] The insured's date of birth is read from the verified structured claims data, and the insured's age is calculated based on the current date and the date of birth.

[0073] The insured's date of birth refers to the year, month, and day of birth recorded in the insurance contract; for example, Mr. A's date of birth is January 1, 1980. The insured's age refers to the age in full years calculated based on the insured's date of birth and the current date; for example, the claims system calculates Mr. A's age as 46 years old based on the claim application date of January 5, 2026.

[0074] The case features are formed by combining the insurance type identifier, the accident type label, the medical expense type label, and the insured's age.

[0075] Among the above-mentioned optional methods, the insurance type identifier, accident type label and medical expense type label are further analyzed from the verified claims data, and the insured's age is calculated. The multi-dimensional information is combined into case features, which solves the technical problem of single-dimensional case feature extraction and improves the accuracy and personalized adaptation capability of the rule engine matching.

[0076] In one alternative approach, the step of inputting the case features into a rule engine and using the rule engine to match and obtain a set of claims settlement rules corresponding to the case features from a pre-set rule model library includes: The case features are input into the rule engine, which then parses the insurance type identifier, accident type label, medical expense type label, and insured age from the case features.

[0077] Based on the insurance type identifier, the rule group corresponding to the insurance type is located in the rule model library. The rule model library pre-stores the following rule types: reimbursement ratio rule model, deductible rule model, deductible rate rule model, medical expense reimbursement scope rule model, liability exemption rule model, and superimposed reimbursement rule model.

[0078] Here, "insurance category" refers to the major categories of insurance products divided according to their coverage attributes, such as accident and health insurance, auto insurance, and property insurance; for example, Mr. A's accidental medical insurance belongs to the accident and health insurance category. "Rule group" refers to the collection of rule models in the rule model library, divided according to insurance category or business area; for example, the rule model library has a rule group for the accident and health insurance category, which contains rule models for all accidental medical insurance, hospitalization medical insurance, and other types of insurance.

[0079] The reimbursement ratio rule model defines the calculation rules for the insurance company's reimbursement based on a percentage of the reimbursable amount after an insured event. For example, Mr. A's reimbursement ratio rule model stipulates that after deducting the deductible, 90% of the remaining expenses will be reimbursed. The deductible rule model specifies the minimum amount of loss the insured must bear for each insured event. For example, Mr. A's deductible rule model sets the deductible at 100 yuan per event. The deductible rate rule model specifies a calculation rule that deducts a certain percentage of the loss amount as the deductible portion. For example, some health insurance policies use a deductible rate rule, i.e., a 5% deductible, but this rule does not apply to Mr. A's insurance policy.

[0080] The medical expense reimbursement scope rule model defines which medical expenses are eligible for reimbursement. For example, Mr. A's reimbursement scope rule model stipulates that only drugs and treatments within the medical insurance catalog are reimbursable, excluding self-paid drugs and treatments. The liability exclusion rule model lists specific circumstances or types of expenses under which the insurance company will not be liable for compensation. For example, Mr. A's exclusion rule model stipulates that expenses for accidents caused by illegal or criminal acts, cosmetic surgery, etc., are not covered. The cumulative compensation rule model specifies how the compensation amounts for each liability are calculated when the same insured event involves multiple coverage liabilities (such as accidental disability and accidental medical expenses). For example, Mr. A's accident insurance includes accidental medical expenses and accidental disability liabilities; the cumulative compensation rule model stipulates that the two liabilities are calculated independently and then paid together.

[0081] Based on the accident type label, the medical expense type label, and the insured's age, at least one rule model matching the case characteristics is traversed and filtered in the rule group.

[0082] Extract corresponding rule parameters from the at least one rule model and combine them to generate the set of reasoning rules corresponding to the case characteristics.

[0083] Among them, rule parameters refer to the numerical values ​​or conditions extracted from the rule model and used for specific calculations; for example, "100" yuan extracted from the deductible rule model and "90%" extracted from the payout ratio rule model are both rule parameters.

[0084] In the above-mentioned optional methods, the rule group is further located based on the insurance type identifier, and the rule model is combined to generate a set of claims settlement rules based on the accident type label, medical expense type label and the insured's age through screening and matching. This solves the technical problem of low rule retrieval efficiency under complex terms and improves the flexibility and convenience of claims settlement rule configuration.

[0085] In one alternative approach, S4 specifically includes: The structured claims data and the claims calculation rule set are input as input features into the business rule model, which is a machine learning model or rule reasoning model trained based on historical accident and health insurance claims data.

[0086] Input features refer to the data set input into the business rule model for reasoning, typically including structured claims data and a set of claims calculation rules. For example, Mr. A's structured claims data (including personal information, accident information, expense details, etc.) and the set of claims calculation rules (90% payout ratio, 100 yuan deductible, etc.) are combined as input to the model. Historical accident and health insurance claims data refer to the data accumulated by the insurance company from past accident and health insurance claims, including policy information, accident information, medical expenses, and manual calculation results. For example, when training the business rule model, nearly 1 million accident and health insurance claims records from the past five years were used, including historical payout data similar to Mr. A's case.

[0087] Machine learning models refer to mathematical models that learn patterns from historical data through training algorithms, enabling them to predict or classify new data. For example, a regression model trained on historical claims data using the gradient boosting tree algorithm is used to output core calculation parameters. Rule-based reasoning models refer to models derived through inference engines based on expert-defined logical rules and knowledge bases. For example, a rule-based reasoning model built on decision trees determines the liability determination coefficient based on conditions such as total cost and accident type.

[0088] The business rule model is used to infer and calculate the input features to generate the cost assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient corresponding to the structured claims data.

[0089] The expense assessment coefficient is a coefficient used to adjust the total reimbursable medical expenses, reflecting the impact of factors such as regional differences and price fluctuations on medical costs. For example, the business rule model outputs an expense assessment coefficient of 1.0 for Mr. A, indicating that the standard expense is used for calculation. The liability determination coefficient is a coefficient used to adjust the compensation liability, reflecting the impact of factors such as the nature of the accident and the insured's fault on compensation liability. For example, since Mr. A is not at fault in the accident, the model outputs a liability determination coefficient of 1.0. The special circumstances adjustment coefficient is a coefficient used to adjust the compensation amount, handling exceptional adjustments in rare or complex situations. For example, since Mr. A's case has no special circumstances, the model outputs a special circumstances adjustment coefficient of 1.0.

[0090] The reasonableness of the value range of the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient is verified respectively.

[0091] When the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient are all within a preset reasonable range, the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient are combined into the core calculation parameters.

[0092] The preset reasonable value range refers to the normal value range set in advance for each coefficient output by the business rule model, which is used to verify the reasonableness of the output results; for example, the preset reasonable range for the cost assessment coefficient is 0.8 to 1.2. If the model output value exceeds this range, an anomaly will be triggered.

[0093] When at least one of the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient is not within the preset reasonable value range, a verification failure flag is generated, and the structured claims data and the claims calculation rule set are marked as pending manual review.

[0094] The "verification failure" flag refers to a system-generated marker indicating that manual intervention is required for a case when the core calculation parameters output by the business rule model exceed a reasonable range. For example, if the claims system verifies Mr. A's coefficients and no failure flag is generated, but the coefficients are abnormal, the system will mark the case as "verification failure." "Pending manual review" refers to a case that is transferred to manual processing due to verification failures or abnormalities in the automated process. For example, if Mr. A's expense assessment coefficients exceed a preset range, the claims system will mark the case as "pending manual review" for further review by a claims specialist.

[0095] Among the above-mentioned optional methods, structured claims data and the set of claims calculation rules are further input into the business rule model to infer and generate cost assessment coefficients, liability determination coefficients and special situation adjustment coefficients, and perform reasonableness verification. This solves the technical problem that pure rule calculation is difficult to cover special situations and improves the level of intelligence in claims assessment and the ability to identify anomalies.

[0096] In one alternative approach, S5 specifically includes: Based on the medical expense reimbursement scope rule model in the claim settlement rule set, the details of reimbursable medical expenses are extracted from the structured claims data, and the total amount of reimbursable medical expenses is obtained by summarizing them.

[0097] The reimbursable medical expense details refer to the expense items identified as covered by insurance and included in the reimbursement calculation after being filtered through the reimbursement scope rules in the structured claims data. For example, in Mr. A's medical expense details, the examination fee of 500 yuan and the treatment fee of 3,000 yuan are reimbursable items, while the out-of-pocket portion of the drug cost of 50 yuan is excluded. The total reimbursable medical expense amount refers to the sum of the amounts of all reimbursable medical expense details. For example, Mr. A's total reimbursable medical expense amount is 100 yuan for registration, 500 yuan for examination, 350 yuan for drugs covered by medical insurance, and 3,000 yuan for treatment, totaling 3,950 yuan (original total cost of 5,000 yuan minus the out-of-pocket portion).

[0098] Based on the liability exemption rule model in the aforementioned claim settlement rule set, expenses that fall within the scope of liability exemption are removed from the total amount of reimbursable medical expenses to obtain the verified total amount of medical expenses.

[0099] The total amount of medical expenses is calculated as follows: the amount obtained after further deducting expenses within the scope of liability exemption from the total amount of reimbursable medical expenses; for example, if Mr. A's medical expenses do not include any items exempt from liability, the total amount of medical expenses calculated is still 3,950 yuan.

[0100] The deductible amount is deducted from the total amount of verified medical expenses according to the deductible rule model or deductible rate rule model in the settlement rule set to obtain the base for reimbursement.

[0101] The deductible amount refers to the amount that the insured must bear themselves, calculated according to the deductible or deductible rate rules. For example, Mr. A's deductible is 100 yuan, which will be deducted from the total assessed medical expenses. The payable base refers to the amount remaining after deducting the deductible, which serves as the basis for calculating the reimbursement ratio. For example, Mr. A's payable base is 3950 yuan minus 100 yuan, which equals 3850 yuan.

[0102] The base amount to be compensated is calculated by proportionally converting the base amount to be compensated based on the compensation ratio rule model in the compensation rule set.

[0103] The basic compensation amount refers to the initial compensation amount obtained by multiplying the base amount to be compensated by the compensation ratio; for example, Mr. A's basic compensation amount is 3,850 yuan multiplied by 90%, which equals 3,465 yuan.

[0104] The basic compensation amount is weighted and adjusted based on the cost assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient in the core calculation parameters to obtain the individual liability compensation amount.

[0105] The single liability compensation amount refers to the compensation amount calculated for a specific coverage liability (such as accidental medical treatment or accidental disability). For example, Mr. A's accidental medical single liability compensation amount is obtained after adjusting the basic compensation amount with core calculation parameters, which is 3465 yuan in this example.

[0106] Identify the number of coverage items contained in the structured claims data. When the structured claims data contains a single coverage item, determine the single coverage item's compensation amount as the final compensation amount.

[0107] When the structured claims data contains multiple coverage liabilities, the following steps are performed for each coverage liability: extracting details of reimbursable medical expenses from the structured claims data according to the medical expense reimbursement scope rule model in the claims calculation rule set; and then weighting and adjusting the basic compensation amount according to the expense assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient in the core calculation parameters to obtain the individual liability compensation amount corresponding to each coverage liability. Finally, the individual liability compensation amounts corresponding to each coverage liability are merged according to the superimposed compensation rule model in the claims calculation rule set to generate the final compensation amount.

[0108] Among the above-mentioned optional methods, the reimbursable expenses are further extracted based on the scope of medical expense reimbursement, the exclusion expenses are removed according to the liability exemption rules, the base amount to be reimbursed is calculated according to the deductible conditions, the basic reimbursement amount is converted according to the reimbursement ratio, and the technical problem of chaotic calculation logic in multi-insurance liability scenarios is solved by weighted adjustment of core calculation parameters and merging of superimposed reimbursement rules, thereby improving the standardization and transparency of the final reimbursement amount calculation.

[0109] In another embodiment of the automatic claims settlement method for health and accident insurance based on rule engine and model fusion of the present invention, the specific steps include: S10. Receive property insurance claim application data from different data sources. The property insurance claim application data includes structured data, semi-structured data, and unstructured data. For unstructured data, extract text information using optical character recognition technology. For semi-structured data, extract key fields using a preset parsing template. Integrate the text information, key fields, and structured data according to a unified storage format to generate standardized structured claim data. The standardized structured claim data includes the time of the accident, the location of the accident, the list of insured property, the list of losses, the insured's identity information, and the policy number field.

[0110] S20. Retrieve policy information associated with the policy identifier in the standardized structured claims data from the core business system; verify whether the date of the incident in the standardized structured claims data is within the coverage period based on the coverage period in the policy information; verify whether the identity information in the standardized structured claims data is consistent with the insured information based on the insured information in the policy information; verify whether the list of insured property in the standardized structured claims data belongs to the insured object based on the scope of the insured object in the policy information; verify whether the accident description in the standardized structured claims data falls under the insurance liability based on the scope of liability in the policy information; verify whether the application date in the standardized structured claims data is within the claim time limit based on the claim time limit provisions in the policy information; when all verifications pass, the standardized structured claims data is deemed to have passed the validity verification.

[0111] S30. Parse the insurance type identifier from the verified standard structured claims data; identify accident description information from the verified standard structured claims data, and extract accident type tags from the accident description information according to the preset accident type classification rules; traverse the loss list items from the verified standard structured claims data, and extract the loss type tags for each item according to the preset loss category mapping relationship; read the property type identifier from the insured property list from the verified standard structured claims data; combine the insurance type identifier, accident type tag, loss type tag, and property type identifier into case features; input the case features into the rule engine, and the rule engine... The process involves analyzing the insurance type identifier, accident type label, loss type label, and property type identifier in the case characteristics; locating the rule group corresponding to the insurance type category in the rule model library based on the insurance type identifier; pre-storing the rule model library according to rule type, including payout ratio rule model, deductible rule model, deductible rate rule model, loss assessment rule model, liability exemption rule model, and cumulative payout rule model; traversing and filtering at least one rule model matching the case characteristics based on the accident type label, loss type label, and property type identifier; extracting the corresponding rule parameters from at least one rule model and combining them to generate a set of claims settlement rules corresponding to the case characteristics.

[0112] S40. Standardized structured claims data and a set of claims calculation rules are input into the business rule model, which is a machine learning model trained on historical property insurance claims data. The business rule model performs inference calculations on the input features to generate loss assessment coefficients, liability determination coefficients, and special circumstance adjustment coefficients corresponding to the standardized structured claims data. The reasonableness of the value range of the loss assessment coefficients, liability determination coefficients, and special circumstance adjustment coefficients is verified. When the loss assessment coefficients, liability determination coefficients, and special circumstance adjustment coefficients are all within the preset reasonable value range, the loss assessment coefficients, liability determination coefficients, and special circumstance adjustment coefficients are combined into core calculation parameters. When at least one of the loss assessment coefficients, liability determination coefficients, and special circumstance adjustment coefficients is not within the preset reasonable value range, a verification failure flag is generated, and the standardized structured claims data and the set of claims calculation rules are marked as awaiting manual review.

[0113] S50. Extract verifiable loss details from standardized structured claims data according to the loss assessment rule model in the claims settlement rule set, and summarize to obtain the total verifiable loss amount; remove losses within the scope of liability exemption from the total verifiable loss amount according to the liability exemption rule model in the claims settlement rule set, to obtain the total assessed loss amount; deduct the corresponding deductible amount from the total assessed loss amount according to the deductible amount rule model or deductible rate rule model in the claims settlement rule set, to obtain the base amount to be paid; convert the base amount to be paid according to the ratio rule model in the claims settlement rule set, to obtain the basic payment amount; adjust the basic payment amount with weights according to the loss assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient in the core calculation parameters, to obtain the individual liability payment amount; identify the target The standardized structured claims data includes a number of coverage items. When the standardized structured claims data contains a single coverage item, the single coverage amount is determined as the final compensation amount. When the standardized structured claims data contains multiple coverage items, for each coverage item, the steps are as follows: extracting verifiable loss details from the standardized structured claims data according to the loss assessment rule model in the claims calculation rule set; then, weighting the basic compensation amount according to the loss assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient in the core calculation parameters; obtaining the single coverage amount corresponding to each coverage item; and finally, merging the single coverage amounts corresponding to each coverage item according to the superimposed compensation rule model in the claims calculation rule set to generate the final compensation amount.

[0114] S60. Store the standardized structured claims data, claims calculation rule set, core calculation parameters, final compensation amount, and full process records of this claims case into the historical database for incremental training of the business rule model and optimization and updating of the rule model library.

[0115] This embodiment receives and structures multi-source property insurance claims data, performs multi-dimensional validity verification based on policy information, extracts case features, matches the claims calculation rule set with a rule engine, and then outputs core calculation parameters in conjunction with the business rule model to automatically complete the calculation of the compensation amount. The entire process data is stored for continuous optimization of the model and rules. This solves the technical problems of low efficiency, high human error rate, poor rule adaptability, high resource consumption, and cumbersome review process caused by the reliance on manual labor in traditional property insurance claims. It realizes the automation and intelligence of the entire claims process, improves the efficiency and accuracy of claims processing, reduces labor costs and system maintenance costs, shortens the claims cycle, and improves the customer claims experience.

[0116] Figure 2 This diagram illustrates the structure of an embodiment of an automatic claims processing system 200 for health and accident insurance based on rule engine and model fusion provided by the present invention. Figure 2 As shown, the automatic claims settlement system 200 for health and accident insurance based on rule engine and model fusion includes: The processing module 201 is used to receive accident and health insurance claim application data and perform structured processing on the accident and health insurance claim application data to obtain structured claim data; The verification module 202 is used to obtain policy information corresponding to the structured claims data, and to perform validity verification on the structured claims data based on the policy information. The matching module 203 is used to extract case features from the verified structured claims data, input the case features into the rule engine, and obtain the set of claims calculation rules corresponding to the case features by matching them in a preset rule model library. Output module 204 is used to input the structured claims data and the claims calculation rule set into a pre-trained business rule model, and output the core calculation parameters required to determine the compensation amount through the business rule model; The calculation module 205 is used to calculate the compensation amount based on the set of claims rules and the core calculation parameters, and obtain the final compensation amount.

[0117] In one alternative embodiment, the processing module 201 is specifically used for: Receive health insurance claim application data from different data sources, including structured data, semi-structured data and unstructured data; For the unstructured data, text information is extracted using optical character recognition technology; For the semi-structured data, key fields are extracted using a preset parsing template; The text information, key fields, and structured data are integrated in a unified storage format to generate the structured claims data.

[0118] In one alternative embodiment, the verification module 202 is specifically used for: Retrieve the policy information associated with the policy identifier in the structured claims data from the core business system; Verify whether the date of the incident in the structured claims data is within the coverage period based on the coverage period in the policy information; Verify whether the identity information in the structured claims data is consistent with the insured information in the policy information; Verify whether the accident description in the structured claims data falls under the insurance liability based on the scope of liability in the policy information; Verify whether the application date in the structured claims data is within the claim statute of limitations according to the claim statute of limitations provisions in the policy information; Verify whether the medical expense details in the structured claims data fall within the reimbursable scope according to the expense type provisions in the policy information; When all verifications pass, the structured claims data is deemed to have passed the validity verification.

[0119] In an alternative embodiment, the matching module 203 is specifically used for: The insurance type identifier is parsed from the structured claims data that has passed verification; Identify accident description information from the verified structured claims data, and extract accident type tags from the accident description information according to preset accident type classification rules; Traverse the medical expense details from the verified structured claims data, and extract the medical expense type label for each item according to the preset expense category mapping relationship; Read the insured's date of birth from the verified structured claims data, and calculate the insured's age based on the current date and the date of birth; The case features are formed by combining the insurance type identifier, the accident type label, the medical expense type label, and the insured's age.

[0120] In an alternative embodiment, the matching module 203 is specifically used for: The case features are input into the rule engine, which parses the insurance type identifier, accident type label, medical expense type label, and insured age from the case features. Based on the insurance type identifier, the rule group corresponding to the insurance type is located in the rule model library. The rule model library pre-stores the following rule types: reimbursement ratio rule model, deductible rule model, deductible rate rule model, medical expense reimbursement scope rule model, liability exemption rule model, and superimposed reimbursement rule model. Based on the accident type label, the medical expense type label, and the insured's age, at least one rule model matching the case characteristics is traversed and filtered in the rule group. Extract corresponding rule parameters from the at least one rule model and combine them to generate the set of reasoning rules corresponding to the case characteristics.

[0121] In an alternative embodiment, the output module 204 is specifically used for: The structured claims data and the claims calculation rule set are used as input features to the business rule model, which is a machine learning model or rule reasoning model trained based on historical accident and health insurance claims data. The business rule model is used to infer and calculate the input features to generate the cost assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient corresponding to the structured claims data. The reasonableness of the value range of the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient shall be verified respectively. When the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient are all within a preset reasonable range, the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient are combined into the core calculation parameter; when at least one of the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient is not within the preset reasonable range, a verification failure flag is generated, and the structured claims data and the claims calculation rule set are marked as pending manual review.

[0122] In an alternative embodiment, the computing module 205 is specifically used for: Based on the medical expense reimbursement scope rule model in the claim settlement rule set, the details of reimbursable medical expenses are extracted from the structured claims data, and the total amount of reimbursable medical expenses is obtained by summarizing them. Based on the liability exemption rule model in the claim settlement rule set, expenses that fall within the scope of liability exemption are removed from the total amount of reimbursable medical expenses to obtain the verified total amount of medical expenses; Based on the deductible rule model or deductible rate rule model in the settlement rule set, the corresponding deductible amount is deducted from the total amount of verified medical expenses to obtain the base for reimbursement. The base amount to be compensated is calculated by proportional conversion based on the compensation ratio rule model in the compensation rule set. The basic compensation amount is weighted and adjusted based on the cost assessment coefficient, liability determination coefficient and special circumstances adjustment coefficient in the core calculation parameters to obtain the individual liability compensation amount. Identify the number of coverage items contained in the structured claims data; when the structured claims data contains a single coverage item, determine the single coverage item's compensation amount as the final compensation amount. When the structured claims data contains multiple coverage liabilities, the following steps are performed for each coverage liability: extracting details of reimbursable medical expenses from the structured claims data according to the medical expense reimbursement scope rule model in the claims calculation rule set; and then weighting and adjusting the basic compensation amount according to the expense assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient in the core calculation parameters to obtain the individual liability compensation amount corresponding to each coverage liability. Finally, the individual liability compensation amounts corresponding to each coverage liability are merged according to the superimposed compensation rule model in the claims calculation rule set to generate the final compensation amount.

[0123] It should be noted that the beneficial effects of the automatic accident and health insurance claims settlement system 200 based on rule engine and model fusion provided in the above embodiments are the same as those of the automatic accident and health insurance claims settlement method based on rule engine and model fusion, and will not be repeated here. Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiments, and will not be repeated here.

[0124] The automatic claims settlement system 200 for accident and health insurance based on rule engine and model fusion of the present invention can be a computer program (including program code) running on a computer device. For example, the automatic claims settlement system 200 for accident and health insurance based on rule engine and model fusion of the present invention is an application software that can be used to execute the corresponding steps in the automatic claims settlement method for accident and health insurance based on rule engine and model fusion of the present invention.

[0125] In some embodiments, the automatic claims settlement system 200 for health and accident insurance based on rule engine and model fusion of the present invention can be implemented in a combination of hardware and software. As an example, the automatic claims settlement system 200 for health and accident insurance based on rule engine and model fusion of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the automatic claims settlement method for health and accident insurance based on rule engine and model fusion of the present invention. For example, the processor in the form of a hardware decoding processor can be one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0126] The modules described in the embodiments of this invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.

[0127] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned automatic claims settlement methods for health and accident insurance based on rule engine and model fusion. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the automatic claims settlement method for health and accident insurance based on rule engine and model fusion shown in any embodiment of the present invention by calling the computer program.

[0128] In one alternative embodiment, an electronic device is provided, such as Figure 3 As shown, Figure 3 The illustrated electronic device 4000 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0129] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0130] Bus 4002 may include a path for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The bus 4002 is represented by only one thick line, but this does not mean that there is only one bus or one type of bus.

[0131] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0132] The memory 4003 stores application code (computer program) for executing the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0133] Among them, electronic devices can also be terminal devices. A terminal device can be any terminal device that can install applications and access web pages through applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0134] It should be noted that, Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0135] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned automatic claims settlement methods for health and accident insurance based on rule engine and model fusion.

[0136] Alternatively, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0137] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. The processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the aforementioned automatic claims processing method for health and accident insurance based on rule engine and model fusion.

[0138] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0139] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0140] The computer-readable storage medium provided in this invention can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0141] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0142] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0143] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0144] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0145] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for automatic settlement of health insurance based on rule engine and model fusion, characterized in that, include: Receive accident and health insurance claim application data, and perform structured processing on the accident and health insurance claim application data to obtain structured claim data; Obtain the policy information corresponding to the structured claims data, and perform validity verification on the structured claims data based on the policy information; Extract case features from the verified structured claims data, input the case features into the rule engine, and use the rule engine to match the set of claims settlement rules corresponding to the case features in a pre-set rule model library; The structured claims data and the claims calculation rule set are input into a pre-trained business rule model, and the business rule model outputs the core calculation parameters required to determine the compensation amount. Based on the set of claims rules and the core calculation parameters, the compensation amount is calculated on the structured claims data to obtain the final compensation amount.

2. The method according to claim 1, wherein, The steps of structuring the aforementioned health and accident insurance claim data to obtain structured claim data include: Receive health insurance claim application data from different data sources, including structured data, semi-structured data and unstructured data; For the unstructured data, text information is extracted using optical character recognition technology; For the semi-structured data, key fields are extracted using a preset parsing template; The text information, key fields, and structured data are integrated in a unified storage format to generate the structured claims data.

3. The automatic claims settlement method for accident and health insurance based on rule engine and model fusion as described in claim 1, characterized in that, The steps of obtaining policy information corresponding to the structured claims data and validating the structured claims data based on the policy information include: Retrieve the policy information associated with the policy identifier in the structured claims data from the core business system; Verify whether the date of the incident in the structured claims data is within the coverage period based on the coverage period in the policy information; Verify whether the identity information in the structured claims data is consistent with the insured information in the policy information; Verify whether the accident description in the structured claims data falls under the insurance liability based on the scope of liability in the policy information; Verify whether the application date in the structured claims data is within the claim statute of limitations according to the claim statute of limitations provisions in the policy information; Verify whether the medical expense details in the structured claims data fall within the reimbursable scope according to the expense type provisions in the policy information; When all verifications pass, the structured claims data is deemed to have passed the validity verification.

4. The automatic claims settlement method for health and accident insurance based on rule engine and model fusion as described in claim 3, characterized in that, The steps for extracting case features from validated structured claims data include: The insurance type identifier is parsed from the structured claims data that has passed verification; Identify accident description information from the verified structured claims data, and extract accident type tags from the accident description information according to preset accident type classification rules; Traverse the medical expense details from the verified structured claims data, and extract the medical expense type label for each item according to the preset expense category mapping relationship; Read the insured's date of birth from the verified structured claims data, and calculate the insured's age based on the current date and the date of birth; The case features are formed by combining the insurance type identifier, the accident type label, the medical expense type label, and the insured's age.

5. The automatic claims settlement method for health and accident insurance based on rule engine and model fusion as described in claim 4, characterized in that, The steps of inputting the case features into the rule engine and using the rule engine to match the case features in a pre-set rule model library to obtain a set of claims settlement rules corresponding to the case features include: The case features are input into the rule engine, which parses the insurance type identifier, accident type label, medical expense type label, and insured age from the case features. Based on the insurance type identifier, the rule group corresponding to the insurance type is located in the rule model library. The rule model library pre-stores the following rule types: reimbursement ratio rule model, deductible rule model, deductible rate rule model, medical expense reimbursement scope rule model, liability exemption rule model, and superimposed reimbursement rule model. Based on the accident type label, the medical expense type label, and the insured's age, at least one rule model matching the case characteristics is traversed and filtered in the rule group. Extract corresponding rule parameters from the at least one rule model and combine them to generate the set of reasoning rules corresponding to the case characteristics.

6. The automatic claims settlement method for health and accident insurance based on rule engine and model fusion as described in claim 5, characterized in that, The steps of inputting the structured claims data and the claims calculation rule set into a pre-trained business rule model, and outputting the core calculation parameters required to determine the compensation amount through the business rule model, include: The structured claims data and the claims calculation rule set are used as input features to the business rule model, which is a machine learning model or rule reasoning model trained based on historical accident and health insurance claims data. The business rule model is used to infer and calculate the input features to generate the cost assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient corresponding to the structured claims data. The reasonableness of the value range of the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient shall be verified respectively. When the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient are all within a preset reasonable range, the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient are combined into the core calculation parameter; when at least one of the cost assessment coefficient, the liability determination coefficient, and the special circumstances adjustment coefficient is not within the preset reasonable range, a verification failure flag is generated, and the structured claims data and the claims calculation rule set are marked as pending manual review.

7. The automatic claims settlement method for health and accident insurance based on rule engine and model fusion as described in claim 6, characterized in that, The steps for calculating the compensation amount based on the set of claims rules and the core calculation parameters to obtain the final compensation amount include: Based on the medical expense reimbursement scope rule model in the claim settlement rule set, the details of reimbursable medical expenses are extracted from the structured claims data, and the total amount of reimbursable medical expenses is obtained by summarizing them. Based on the liability exemption rule model in the claim settlement rule set, expenses that fall within the scope of liability exemption are removed from the total amount of reimbursable medical expenses to obtain the verified total amount of medical expenses; Based on the deductible rule model or deductible rate rule model in the settlement rule set, the corresponding deductible amount is deducted from the total amount of verified medical expenses to obtain the base for reimbursement. The base amount to be compensated is calculated by proportional conversion based on the compensation ratio rule model in the compensation rule set. The basic compensation amount is weighted and adjusted based on the cost assessment coefficient, liability determination coefficient and special circumstances adjustment coefficient in the core calculation parameters to obtain the individual liability compensation amount. Identify the number of coverage items contained in the structured claims data; when the structured claims data contains a single coverage item, determine the single coverage item's compensation amount as the final compensation amount. When the structured claims data contains multiple coverage liabilities, the following steps are performed for each coverage liability: extracting details of reimbursable medical expenses from the structured claims data according to the medical expense reimbursement scope rule model in the claims calculation rule set; and then weighting and adjusting the basic compensation amount according to the expense assessment coefficient, liability determination coefficient, and special circumstances adjustment coefficient in the core calculation parameters to obtain the individual liability compensation amount corresponding to each coverage liability. Finally, the individual liability compensation amounts corresponding to each coverage liability are merged according to the superimposed compensation rule model in the claims calculation rule set to generate the final compensation amount.

8. An automated claims settlement system for accident and health insurance based on the fusion of rule engine and model, characterized in that, include: The processing module is used to receive accident and health insurance claim application data and perform structured processing on the accident and health insurance claim application data to obtain structured claim data. The verification module is used to obtain policy information corresponding to the structured claims data and to verify the validity of the structured claims data based on the policy information. The matching module is used to extract case features from the verified structured claims data, input the case features into the rule engine, and use the rule engine to match the case features in a pre-set rule model library to obtain a set of claims calculation rules corresponding to the case features. The output module is used to input the structured claims data and the claims calculation rule set into a pre-trained business rule model, and output the core calculation parameters required to determine the compensation amount through the business rule model; The calculation module is used to calculate the compensation amount based on the set of compensation rules and the core calculation parameters, and obtain the final compensation amount.

9. An electronic device, characterized in that, The electronic device includes a processor coupled to a memory, the memory storing at least one computer program, which is loaded and executed by the processor to enable the electronic device to implement the automatic claims settlement method for health and accident insurance based on rule engine and model fusion as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which, when executed by a processor, implements the automatic claims settlement method for health and accident insurance based on rule engine and model fusion as described in any one of claims 1 to 7.

Images