Text generation method and device, electronic equipment and storage medium
By analyzing enterprise quotation documents and generating risk profiles, the problem of relying on human experience for underwriting recommendations has been solved, thus achieving comprehensiveness and accuracy in underwriting recommendations.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA PING AN PROPERTY INSURANCE CO LTD
- Filing Date
- 2026-03-06
- Publication Date
- 2026-06-12
AI Technical Summary
The underwriting recommendations generated by existing technologies rely on human experience, are highly subjective, lack comprehensiveness, and are difficult to fully understand unstructured text information.
By obtaining the text of the target company's inquiry form, the text is parsed to extract fields such as company type, employee job distribution, number of employees, employees working across regions, and existing insurance configuration, to generate a risk profile, and an underwriting recommendation report is generated based on the risk profile.
It improves the comprehensiveness and accuracy of underwriting recommendation reports, reduces the differences in subjective human judgment, and provides objective evaluation standards based on structured data.
Smart Images
Figure CN122197829A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and is applicable to the financial technology field, particularly to a text generation method, apparatus, electronic device, and storage medium. Background Technology
[0002] When receiving a customer's insurance application, insurance companies often need to assess the customer's insurable risks to generate underwriting recommendations. For example, in the context of employer's liability insurance in the fintech sector, for customers (such as companies) applying for employer's liability insurance, an underwriting recommendation needs to be generated. Employer's liability insurance is insurance purchased by companies to protect their liability for compensation arising from work-related injuries or occupational diseases suffered by employees during work hours.
[0003] However, current underwriting recommendations are often not comprehensive enough. For example, they are usually generated based on a client's industry description, which relies on human experience, is highly subjective, and results in rather one-sided recommendations. Summary of the Invention
[0004] The main objective of this application is to provide a text generation method, apparatus, electronic device, and storage medium, which aims to solve the technical problem that underwriting recommendations rely on human experience and are somewhat one-sided, thereby improving the comprehensiveness of the generated underwriting recommendation report.
[0005] To achieve the above objectives, a first aspect of this application proposes a text generation method, the method comprising: Obtain the text of the inquiry form from the target company; wherein the text of the inquiry form is used to instruct the target company to request the purchase of employer's liability insurance; The text of the inquiry form is parsed to obtain the target fields; wherein, the target fields include at least one of the following: enterprise type field, employee job distribution field, number of employees field, employee cross-regional operation field, and existing insurance configuration field; Based on the target fields, generate a risk profile for the target company; Based on the risk profile, a target underwriting recommendation report is generated; wherein, the target underwriting recommendation report is used to underwrite the employer's liability insurance for the target company.
[0006] In some embodiments, the risk profile includes the target risk score of the target enterprise; The process of generating a target underwriting recommendation report based on the risk profile includes: The target risk score is compared with a preset risk score threshold. If the target risk score is less than the risk score threshold, the first deductible value will be determined as the target deductible value for the target enterprise. If the target risk score is greater than or equal to the risk score threshold, the second deductible value is determined as the target deductible value for the target enterprise; wherein the second deductible value is less than the first deductible value; A report is generated based on the target deductible, resulting in the target underwriting recommendation report.
[0007] In some embodiments, generating the target underwriting recommendation report based on the target deductible includes: If the target field includes the enterprise type field, determine the enterprise type protection restriction rule corresponding to the enterprise type field, and determine the enterprise type protection restriction rule as the underwriting restriction rule for the employer's liability insurance; If the target field includes the number of employees field, and the number of employees field is greater than or equal to a preset number of employees threshold, the employee management security restriction rule is added to the underwriting restriction rule; wherein, the employee management security restriction rule is used to determine whether the target enterprise has an employee management security certificate; A report is generated based on the underwriting limitation rules and the target deductible, resulting in the target underwriting recommendation report.
[0008] In some embodiments, generating a target underwriting recommendation report based on the risk profile includes: If the target field includes the employee cross-regional operation field, and the employee cross-regional operation field indicates that the employees of the target enterprise are performing cross-regional operations, then obtain the number of operation areas of the target enterprise. Based on the number of work areas, an additional cross-regional insurance value is generated, and a report is generated based on the additional cross-regional insurance value to obtain the target underwriting recommendation report.
[0009] In some embodiments, generating a risk profile of the target enterprise based on the target field includes: Determine whether the target field includes the target company's historical claims records. If the target field includes the historical claims records, obtain the number of historical claims from the historical claims records. An initial historical claims risk score is determined based on the number of historical claims, and the initial historical claims risk score is used as the risk profile; wherein, the initial historical claims risk score is positively correlated with the number of historical claims.
[0010] In some embodiments, after determining the initial historical claims risk score based on the number of historical claims, the method further includes: If the target field includes the existing insurance configuration field, compare the existing insurance configuration field with the preset basic insurance type; If the existing insurance configuration field does not include the basic insurance type, the initial historical claims risk score is increased to obtain the target historical claims risk score, and the target historical claims risk score is determined as the risk profile.
[0011] In some embodiments, generating a risk profile of the target enterprise based on the target field includes: If the target field includes the employee job distribution field, obtain the target job type in the employee job distribution field and the number of employees for each target job type; The number of employees belonging to the target job type that falls under the preset risk job type is determined as the number of risk job employees; The ratio of the number of employees in the risk positions to the total number of employees in the target company is calculated to obtain the percentage of employees in the risk positions. A risk score is calculated based on the percentage of employees in the risky positions; the risk score is positively correlated with the percentage of employees in the risky positions.
[0012] To achieve the above objectives, a second aspect of this application provides a text generation apparatus, the apparatus comprising: The text acquisition module is used to acquire the text of the inquiry form from the target company; wherein, the text of the inquiry form is used to instruct the target company to request the purchase of employer's liability insurance; The text parsing module is used to parse the text of the inquiry form to obtain target fields; wherein, the target fields include at least one of the following: enterprise type field, employee job distribution field, number of employees field, employee cross-regional operation field, and existing insurance configuration field; The profile generation module is used to generate a risk profile of the target enterprise based on the target fields. The report generation module is used to generate a target underwriting recommendation report based on the risk profile; wherein the target underwriting recommendation report is used to underwrite the employer's liability insurance for the target enterprise.
[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.
[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.
[0015] The text generation method, apparatus, electronic device, and storage medium proposed in this application provide the raw data foundation for generating underwriting recommendations for employer's liability insurance by acquiring the text of a quotation request from a target company. The quotation request text is parsed to obtain target fields, including at least one field such as company type and employee job distribution. This transforms unstructured text information into structured risk assessment indicators, accurately extracting key information for generating a risk profile. A risk profile of the target company is generated based on the target fields. This breaks down the target company's risks into multiple categories, clearly quantifying the risks brought about by different categories of influencing factors (such as company type and employee job distribution), thereby improving the comprehensiveness of the risk profile and, consequently, the comprehensiveness of the target underwriting recommendation report generated based on the risk profile. Attached Figure Description
[0016] Figure 1 This is a flowchart of the text generation method provided in the embodiments of this application; Figure 2 This is an embodiment provided by this application. Figure 1 The flowchart for step 103 in the text; Figure 3 This is a flowchart of a text generation method provided in another embodiment of this application; Figure 4 This is provided by another embodiment of the present application. Figure 1 The flowchart for step 103 in the text; Figure 5 This is an embodiment provided by this application. Figure 1 The flowchart for step 104 in the document; Figure 6 yes Figure 5 The flowchart for step 504 in the document; Figure 7 This is provided by another embodiment of the present application. Figure 1 The flowchart for step 104 in the document; Figure 8 This is a flowchart illustrating an application example provided in an embodiment of this application; Figure 9 This is a schematic diagram of the structure of the text generation device provided in the embodiments of this application; Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation
[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0018] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.
[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0020] First, let's analyze some of the terms used in this application: Artificial Intelligence (AI) is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can be a simulation of the information processes of human consciousness and thought. AI can also be the theory, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results. Basic AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operating / interactive systems, and mechatronics. AI software technologies mainly include computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning. This application can acquire and process relevant data based on AI technology.
[0021] Natural Language Processing (NLP) is an important research area in the field of artificial intelligence. It integrates knowledge from multiple disciplines such as linguistics, computer science, machine learning, mathematics, and cognitive psychology. NLP aims to enable machines to understand, interpret, and generate human language, achieve effective communication between humans and machines, and enable computers to perform tasks such as language translation, sentiment analysis, and text summarization.
[0022] Employer's liability insurance is insurance purchased by companies to protect their liability for compensation arising from work-related injuries or occupational diseases suffered by employees during work hours. When an insurance company receives an application for employer's liability insurance from a client (i.e., a company), it typically needs to assess the client's insurable risks in order to generate an underwriting recommendation for the employer's liability insurance.
[0023] However, current underwriting recommendations are often incomplete. For example, they are typically generated based on client industry descriptions, a method heavily reliant on human experience, highly subjective, and inefficient. Furthermore, current underwriting recommendation generation methods lack interpretability and struggle to fully understand unstructured text information. Therefore, the generated underwriting recommendations are rather one-sided.
[0024] Based on this, embodiments of this application provide a text generation method, apparatus, electronic device, and storage medium, which can improve the comprehensiveness of the generated underwriting recommendation report.
[0025] The text generation method, apparatus, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the text generation method in this application is described.
[0026] The text generation method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms; the software can be an application that implements the text generation method, but is not limited to the above forms.
[0027] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
[0028] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.
[0029] Figure 1 This is an optional flowchart of the text generation method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps 101 to 104.
[0030] Step 101: Obtain the text of the inquiry form from the target company; wherein, the text of the inquiry form is used to instruct the target company to request the purchase of employer's liability insurance; Step 102: Parse the inquiry form text to obtain the target fields; wherein, the target fields include at least one of the following: enterprise type field, employee job distribution field, number of employees field, employee cross-regional operation field, and existing insurance configuration field; Step 103: Generate a risk profile of the target company based on the target fields; Step 104: Generate a target underwriting recommendation report based on the risk profile; the target underwriting recommendation report is used to underwrite employer's liability insurance for the target company.
[0031] The beneficial effects of this application's embodiments include, but are not limited to: providing a raw data foundation for generating underwriting recommendations for employer's liability insurance by obtaining the text of the target company's inquiry form. Text parsing of the inquiry form text yields target fields, including at least one field such as company type and employee job distribution. This transforms unstructured text information into structured risk assessment indicators, accurately extracting key information for generating a risk profile. Generating a risk profile of the target company based on the target fields breaks down the target company's risks into multiple categories, clearly quantifying the risks brought about by different categories of influencing factors (such as company type and employee job distribution), thereby improving the comprehensiveness of the risk profile and, consequently, the comprehensiveness of the target underwriting recommendation report generated based on the risk profile.
[0032] In step 101 of some embodiments, the target company is the company applying for employer's liability insurance. It should be noted that employer's liability insurance is insurance purchased by a company to protect its employees from liability for compensation arising from work-related injuries or occupational diseases during work hours. The request for quotation is a document instructing the target company to request the purchase of employer's liability insurance. The request for quotation can be unstructured text. For example, when requesting employer's liability insurance, the target company can fill out a request for quotation and send it to an insurance company providing employer's liability insurance.
[0033] In step 102 of some embodiments, the target field is structured text parsed from the inquiry form text. In some embodiments, the inquiry form text can be parsed using a natural language processing model (such as a large language model) to obtain the target field. Other methods of text parsing are also possible, and are not limited to these.
[0034] In some embodiments, the target field may include at least one of the following: a business type field, an employee job distribution field, a number of employees field, an employee cross-regional work field, and an existing insurance configuration field. The business type field represents the type of the target business, such as hazardous materials transportation, logistics warehousing, manufacturing, internet, logistics, construction, education, etc. The employee job distribution field represents the job type of each employee in the target business, such as frontline operator, driver, customer service representative, programmer, management, etc. The number of employees field represents the total number of employees in the target business. The employee cross-regional work field represents whether the employees of the target business work in different geographical areas. The existing insurance configuration field represents the current insurance coverage of the target business, such as no insurance record, only accident insurance, accident insurance and employer's liability insurance, etc.
[0035] In step 103 of some embodiments, the risk profile is the result of a multi-dimensional description and analysis of the risk characteristics of the target enterprise. Specifically, the risk profile may include a risk score for the target enterprise.
[0036] In step 104 of some embodiments, the target underwriting recommendation report is a recommendation report made by the insurance company after assessing the target company's application for employer's liability insurance, regarding whether to underwrite, how to underwrite (including the sum insured, premium, deductible, etc.), and risk control measures.
[0037] Please see Figure 2 In some embodiments, step 103 may include, but is not limited to, steps 201 to 202: Step 201: Determine whether the target field includes the target company's historical claims records. If the target field includes historical claims records, obtain the number of historical claims from the historical claims records. Step 202: Determine the initial historical claim risk score based on the number of historical claims, and define the initial historical claim risk score as the risk profile; wherein, the initial historical claim risk score is positively correlated with the number of historical claims.
[0038] The advantage of this embodiment lies in its ability to identify and extract a company's historical claims data by determining whether the target field includes historical claims records and, if so, obtaining the number of historical claims. This provides foundational information for risk quantification. Then, an initial historical claims risk score is determined based on the number of historical claims, which is then used to create a risk profile. This quantifies the risk posed by the historical claims frequency as a contributing factor, thereby improving the comprehensiveness of the risk profile and ultimately enhancing the comprehensiveness of the target underwriting recommendation report generated based on the risk profile.
[0039] In step 201 of some embodiments, historical claims records are used to indicate the target company's claims history prior to generating the quotation text. Specifically, historical claims records may include the number of historical claims. In some embodiments, the number of historical claims may be the number of claims made by the target company within a preset time period. For example, the number of historical claims may be the number of claims made by the target company within the past year.
[0040] In step 202 of some embodiments, the initial historical claims risk score is positively correlated with the number of historical claims. For example, the number of historical claims refers to the number of claims made by the target company in the past year.
[0041] In some embodiments, a preset table showing the relationship between historical claim counts and risk scores can be consulted to obtain the score corresponding to the historical claim count, which serves as the initial historical claim risk score. For example, assuming there are 0 historical claims, the initial historical claim risk score could be 0. Assuming there is 1 historical claim, the initial historical claim risk score could be 20. Assuming there are 2 historical claims, the initial historical claim risk score could be 50. Assuming there are 3 or more historical claims, the initial historical claim risk score could be 80.
[0042] Please see Figure 3 In some embodiments, after step 202, the text generation method may also include, but is not limited to, steps 301 to 302: Step 301: If the target field includes an existing insurance configuration field, compare the existing insurance configuration field with the preset basic insurance type; Step 302: If the existing insurance configuration fields do not include the basic insurance type, increase the initial historical claims risk score to obtain the target historical claims risk score, and determine the target historical claims risk score as the risk profile.
[0043] The advantage of this embodiment lies in that, by comparing the existing insurance configuration fields with the preset basic insurance types when the target field includes them, it can verify whether the company's configured insurance covers the basic risk types and identify protection gaps. If the existing insurance configuration fields do not include the basic insurance types, the initial historical claims risk score is increased to obtain the target historical claims risk score. This dynamically reflects the company's insufficient coverage as a risk factor in the risk score, thereby considering the company's existing insurance for comprehensive risk analysis, improving the comprehensiveness of the risk profile, and ultimately improving the comprehensiveness of the target underwriting recommendation report generated based on the risk profile.
[0044] In step 301 of some embodiments, the preset basic insurance type may include any one of the following insurance types: accident insurance, employer's liability insurance, etc.
[0045] In step 302 of some embodiments, for example, if the basic insurance type includes accident insurance, but the existing insurance configuration field does not include accident insurance, this indicates that the target company has not purchased accident insurance. The initial historical claims risk score is then increased, for example, by 30 points.
[0046] In some embodiments, if the insurance configuration field cannot be extracted from the target company's inquiry text, that is, if the target company has no insurance record, an underwriting warning can be triggered to alert the insurance company that the company is a high-risk customer.
[0047] Please see Figure 4 In some embodiments, step 103 may include, but is not limited to, steps 401 to 404: Step 401: If the target field includes the employee job distribution field, obtain the target job type and the number of employees for each target job type from the employee job distribution field. Step 402: Determine the number of employees belonging to the target job type of the preset risk job type as the number of risk job employees; Step 403: Calculate the ratio of the number of employees in risky positions to the total number of employees in the target company to obtain the percentage of employees in risky positions; Step 404: Calculate the risk score based on the proportion of employees in risky positions to obtain the position risk score; where the position risk score is positively correlated with the proportion of employees in risky positions.
[0048] The advantage of this embodiment lies in its ability to accurately extract specific data on the composition of an enterprise's employee positions by obtaining the target position types and the number of employees in each target position type when the target field includes an employee position distribution field. This provides a data basis for risk assessment. Then, the number of employees belonging to preset risk position types is determined as the number of employees in risk positions, and this number is compared with the total number of employees to obtain the proportion of employees in risk positions. This transforms position distribution into a quantifiable risk indicator. A risk score is calculated based on the proportion of employees in risk positions, and the score is positively correlated with the proportion. This allows for the analysis of the enterprise's risk level based on the proportion of high-risk positions, thereby improving the comprehensiveness of the risk profile and, consequently, the comprehensiveness of the target underwriting recommendation report generated based on the risk profile.
[0049] In step 401 of some embodiments, the target job type refers to the job type of an employee in the target enterprise. The target job type may include, but is not limited to, the following job types: front-line operators, high-altitude workers, hazardous chemical transport drivers, customer service representatives, programmers, management, etc.
[0050] In step 402 of some embodiments, the risky job type may include front-line operators, high-altitude workers, and hazardous materials transport drivers. In some embodiments, the number of employees in risky jobs may be the sum of the number of employees in multiple target job types belonging to the risky job type. For example, suppose the target job types of a target enterprise include front-line operators, hazardous materials transport drivers, and customer service personnel, with N1 front-line operators, N2 hazardous materials transport drivers, and N3 customer service personnel. Since front-line operators and hazardous materials transport drivers belong to the preset risky job types, the number of employees in risky jobs is (N1 + N2).
[0051] In step 403 of some embodiments, the number of employees in risky positions can be used as the numerator and the total number of employees can be used as the denominator to calculate the ratio and obtain the proportion of employees in risky positions.
[0052] In step 404 of some embodiments, for example, assuming the percentage of employees in high-risk positions is 0%, the position risk score can be 0. Assuming the percentage of employees in high-risk positions is 1%-20%, the position risk score can be 20. Assuming the percentage of employees in high-risk positions is 21%-50%, the position risk score can be 50. Assuming the percentage of employees in high-risk positions is 51%-80%, the position risk score can be 75. Assuming the percentage of employees in high-risk positions is greater than 80%, the position risk score can be 100.
[0053] Please see Figure 5 In some embodiments, the risk profile includes the target enterprise's target risk score; Step 104 may include, but is not limited to, steps 501 through 504: Step 501: Compare the target risk score with the preset risk score threshold; Step 502: If the target risk score is less than the risk score threshold, the first deductible value is determined as the target deductible value for the target company. Step 503: If the target risk score is greater than or equal to the risk score threshold, the second deductible value is determined as the target deductible value for the target company; wherein the second deductible value is less than the first deductible value; Step 504: Generate a report based on the target deductible to obtain the target underwriting recommendation report.
[0054] The advantage of this embodiment lies in that by comparing the target risk score with a preset risk score threshold, the enterprise's risk status can be transformed into an objective and unified assessment standard, reducing the differences in subjective human judgment and providing a precise basis for subsequent decision-making. Then, a target deductible is determined based on the comparison results. Specifically, a higher first deductible is used when the target risk score is less than the risk score threshold, and a lower second deductible is used when the target risk score is greater than or equal to the risk score threshold. This ensures that the deductible setting matches the actual risk level of the target enterprise. An underwriting recommendation report is then generated based on the determined target deductible, which improves the reasonableness, personalization, and comprehensiveness of the underwriting recommendation report.
[0055] In step 501 of some embodiments, the target risk score may be compared to a preset risk score threshold.
[0056] In step 502 of some embodiments, if the target risk score is less than the risk score threshold, it indicates that the user's risk is low.
[0057] In step 503 of some embodiments, if the target risk score is greater than or equal to the risk score threshold, it indicates that the user has a high risk. The second deductible is a deductible that is less than the first deductible.
[0058] In step 504 of some embodiments, the target underwriting recommendation report may include a target deductible. In some embodiments, the risk level of a target enterprise may be classified based on a target risk score and a risk score threshold. For example, a target enterprise with a target risk score less than or equal to 30 may be identified as a low-risk enterprise; a target enterprise with a target risk score greater than 30 and less than or equal to 65 may be identified as a medium-risk enterprise; and a target enterprise with a target risk score greater than 65 may be identified as a high-risk enterprise.
[0059] Please see Figure 6 In some embodiments, step 504 may include, but is not limited to, steps 601 to 603: Step 601: If the target field includes an enterprise type field, determine the enterprise type protection restriction rule corresponding to the enterprise type field, and determine the enterprise type protection restriction rule as the underwriting restriction rule for employer's liability insurance; Step 602: If the target field includes the number of employees field, and the number of employees field is greater than or equal to the preset number of employees threshold, add the employee management security restriction rule to the underwriting restriction rule; wherein, the employee management security restriction rule is used to determine whether the target company has an employee management security certificate; Step 603: Generate a report based on the underwriting limitation rules and the target deductible to obtain the target underwriting recommendation report.
[0060] The advantage of this embodiment lies in that, when the target field includes a company type field, it determines the corresponding company type protection restriction rules (e.g., for construction companies, determining a restriction rule such as "the company needs to purchase high-risk operation supplementary insurance") and uses it as the underwriting restriction rule. This allows for differentiated risk management measures to be applied to different types of companies, making the underwriting recommendations more aligned with the company's specific risks. Furthermore, if the target field includes a number of employees field, and the number of employees is greater than or equal to a preset employee number threshold, then employee management safety restriction rules are added to the underwriting restriction rules. This automatically triggers stricter safety qualification review requirements for companies with larger workforces, thereby achieving targeted control over key risk factors. Then, the underwriting restriction rules and the target deductible are combined to generate a report. This ensures that the generated target underwriting recommendation report not only includes a deductible scheme based on risk score but also multi-dimensional underwriting conditions matched to the company type and size, thus improving the comprehensiveness of the underwriting recommendation report.
[0061] In step 601 of some embodiments, for example, assuming the enterprise type field is manufacturing, the enterprise type protection restriction rule corresponding to this enterprise type field could be to suggest adding occupational disease protection. Assuming the enterprise type field is construction or building, the enterprise type protection restriction rule corresponding to this enterprise type field could be to suggest purchasing additional high-risk operation insurance, or to suggest increasing the accidental death limit, etc.
[0062] In some embodiments, supplementary insurance recommendations can be generated based on the target company's business type. For example, assuming the company is in manufacturing, chemical, or mining, the supplementary insurance recommendations could include a recommendation to purchase occupational disease coverage. As another example, assuming the company is in the internet, finance, or service industries, the supplementary insurance recommendations could include a recommendation to purchase mental health coverage. Yet another example, assuming the company is in the field operations or transportation industry, the supplementary insurance recommendations could include a recommendation to purchase emergency rescue service coverage.
[0063] In some embodiments, targeted suggestions can be generated based on whether the target enterprise's target job types include special job types (such as drone operators, AI trainers, etc.), such as suggestions to add a sub-item for emerging job liability insurance. It can also be suggested to clarify the scope of coverage stipulated in the employer's liability insurance contract, i.e., the scope of the insured object, such as whether it includes interns, part-time workers, outsourced workers, etc.
[0064] In step 602 of some embodiments, specifically, the preset employee number threshold can be 500. The employee number threshold can also be set or adjusted to other values, and this application embodiment does not limit this. For example, the employee management safety certificate can be the enterprise's safety production management system certification.
[0065] In step 603 of some embodiments, the target underwriting recommendation report may include underwriting limitation rules and a target deductible.
[0066] Please see Figure 7 In some embodiments, step 104 may include, but is not limited to, steps 701 to 702: Step 701: If the target field includes the "Employee Cross-Regional Operation" field, and the "Employee Cross-Regional Operation" field represents that the employees of the target company are performing cross-regional operations, obtain the number of operation areas of the target company. Step 702: Generate cross-regional supplementary insurance value based on the number of work areas, and generate a report based on the cross-regional supplementary insurance value to obtain the target underwriting recommendation report.
[0067] The advantage of this embodiment is that, when generating the report, if the target field includes an "Employee Cross-Regional Work" field, and this field indicates that employees are performing cross-regional work, the number of operating regions for the target company is obtained. This allows for the identification and quantification of potential risks arising from the complexity of business geographical distribution. Then, an additional cross-regional insurance value is generated based on the number of operating regions, and a target underwriting recommendation report is generated based on this value. This transforms the specific risk factor of cross-regional work into concrete insurance premium adjustment parameters, taking into account the risks associated with employees performing cross-regional work, thereby improving the comprehensiveness of the underwriting recommendation report.
[0068] In step 701 of some embodiments, the number of work areas may be greater than or equal to two. For example, assuming the employee cross-regional work field represents employees of the target company performing cross-regional work, and the target company's work areas include area A1, area A2, and area A3, then the number of work areas is three.
[0069] In step 702 of some embodiments, the target underwriting recommendation report includes an additional cross-regional insurance value. In some embodiments, the additional cross-regional insurance value may be positively correlated with the number of operating areas.
[0070] Please see Figure 8 In one application example, the text generation method includes the following steps: The system receives inquiry texts from clients (i.e., target companies); it then performs structured parsing of the inquiry text using a large-scale model, extracting fields such as client type, employee positions, number of employees, work location, and historical claims records; based on these fields, it builds a risk profile, which may include a comprehensive risk score and risk type classification; it calls the underwriting strategy library and the large-scale language model (referred to as the large-scale model) to perform underwriting recommendation reasoning on the risk profile, generating underwriting recommendations; these recommendations may include suggestions based on client type, employee positions, supplementary insurance, deductibles, and insurance limits; and it outputs a structured underwriting recommendation report and a visual underwriting recommendation card. In another application example, the system can also choose to push the underwriting recommendations to the insurance company's underwriting system or generate a quotation.
[0071] Please see Figure 9 This application also provides a text generation apparatus that can implement the above-described text generation method. The apparatus includes: The text acquisition module 801 is used to acquire the text of the inquiry form from the target company; wherein, the text of the inquiry form is used to instruct the target company to request the purchase of employer's liability insurance; The text parsing module 802 is used to parse the inquiry form text to obtain the target fields; wherein, the target fields include at least one of the following: enterprise type field, employee position distribution field, number of employees field, employee cross-regional operation field, and existing insurance configuration field; The profile generation module 803 is used to generate a risk profile of the target enterprise based on the target fields. The report generation module 804 is used to generate a target underwriting recommendation report based on the risk profile; the target underwriting recommendation report is used to underwrite employer's liability insurance for the target company.
[0072] In one embodiment, the text generation device further includes a claims risk generation module, configured to: if the target field includes an existing insurance configuration field, compare the existing insurance configuration field with a preset basic insurance type; if the existing insurance configuration field does not include a basic insurance type, increase the initial historical claims risk score to obtain a target historical claims risk score, and determine the target historical claims risk score as a risk profile.
[0073] The specific implementation of this text generation device is basically the same as the specific embodiment of the text generation method described above, and will not be repeated here.
[0074] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described text generation method. This electronic device can include any smart terminal such as a tablet computer or an in-vehicle computer.
[0075] Please see Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 902 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the text generation method of the embodiments of this application. The input / output interface 903 is used to implement information input and output; The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904); The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.
[0076] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described text generation method.
[0077] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0078] It should be noted that the software tools or components not belonging to our company that appear in the embodiments of this application are merely examples and do not represent actual use.
[0079] The embodiments described in this application are intended to more clearly illustrate the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.
[0080] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.
[0081] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0082] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.
[0083] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0084] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0085] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. The coupling or direct coupling or communication connection between the shown or discussed units may be through some interfaces, or indirect coupling or communication connection between the apparatus or units, and may be electrical, mechanical, or other forms.
[0086] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0087] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0088] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0089] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.
Claims
1. A text generation method, characterized in that, The method includes: Obtain the text of the inquiry form from the target company; wherein the text of the inquiry form is used to instruct the target company to request the purchase of employer's liability insurance; The text of the inquiry form is parsed to obtain the target fields; wherein, the target fields include at least one of the following: enterprise type field, employee job distribution field, number of employees field, employee cross-regional operation field, and existing insurance configuration field; Based on the target fields, generate a risk profile for the target company; Based on the risk profile, a target underwriting recommendation report is generated; wherein, the target underwriting recommendation report is used to underwrite the employer's liability insurance for the target company.
2. The method according to claim 1, characterized in that, The risk profile includes the target risk score of the target enterprise; The process of generating a target underwriting recommendation report based on the risk profile includes: The target risk score is compared with a preset risk score threshold. If the target risk score is less than the risk score threshold, the first deductible value will be determined as the target deductible value for the target enterprise. If the target risk score is greater than or equal to the risk score threshold, the second deductible value is determined as the target deductible value for the target enterprise; wherein the second deductible value is less than the first deductible value; A report is generated based on the target deductible, resulting in the target underwriting recommendation report.
3. The method according to claim 2, characterized in that, The process of generating a report based on the target deductible to obtain the target underwriting recommendation report includes: If the target field includes the enterprise type field, determine the enterprise type protection restriction rule corresponding to the enterprise type field, and determine the enterprise type protection restriction rule as the underwriting restriction rule for the employer's liability insurance; If the target field includes the number of employees field, and the number of employees field is greater than or equal to a preset number of employees threshold, the employee management security restriction rule is added to the underwriting restriction rule; wherein, the employee management security restriction rule is used to determine whether the target enterprise has an employee management security certificate; A report is generated based on the underwriting limitation rules and the target deductible, resulting in the target underwriting recommendation report.
4. The method according to any one of claims 1 to 3, characterized in that, The process of generating a target underwriting recommendation report based on the risk profile includes: If the target field includes the employee cross-regional operation field, and the employee cross-regional operation field indicates that the employees of the target enterprise are performing cross-regional operations, then obtain the number of operation areas of the target enterprise. Based on the number of work areas, an additional cross-regional insurance value is generated, and a report is generated based on the additional cross-regional insurance value to obtain the target underwriting recommendation report.
5. The method according to any one of claims 1 to 3, characterized in that, The step of generating a risk profile for the target enterprise based on the target field includes: Determine whether the target field includes the target company's historical claims records. If the target field includes the historical claims records, obtain the number of historical claims from the historical claims records. An initial historical claims risk score is determined based on the number of historical claims, and the initial historical claims risk score is used as the risk profile; wherein, the initial historical claims risk score is positively correlated with the number of historical claims.
6. The method according to claim 5, characterized in that, After determining the initial historical claims risk score based on the number of historical claims, the method further includes: If the target field includes the existing insurance configuration field, compare the existing insurance configuration field with the preset basic insurance type; If the existing insurance configuration field does not include the basic insurance type, the initial historical claims risk score is increased to obtain the target historical claims risk score, and the target historical claims risk score is determined as the risk profile.
7. The method according to any one of claims 1 to 3, characterized in that, The step of generating a risk profile for the target enterprise based on the target field includes: If the target field includes the employee job distribution field, obtain the target job type in the employee job distribution field and the number of employees for each target job type; The number of employees belonging to the target job type that falls under the preset risk job type is determined as the number of risk job employees; The ratio of the number of employees in the risk positions to the total number of employees in the target company is calculated to obtain the percentage of employees in the risk positions. A risk score is calculated based on the percentage of employees in the risky positions; the risk score is positively correlated with the percentage of employees in the risky positions.
8. A text generation device, characterized in that, The device includes: The text acquisition module is used to acquire the text of the inquiry form from the target company; wherein, the text of the inquiry form is used to instruct the target company to request the purchase of employer's liability insurance; The text parsing module is used to parse the text of the inquiry form to obtain target fields; wherein, the target fields include at least one of the following: enterprise type field, employee job distribution field, number of employees field, employee cross-regional operation field, and existing insurance configuration field; The profile generation module is used to generate a risk profile of the target enterprise based on the target fields. The report generation module is used to generate a target underwriting recommendation report based on the risk profile; wherein the target underwriting recommendation report is used to underwrite the employer's liability insurance for the target enterprise.
9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the text generation method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the text generation method according to any one of claims 1 to 7.