Intelligent underwriting method and device based on knowledge graph reasoning, equipment and medium

Abstract

The present disclosure provides an intelligent underwriting method and device based on knowledge graph reasoning, equipment and medium, comprising: obtaining health characterization data of a target entity object; performing medical reasoning analysis on the health characterization data through a medical knowledge graph to obtain a medical evolution factor sequence; determining a medical reasoning factor from the medical evolution factor sequence according to the health characterization data and the insurance time information of the target entity object; performing intelligent underwriting analysis on the health characterization data to obtain intelligent underwriting information; determining the dynamic evolution probability of the target entity object corresponding to the medical reasoning factor according to the historical evolution data between the health characterization data and the medical reasoning factor; and performing information optimization on the intelligent underwriting information corresponding to the target entity object based on the dynamic evolution probability of the target entity object corresponding to the medical reasoning factor to obtain target underwriting information corresponding to the target entity object. Thus, the accuracy of intelligent underwriting is effectively improved.

Description

Technical Field

[0001] The embodiments of this disclosure relate to the field of intelligent underwriting technology, and more specifically, to an intelligent underwriting method, apparatus, device, and medium applicable to knowledge graph-based deduction. Background Technology

[0002] Underwriting protects the interests of both insurance companies and policyholders while ensuring the fairness and legality of insurance transactions. It also helps insurance companies accurately assess risk levels and prevent the accumulation of adverse risks. Therefore, when purchasing insurance, policyholders should understand and cooperate with the underwriting process, providing truthful and accurate personal information to obtain appropriate insurance coverage.

[0003] In related technologies, most existing underwriting systems adopt a layered model of database + platform layer + application layer. The data layer integrates multi-source data such as policyholder historical data, medical records, and financial information; the platform layer includes a rule engine and AI (Artificial Intelligence) models to support automated decision-making; and the application layer uses technologies such as OCR (Optical Character Recognition) to realize an interactive underwriting process.

[0004] However, the accuracy of underwriting using existing methods is not high. Summary of the Invention

[0005] The embodiments described herein provide an intelligent underwriting method, apparatus, device, and medium based on knowledge graph inference, which overcomes the above-mentioned problems.

[0006] Firstly, based on the content of this disclosure, an intelligent underwriting method based on knowledge graph inference is provided, including:

[0007] Obtain health representation data of the target entity object, wherein the health representation data of the target entity object is obtained by filtering from the historical medical information of the target entity object;

[0008] By using a medical knowledge graph to perform medical extrapolation and analysis on the health representation data of the target entity, a sequence of medical evolution factors associated with the health representation data of the target entity is obtained.

[0009] Obtain the insurance application time information of the target entity object, and based on the insurance application time information and the health characterization data of the target entity object, determine the medical inference factor corresponding to the insurance application time information of the target entity object from the medical evolution factor sequence;

[0010] Intelligent underwriting analysis is performed on the health characterization data of the target entity to obtain intelligent underwriting information corresponding to the target entity.

[0011] Obtain the health representation data of the target entity object and the historical evolution data between the medical inference factors, and determine the dynamic evolution probability of the target entity object corresponding to the medical inference factors based on the historical evolution data between the health representation data of the target entity object and the medical inference factors.

[0012] Based on the dynamic evolution probability of the target entity object corresponding to the medical inference factor, the intelligent underwriting information corresponding to the target entity object is optimized to obtain the target underwriting information corresponding to the target entity object.

[0013] Secondly, according to the content of this disclosure, an intelligent underwriting device based on knowledge graph inference is provided, comprising:

[0014] The acquisition module is used to acquire health representation data of the target entity object, wherein the health representation data of the target entity object is obtained by filtering from the historical medical information of the target entity object;

[0015] The first analysis module is used to perform medical extrapolation analysis on the health representation data of the target entity object through a medical knowledge graph, and obtain the sequence of medical evolution factors associated with the health representation data of the target entity object.

[0016] The first determining module is used to obtain the insurance time information of the target entity object, and determine the medical inference factor corresponding to the insurance time information of the target entity object from the medical evolution factor sequence based on the insurance time information and the health characterization data of the target entity object;

[0017] The second analysis module is used to perform intelligent underwriting analysis on the health characterization data of the target entity object to obtain the intelligent underwriting information corresponding to the target entity object.

[0018] The second determining module is used to acquire the health characterization data of the target entity object and the historical evolution data between the medical inference factors, and to determine the dynamic evolution probability of the target entity object corresponding to the medical inference factors based on the historical evolution data between the health characterization data of the target entity object and the medical inference factors.

[0019] The optimization module is used to optimize the intelligent underwriting information corresponding to the target entity object based on the dynamic evolution probability of the target entity object corresponding to the medical inference factor, so as to obtain the target underwriting information corresponding to the target entity object.

[0020] Thirdly, a computer device is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the intelligent underwriting method based on knowledge graph inference as described in any of the above embodiments.

[0021] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the intelligent underwriting method based on knowledge graph inference as described in any of the above embodiments.

[0022] The intelligent underwriting method based on knowledge graph deduction provided in this application embodiment obtains health representation data of a target entity object, which is filtered from the target entity object's historical medical information; performs medical deduction analysis on the health representation data of the target entity object through a medical knowledge graph to obtain a sequence of medical evolution factors associated with the health representation data of the target entity object; obtains the insurance application time information of the target entity object, and determines the medical deduction factor mapped to the insurance application time information of the target entity object from the sequence of medical evolution factors based on the insurance application time information and the health representation data of the target entity object; performs intelligent underwriting analysis on the health representation data of the target entity object to obtain intelligent underwriting information corresponding to the target entity object; obtains historical evolution data between the health representation data of the target entity object and the medical deduction factors, and determines the dynamic evolution probability of the target entity object corresponding to the medical deduction factors based on the historical evolution data between the health representation data of the target entity object and the medical deduction factors; and optimizes the intelligent underwriting information corresponding to the target entity object based on the dynamic evolution probability of the target entity object corresponding to the medical deduction factors to obtain the target underwriting information corresponding to the target entity object. In this way, by performing knowledge graph deduction on the health representation data of the target entity, medical deduction factors with similar development trends to the health representation data are obtained. Based on the historical evolution data between the medical deduction factors and the health representation data, the preliminary intelligent underwriting information is optimized to obtain the target underwriting information, effectively improving the accuracy of intelligent underwriting.

[0023] The above description is merely an overview of the technical solutions of the embodiments of this application. In order to better understand the technical means of the embodiments of this application and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of this application more obvious and understandable, specific implementation methods of this application are described below. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments will be briefly described below. It should be understood that the drawings described below only relate to some embodiments of this disclosure and are not intended to limit this disclosure, wherein:

[0025] Figure 1 This is a flowchart illustrating an intelligent underwriting method based on knowledge graph inference, which is disclosed in this publication.

[0026] Figure 2 This is a schematic diagram of the structure of an intelligent underwriting device based on knowledge graph deduction provided in this disclosure.

[0027] Figure 3 This is a schematic diagram of the structure of a computer device provided in this disclosure.

[0028] It should be noted that the elements in the attached diagram are schematic and not drawn to scale. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the described embodiments of this disclosure without creative effort are also within the scope of protection of this disclosure.

[0030] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this subject matter pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having the meaning consistent with their meaning in the context of the specification and in the relevant art, and shall not be interpreted in an idealized or overly formal form unless otherwise explicitly defined herein. As used herein, the statement of “connecting” or “coupling” two or more parts together shall mean that these parts are directly joined together or joined through one or more intermediate components.

[0031] The term "embodiment" as used herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of the phrase "embodiment" in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0032] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists, A and B exist simultaneously, or B exists. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Terms such as "first" and "second" are only used to distinguish one component (or part of a component) from another component (or another part of a component).

[0033] In the description of this application, unless otherwise stated, "multiple" means two or more (including two), and similarly, "multiple groups" means two or more (including two groups).

[0034] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0035] Figure 1 This is a flowchart illustrating an intelligent underwriting method based on knowledge graph inference provided in this disclosure embodiment, such as... Figure 1 As shown, the specific process of the intelligent underwriting method based on knowledge graph inference includes:

[0036] S110. Obtain the health representation data of the target entity object.

[0037] The target entity can be an insured living being or a part of its limb structure, such as a female / male, an animal, or a female's hand. The health profile data of the target entity is derived from its historical medical records. If the target entity is a user with a pre-existing condition, its health profile data can be derived from the user's health record.

[0038] S120. Through medical knowledge graph, perform medical extrapolation analysis on the health representation data of the target entity object to obtain the sequence of medical evolution factors associated with the health representation data of the target entity object.

[0039] Among them, medical knowledge graphs can integrate and analyze large amounts of medical data as well as integrate scattered medical data and literature, and build knowledge graphs of the health status of patients for different patients. In public health event scenarios such as epidemiological investigation and analysis and early warning of epidemic events, knowledge graphs can intuitively represent information such as personnel distribution, activity trajectory, and onset time, which helps to analyze and sort out the source of infection.

[0040] The medical evolution factor sequence includes multiple medical evolution factors, which can be diseases that develop over time from the health representation data of the target entity, such as pulmonary nodules, pneumonia, bronchial asthma, and emphysema.

[0041] In some embodiments, a medical extrapolation analysis is performed on the health representation data of the target entity object using a medical knowledge graph to obtain a sequence of medical evolution factors associated with the health representation data of the target entity object, including:

[0042] Based on the health representation data of the target entity object, multiple related evolution factors are matched from the medical knowledge graph, and the evolution fusion value between the health representation data of the target entity object and each related evolution factor is calculated. Based on the evolution fusion value between the health representation data of the target entity object and each related evolution factor, multiple node evolution factors related to the health representation data of the target entity object are determined. Based on the multiple node evolution factors related to the health representation data of the target entity object, a sequence of medical evolution factors associated with the health representation data of the target entity object is generated.

[0043] Specifically, if the health representation data of the target entity object is associated with a certain evolution factor in the medical knowledge graph, such as the health representation data being "cough", then it is associated with evolution factor A (e.g., "emphysema"), evolution factor B (e.g., "pneumonia"), and evolution factor C (e.g., "bronchitis").

[0044] The evolution fusion value between the health representation data of a target entity and its associated evolution factors is the probability that the target entity will evolve into an associated evolution factor based on the health representation data. For example, the evolution fusion value between the health representation data of a target entity and evolution factor A is probability 1, the evolution fusion value between the health representation data of a target entity and evolution factor B is probability 2, and the evolution fusion value between the health representation data of a target entity and evolution factor C is probability 3. Probabilities 1 and 2 are both greater than 50%, and probability 3 is less than 50%. In this case, the multiple node evolution factors related to the health representation data of the target entity are evolution factors A and B, and the medical evolution factor sequence is [evolution factor A, evolution factor B].

[0045] Therefore, by using factor matching in the medical knowledge graph, multiple node evolution factors that are most compatible with the development trend of the health representation data of the target entity are selected from the medical knowledge graph, and then a medical evolution factor sequence that conforms to the target entity is generated.

[0046] S130. Obtain the insurance application time information of the target entity object, and based on the insurance application time information and the health representation data of the target entity object, determine the medical inference factor mapped by the insurance application time information of the target entity object from the medical evolution factor sequence.

[0047] Among them, the insurance period information of the target entity can be the insurance duration T, such as three years, five years, ten years and twenty years (the insurance period is used as the start time).

[0048] In some embodiments, based on the insurance application time information and health characterization data of the target entity, the medical inference factors corresponding to the insurance application time information of the target entity are determined from the medical evolution factor sequence, including:

[0049] Based on the health representation data of the target entity, the evolution time of multiple node evolution factors contained in the medical evolution factor sequence is estimated to obtain the estimated evolution time corresponding to each node evolution factor. The estimated evolution time corresponding to each node evolution factor is matched with the insurance time information of the target entity to obtain the medical inference factor mapped by the insurance time information of the target entity.

[0050] Based on the above example, and using the health representation data of the target entity, the estimated evolution time corresponding to the development of the target entity to evolution factor A is estimated as T1 (starting from the current estimated time), and the estimated evolution time corresponding to the development of the target entity to evolution factor B is estimated as T2. Where T1 is less than the insurance period T and T2 is greater than the insurance period T, the medical inference factor mapped by the insurance period information of the target entity is determined to be evolution factor A.

[0051] Therefore, by matching the insurance application time information of the target entity with the estimated evolution time corresponding to each evolution factor at each node, it is possible to effectively derive the medical inference factors that the target entity may generate during the insurance period, and thus make a reasonable prediction of the target entity's physical health.

[0052] S140. Perform intelligent underwriting analysis on the health characterization data of the target entity to obtain the intelligent underwriting information corresponding to the target entity.

[0053] Among them, intelligent underwriting analysis is to analyze the insurance situation based on the health characteristics data of the target entity. Intelligent underwriting information can be used to describe the underwriting conclusion of the target entity, such as rejection, additional premium, standard coverage, exclusion, etc.

[0054] In some embodiments, intelligent underwriting analysis is performed on the health characterization data of the target entity to obtain intelligent underwriting information corresponding to the target entity, including:

[0055] Based on the medical underwriting type of the target entity, the corresponding intelligent underwriting analysis model is determined from the disease model library, and the health representation features related to the medical underwriting type are determined from the health representation data of the target entity. The health representation features related to the medical underwriting type are input into the intelligent underwriting analysis model, and the underwriting risk probability of the target entity corresponding to the medical underwriting type is determined based on the output of the intelligent underwriting analysis model. Based on the underwriting risk probability of the target entity corresponding to the medical underwriting type, the intelligent underwriting information corresponding to the target entity is determined.

[0056] In particular, the integration of multiple medical knowledge graphs provides domain knowledge support for the intelligent underwriting analysis model. This model can effectively determine the underwriting risk probability of a target entity corresponding to a specific medical underwriting type. Specifically, the weights of the predicted probabilities in the model can be dynamically adjusted for different health characteristics to effectively adapt to different features.

[0057] In addition, it supports independent partitioning of disease knowledge (the knowledge bases for pulmonary nodules and thyroid nodules do not interfere with each other), and the clinical evidence output by the knowledge base can be embedded into the decision support of the manual review process.

[0058] This embodiment also controls the modification permissions of different roles for underwriting rules. For example, underwriters can only view the calculation rules, while engineers can adjust the model parameters. It also pushes underwriting task status changes (such as manual review tasks / timeout reminders), links with the task management event bus, and records the rule modification history to meet audit compliance requirements.

[0059] In some embodiments, before inputting health characterization features related to the type of medical underwriting into the intelligent underwriting analysis model, the following steps are also included:

[0060] Medical characteristics are queried for the target entity to obtain medical response characteristics; based on the medical underwriting type, the medical response characteristics are filtered to obtain underwriting-related characteristics related to the medical underwriting type; based on the underwriting-related characteristics related to the medical underwriting type, the health representation characteristics related to the medical underwriting type are adjusted.

[0061] Based on the above examples, if the medical response characteristics reported by the target entity are "headache" and "sore throat", and feature filtering is performed on the medical response characteristics, the underwriting association characteristic related to the medical underwriting type is "sore throat". Then, the feature "sore throat" can be added to the health representation characteristics, thereby enriching the feature details of the health representation characteristics.

[0062] S150. Obtain the historical evolution data between the health representation data of the target entity and the medical inference factors, and determine the dynamic evolution probability of the target entity corresponding to the medical inference factors based on the historical evolution data between the health representation data of the target entity and the medical inference factors.

[0063] Among them, the dynamic evolution probability of the target entity object corresponding to the medical inference factor can be used to describe the predicted probability of the target entity object evolving from its current state to the state corresponding to the medical inference factor.

[0064] In some embodiments, acquiring historical evolution data between the health representation data of the target entity and medical extrapolation factors includes:

[0065] Obtain a first entity object set, which includes multiple other entity objects, each of which is associated with a medical projection factor; obtain health representation data for each other entity object in the first entity object set; based on the insurance time information of the target entity object, perform time matching from the health representation data of the multiple other entity objects to obtain a second entity object set; based on the number of other entity objects in the second entity object set and the number of other entity objects in the first entity object set, determine the historical evolution data between the health representation data of the target entity object and the medical projection factor.

[0066] Specifically, if the ratio of the number of other entity objects in the second entity object set to the number of other entity objects in the first entity object set is greater than or equal to a preset threshold, then the historical evolution data between the health representation data and the medical inference factor of the target entity object is determined as the first representation value; if the ratio of the number of other entity objects in the second entity object set to the number of other entity objects in the first entity object set is less than the preset threshold, then the historical evolution data between the health representation data and the medical inference factor of the target entity object is determined as the second representation value.

[0067] When the historical evolution data between the health representation data and the medical inference factor of the target entity is a first representation value, the dynamic evolution probability of the target entity corresponding to the medical inference factor is a first probability. When the historical evolution data between the health representation data and the medical inference factor of the target entity is a second representation value, the dynamic evolution probability of the target entity corresponding to the medical inference factor is a second probability, and the first probability is greater than the second probability.

[0068] It is understandable that when the dynamic evolution probability of the target entity object corresponding to the medical inference factor is the first probability, it means that the target entity object has a relatively high probability of developing into a medical inference factor; when the dynamic evolution probability of the target entity object corresponding to the medical inference factor is the second probability, it means that the target entity object has a relatively low probability of developing into a medical inference factor.

[0069] S160. Based on the dynamic evolution probability of the medical inference factor corresponding to the target entity object, optimize the intelligent underwriting information corresponding to the target entity object to obtain the target underwriting information corresponding to the target entity object.

[0070] Information optimization refers to adjusting the intelligent underwriting information, such as changing "standard body" to "surcharge" or "rejection of insurance".

[0071] In some embodiments, based on the dynamic evolution probability of the medical extrapolation factor corresponding to the target entity object, the intelligent underwriting information corresponding to the target entity object is optimized to obtain the target underwriting information corresponding to the target entity object, including:

[0072] Based on the dynamic evolution probability of the target entity object corresponding to the medical inference factor, the corresponding underwriting optimization parameters and parameter optimization data are determined from the evolution rule set; based on the parameter optimization data, the underwriting optimization parameters included in the intelligent underwriting information corresponding to the target entity object are adjusted to obtain the target underwriting information corresponding to the target entity object.

[0073] Specifically, if the dynamic evolution probability of the target entity corresponding to the medical deduction factor is the first probability, then the corresponding underwriting optimization parameter determined from the evolution rule set is: the underwriting conclusion, and the parameter optimization data is: the conclusion modification instruction. For example, if the intelligent underwriting information is "standard body", it is modified to "rejection". If the dynamic evolution probability of the target entity corresponding to the medical deduction factor is the second probability, then the corresponding underwriting optimization parameter determined from the evolution rule set is: the underwriting cost, and the parameter optimization data is: the cost increase instruction. For example, if the intelligent underwriting information is "standard body", it is modified to "30% increase in cost".

[0074] Therefore, by analyzing the dynamic evolution probability of the target entity object corresponding to the medical inference factor, the corresponding underwriting optimization parameters and parameter optimization data are determined from the set of evolution rules, so as to accurately adjust the intelligent underwriting information generated by intelligent underwriting and make it easier to determine the most accurate target underwriting information.

[0075] In this embodiment, health representation data of the target entity is obtained, which is filtered from the target entity's historical medical information. Medical extrapolation analysis is performed on the health representation data of the target entity using a medical knowledge graph to obtain a sequence of medical evolution factors associated with the health representation data. Insurance application time information of the target entity is obtained, and based on the insurance application time information and the health representation data, the medical extrapolation factors corresponding to the insurance application time information are determined from the medical evolution factor sequence. Intelligent underwriting analysis is performed on the health representation data of the target entity to obtain intelligent underwriting information corresponding to the target entity. Historical evolution data between the health representation data and the medical extrapolation factors of the target entity is obtained, and the dynamic evolution probability of the target entity corresponding to the medical extrapolation factors is determined based on this historical evolution data. Based on the dynamic evolution probability of the target entity corresponding to the medical extrapolation factors, the intelligent underwriting information corresponding to the target entity is optimized to obtain the target underwriting information corresponding to the target entity. In this way, by performing knowledge graph deduction on the health representation data of the target entity, medical deduction factors with similar development trends to the health representation data are obtained. Based on the historical evolution data between the medical deduction factors and the health representation data, the preliminary intelligent underwriting information is optimized to obtain the target underwriting information, effectively improving the accuracy of intelligent underwriting.

[0076] In addition, this embodiment can define the differences in underwriting permissions and processes for different sales channels, use a multimodal data pool to store the health records of customers with illnesses, and configure the binding relationship of underwriting rules for different insurance products. For underwriting tasks, the verification rules can be automatically invoked, computing resources can be dynamically allocated according to task priority, and a manual review process can be initiated for low-confidence tasks (such as RB values ​​in the threshold critical zone), and a self-built knowledge base can be invoked to provide clinical guideline references and feedback review conclusions for model iteration and optimization.

[0077] For health characterization data, multimodal data validation rules can be set to detect contradictions between text data and user claims; cross-disease extensions are supported, such as requiring validation of ultrasound report formats for thyroid nodules / breast nodules; a built-in OCR+structured engine converts medical parameters (such as nodule long axis and spiculation) into machine-processable structured data; different disease mapping templates are stored independently, such as lung nodules → CT parameters, thyroid nodules → ultrasound parameters, TI-RADS classification, breast nodules → ultrasound parameters, BI-RADS classification, etc. During data auditing, blockchain is used to record sensitive operations, such as hashing underwriting rule version change records on the blockchain; a monitoring mechanism is also set up, using Prometheus to collect rule engine performance indicators and using ELK to implement real-time alerts for abnormal logs, such as issuing an alert if the lung nodule model's calculation results exceed the reasonable range five times consecutively.

[0078] In summary, this embodiment achieves significant improvements in underwriting efficiency, accuracy, scalability, and compliance by integrating multimodal data, dynamic risk quantification, and rule engine technology. Specifically, regarding the automated underwriting decision rate, through the rule engine's verification, mapping, calculation, and chained processing of underwriting rules, the automated processing rate of underwriting cases involving individuals with pre-existing conditions can be increased from 35% in the traditional model to 90% or even higher, with manual review only applied to low-confidence cases; based on a dynamic GPU resource allocation algorithm, the processing speed of underwriting tasks involving individuals with pre-existing conditions can be significantly improved. Regarding labor costs, under the human-machine collaboration mechanism, the average daily processing capacity of a single underwriter can increase from 50 to 300 cases, significantly reducing labor costs. Furthermore, through adversarial example annotation tools, manually fed-in data can be directly used for model optimization, forming a training closed loop. Regarding multimodal contradiction detection, the verification rule module incorporates a medical knowledge graph, which can automatically identify various data contradiction scenarios, improving the contradiction detection rate compared to manual review. Regarding dynamic risk quantification, the lung nodule RB value model integrates quantitative assessments of multiple lung cancer-related factors (age, gender, smoking history, family history of lung cancer, nodule CT manifestations, dynamic progression, etc.) to make risk assessment more accurate. The breast nodule model dynamically links BI-RADS classification with breast cancer high-risk factor data (family history, menstrual history, age at menarche, nodule progression, etc.), resulting in significantly higher accuracy in assessing breast cancer risk compared to simple classification. This embodiment upgrades underwriting decisions from experience-driven to data-driven through multimodal fusion and dynamic quantification models; the rule engine's chained processing enables standardized and configurable underwriting processes; differentiated insurance plans are provided based on individual customer risk profiles; and dynamic monitoring of disease-related risk factors and clinical progression enables early risk screening.

[0079] Figure 2 This embodiment provides a schematic diagram of a knowledge graph-based intelligent underwriting device. The knowledge graph-based intelligent underwriting device may include:

[0080] The acquisition module 210 is used to acquire the health representation data of the target entity object, which is obtained by filtering the target entity object's historical medical information.

[0081] The first analysis module 220 is used to perform medical extrapolation analysis on the health representation data of the target entity object through a medical knowledge graph, and obtain the sequence of medical evolution factors associated with the health representation data of the target entity object.

[0082] The first determining module 230 is used to obtain the insurance time information of the target entity object, and determine the medical inference factor corresponding to the insurance time information of the target entity object from the medical evolution factor sequence based on the insurance time information and the health characterization data of the target entity object.

[0083] The second analysis module 240 is used to perform intelligent underwriting analysis on the health characterization data of the target entity to obtain the intelligent underwriting information corresponding to the target entity.

[0084] The second determining module 250 is used to acquire the historical evolution data between the health representation data of the target entity and the medical inference factors, and to determine the dynamic evolution probability of the target entity corresponding to the medical inference factors based on the historical evolution data between the health representation data of the target entity and the medical inference factors.

[0085] The optimization module 260 is used to optimize the intelligent underwriting information corresponding to the target entity object based on the dynamic evolution probability of the medical inference factor corresponding to the target entity object, so as to obtain the target underwriting information corresponding to the target entity object.

[0086] In this embodiment, optionally, the first analysis module 220 is specifically used for:

[0087] Based on the health representation data of the target entity object, multiple related evolution factors are matched from the medical knowledge graph, and the evolution fusion value between the health representation data of the target entity object and each related evolution factor is calculated. Based on the evolution fusion value between the health representation data of the target entity object and each related evolution factor, multiple node evolution factors related to the health representation data of the target entity object are determined. Based on the multiple node evolution factors related to the health representation data of the target entity object, a sequence of medical evolution factors associated with the health representation data of the target entity object is generated.

[0088] In this embodiment, optionally, the first determining module 230 is specifically used for:

[0089] Based on the health representation data of the target entity, the evolution time of multiple node evolution factors contained in the medical evolution factor sequence is estimated to obtain the estimated evolution time corresponding to each node evolution factor. The estimated evolution time corresponding to each node evolution factor is matched with the insurance time information of the target entity to obtain the medical inference factor mapped by the insurance time information of the target entity.

[0090] In this embodiment, optionally, the second analysis module 240 is specifically used for:

[0091] Based on the medical underwriting type of the target entity, the corresponding intelligent underwriting analysis model is determined from the disease model library, and the health representation features related to the medical underwriting type are determined from the health representation data of the target entity. The health representation features related to the medical underwriting type are input into the intelligent underwriting analysis model, and the underwriting risk probability of the target entity corresponding to the medical underwriting type is determined based on the output of the intelligent underwriting analysis model. Based on the underwriting risk probability of the target entity corresponding to the medical underwriting type, the intelligent underwriting information corresponding to the target entity is determined.

[0092] In this embodiment, optionally, the second determining module 250 is specifically used for:

[0093] Obtain a first entity object set, which includes multiple other entity objects, each of which is associated with a medical projection factor; obtain health representation data for each other entity object in the first entity object set; based on the insurance time information of the target entity object, perform time matching from the health representation data of the multiple other entity objects to obtain a second entity object set; based on the number of other entity objects in the second entity object set and the number of other entity objects in the first entity object set, determine the historical evolution data between the health representation data of the target entity object and the medical projection factor.

[0094] In this embodiment, optionally, the optimization module 260 is specifically used for:

[0095] Based on the dynamic evolution probability of the target entity object corresponding to the medical inference factor, the corresponding underwriting optimization parameters and parameter optimization data are determined from the evolution rule set; based on the parameter optimization data, the underwriting optimization parameters included in the intelligent underwriting information corresponding to the target entity object are adjusted to obtain the target underwriting information corresponding to the target entity object.

[0096] In this embodiment, optionally, a processing module may also be included.

[0097] The processing module is used to query the target entity object for medical characteristics to obtain medical response characteristics; to filter the medical response characteristics based on the medical underwriting type to obtain underwriting-related characteristics related to the medical underwriting type; and to adjust the health representation characteristics related to the medical underwriting type based on the underwriting-related characteristics related to the medical underwriting type.

[0098] The intelligent underwriting device based on knowledge graph inference provided in this disclosure can execute the above-described method embodiments. Its specific implementation principle and technical effects can be found in the above-described method embodiments, and will not be repeated here.

[0099] This application also provides a computer device. Please refer to the following for details. Figure 3 , Figure 3 This is a basic structural block diagram of the computer device in this embodiment.

[0100] The computer device includes a memory 310 and a processor 320 that are communicatively connected to each other via a system bus. It should be noted that only a computer device with memory 310 and processor 320 is shown in the figure; however, it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented alternatively. Those skilled in the art will understand that the computer device described herein is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0101] Computer devices can include desktop computers, laptops, handheld computers, and cloud servers. These devices allow for human-computer interaction with users through keyboards, mice, remote controls, touchpads, or voice-activated devices.

[0102] The memory 310 includes at least one type of readable storage medium, including non-volatile memory or volatile memory, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. RAM may include static RAM or dynamic RAM. In some embodiments, the memory 310 may be an internal storage unit of a computer device, such as the hard disk or RAM of the computer device. In other embodiments, the memory 310 may also be an external storage device of the computer device, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, or flash card equipped on the computer device. Of course, the memory 310 may include both internal storage units and external storage devices of the computer device. In this embodiment, the memory 310 is typically used to store the operating system and various application software installed on the computer device, such as the program code of the methods described above. Furthermore, the memory 310 may also be used to temporarily store various types of data that have been output or will be output.

[0103] Processor 320 is typically used to perform overall operations of a computer device. In this embodiment, memory 310 is used to store program code or instructions, including computer operation instructions, and processor 320 is used to execute the program code or instructions stored in memory 310 or process data, such as program code that runs the methods described above.

[0104] In this article, the bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. This bus system can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0105] Another embodiment of this application also provides a computer-readable medium, which may be a computer-readable signal medium or a computer-readable medium. A processor in a computer reads computer-readable program code stored in the computer-readable medium, enabling the processor to execute the functional actions specified in each step or combination of steps in the above method; and to generate means for implementing the functional actions specified in each block or combination of blocks in the block diagram.

[0106] Computer-readable media include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared memory or semiconductor systems, devices or apparatuses, or any suitable combination thereof, wherein the memory is used to store program code or instructions, the program code including computer operation instructions, and the processor is used to execute the program code or instructions of the above-described methods stored in the memory.

[0107] The definitions of memory and processor can be found in the description of the foregoing computer device embodiments, and will not be repeated here.

[0108] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units 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. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0109] In the various embodiments of this application, the functional units or modules 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.

[0110] 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 several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor 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 program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0111] In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" as described in this application does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims listing several means, several units of these means may be embodied by the same item of hardware. The use of "first," "second," and "third," etc., does not indicate any order and these words should be interpreted as names. Unless otherwise specified, the steps in the above embodiments should not be construed as limiting the order of execution.

[0112] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An intelligent underwriting method based on knowledge graph inference, characterized in that, include: Obtain health representation data of the target entity object, wherein the health representation data of the target entity object is obtained by filtering from the historical medical information of the target entity object; By using a medical knowledge graph to perform medical extrapolation and analysis on the health representation data of the target entity, a sequence of medical evolution factors associated with the health representation data of the target entity is obtained. The step of performing medical extrapolation analysis on the health representation data of the target entity object using a medical knowledge graph to obtain a sequence of medical evolution factors associated with the health representation data of the target entity object includes: matching multiple corresponding associated evolution factors from the medical knowledge graph based on the health representation data of the target entity object, and calculating the evolution fusion value between the health representation data of the target entity object and each associated evolution factor; determining multiple node evolution factors associated with the health representation data of the target entity object based on the evolution fusion value between the health representation data of the target entity object and each associated evolution factor; and generating a sequence of medical evolution factors associated with the health representation data of the target entity object based on the multiple node evolution factors associated with the health representation data of the target entity object. Obtain the insurance application time information of the target entity object, and based on the insurance application time information and the health characterization data of the target entity object, determine the medical inference factor corresponding to the insurance application time information of the target entity object from the medical evolution factor sequence; Intelligent underwriting analysis is performed on the health characterization data of the target entity to obtain intelligent underwriting information corresponding to the target entity. Obtain the health representation data of the target entity object and the historical evolution data between the medical inference factors, and determine the dynamic evolution probability of the target entity object corresponding to the medical inference factors based on the historical evolution data between the health representation data of the target entity object and the medical inference factors. Based on the dynamic evolution probability of the target entity object corresponding to the medical inference factor, the intelligent underwriting information corresponding to the target entity object is optimized to obtain the target underwriting information corresponding to the target entity object.

2. The method according to claim 1, characterized in that, The step of determining the medical evolution factor corresponding to the insurance application time information of the target entity from the medical evolution factor sequence based on the insurance application time information and the health characterization data of the target entity includes: Based on the health representation data of the target entity object, the evolution time of multiple node evolution factors contained in the medical evolution factor sequence is estimated to obtain the estimated evolution time corresponding to each node evolution factor. Match the estimated evolution time corresponding to each node evolution factor with the insurance time information of the target entity object to obtain the medical projection factor mapped by the insurance time information of the target entity object.

3. The method according to claim 1, characterized in that, The intelligent underwriting analysis of the health characterization data of the target entity to obtain the intelligent underwriting information corresponding to the target entity includes: Based on the medical underwriting type of the target entity, the corresponding intelligent underwriting analysis model is determined from the disease model library, and the health representation features related to the medical underwriting type are determined from the health representation data of the target entity. The health characteristics associated with the medical underwriting type are input into the intelligent underwriting analysis model, and the underwriting risk probability of the target entity object corresponding to the medical underwriting type is determined based on the output of the intelligent underwriting analysis model. Based on the underwriting risk probability of the target entity object corresponding to the medical underwriting type, the intelligent underwriting information corresponding to the target entity object is determined.

4. The method according to claim 1, characterized in that, The acquisition of historical evolution data between the health representation data of the target entity and the medical extrapolation factors includes: Obtain a first set of entity objects, which includes multiple other entity objects, each of which is associated with the medical deduction factor; Obtain the health characterization data of each of the other entity objects in the first entity object set; Based on the insurance time information of the target entity object, time matching is performed from the health characterization data of multiple other entity objects to obtain a second entity object set; Based on the number of other entity objects in the second entity object set and the number of other entity objects in the first entity object set, the historical evolution data between the health representation data of the target entity object and the medical extrapolation factor is determined.

5. The method according to claim 1, characterized in that, The step of optimizing the intelligent underwriting information corresponding to the target entity object based on the dynamic evolution probability of the medical inference factor corresponding to the target entity object to obtain the target underwriting information corresponding to the target entity object includes: Based on the dynamic evolution probability of the target entity object corresponding to the medical inference factor, the corresponding underwriting optimization parameters and parameter optimization data are determined from the evolution rule set; Based on the parameter optimization data, the underwriting optimization parameters included in the intelligent underwriting information corresponding to the target entity object are adjusted to obtain the target underwriting information corresponding to the target entity object.

6. The method according to claim 3, characterized in that, Also includes: Perform medical feature queries on the target entity object to obtain medical response features; Based on the medical underwriting type, feature filtering is performed on the medical response characteristics to obtain underwriting association characteristics related to the medical underwriting type; Based on the underwriting association features related to the medical underwriting type, the health representation features related to the medical underwriting type are adjusted.

7. An intelligent underwriting device based on knowledge graph deduction, characterized in that, include: The acquisition module is used to acquire health representation data of the target entity object, wherein the health representation data of the target entity object is obtained by filtering from the historical medical information of the target entity object; The first analysis module is used to perform medical extrapolation analysis on the health representation data of the target entity object through a medical knowledge graph to obtain a sequence of medical evolution factors associated with the health representation data of the target entity object; the first analysis module is specifically used to: match multiple corresponding associated evolution factors from the medical knowledge graph based on the health representation data of the target entity object, and calculate the evolution fusion value between the health representation data of the target entity object and each of the associated evolution factors. Based on the evolution fusion value between the health representation data of the target entity and each of the associated evolution factors, multiple node evolution factors related to the health representation data of the target entity are determined. Based on multiple node evolution factors related to the health representation data of the target entity, a sequence of medical evolution factors associated with the health representation data of the target entity is generated. The first determining module is used to obtain the insurance time information of the target entity object, and determine the medical inference factor corresponding to the insurance time information of the target entity object from the medical evolution factor sequence based on the insurance time information and the health characterization data of the target entity object; The second analysis module is used to perform intelligent underwriting analysis on the health characterization data of the target entity object to obtain the intelligent underwriting information corresponding to the target entity object. The second determining module is used to acquire the health characterization data of the target entity object and the historical evolution data between the medical inference factors, and to determine the dynamic evolution probability of the target entity object corresponding to the medical inference factors based on the historical evolution data between the health characterization data of the target entity object and the medical inference factors. The optimization module is used to optimize the intelligent underwriting information corresponding to the target entity object based on the dynamic evolution probability of the target entity object corresponding to the medical inference factor, so as to obtain the target underwriting information corresponding to the target entity object.

8. A computer device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the intelligent underwriting method based on knowledge graph inference as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the intelligent underwriting method based on knowledge graph inference as described in any one of claims 1 to 6.

Images