Claim settlement operation platform digital decision-making system based on large model agent Agent

Abstract

The invention relates to the technical field of claim settlement operation data processing, in particular to a claim settlement operation platform digital decision-making system based on a large model Agent. The data processing module comprises an associated data determining unit used for determining associated data in the business data; the data setting unit is used for determining setting of the business data; the exception analysis unit is used for carrying out exception analysis on the business data; the data cleaning unit is used for extracting abnormal data and normal data in the business data and cleaning the abnormal data and the normal data; the data adjusting unit is used for adjusting the service data setting; and the adjusting module is used for adjusting the model or the data processing process. The problem of low claim decision making efficiency caused by low claim decision making accuracy in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of claims operation data processing, and in particular to a digital decision-making system for a claims operation platform based on a large model intelligent agent. Background Art

[0002] As digital transformation deepens, the insurance industry faces increasingly complex data processing and decision-making challenges. While traditional business intelligence (BI) platforms offer data analysis and reporting capabilities, they are limited in handling complex business scenarios, real-time data analysis, and automated decision support. Furthermore, operations managers need to quickly understand business data to make timely and accurate decisions. This requires systems that not only possess advanced analytical capabilities but also present the results in an easily understandable format.

[0003] Chinese Patent Publication No. CN118014739A discloses a method and apparatus for generating claims decision rules based on a dual-layer graph. The method comprises: constructing a dual-layer knowledge graph based on a business graph and a decision graph; building risk combination rules based on this dual-layer knowledge graph; and making claims decisions based on the risk combination rules for insurance claims cases. This method leverages the characteristics of the knowledge graph to integrate various business data and rules into the claims decision-making process, making claims decisions more accurate, comprehensive, and effective.

[0004] It can be seen that the existing technology has the problem of low efficiency of claim settlement decision-making due to low accuracy of claim settlement decision-making. Summary of the Invention

[0005] To this end, the present invention provides a digital decision system for a claims operation platform based on a large model agent, so as to overcome the problem of low efficiency of claims decision-making due to low accuracy of claims decision-making in the prior art.

[0006] To achieve the above objectives, the present invention provides a digital decision-making system for a claims operation platform based on a large model agent, comprising:

[0007] Data collection module, which is used to collect business data from each business terminal;

[0008] A data processing module, connected to the data acquisition module and used to process the business data, includes:

[0009] a related data determining unit, configured to determine whether related data exists in the business data and determine the type of the related data;

[0010] a data setting unit connected to the associated data determining unit, for determining a setting of a ratio of structured data to unstructured data of the business data according to a type of the associated data;

[0011] an anomaly analysis unit connected to the data setting unit, configured to determine the consistency of associated data in the business data and determine a proportion of consistent data in the business data, so as to determine an anomaly analysis of the business data based on the proportion of consistent data;

[0012] a data cleaning unit connected to the abnormality analysis unit, configured to extract abnormal data and normal data from the business data, and perform cleaning processing on the abnormal data and the normal data;

[0013] a data adjustment unit connected to the data cleaning unit, for determining an adjustment of the ratio of structured data to unstructured data in the business data according to the importance of the abnormal data;

[0014] An adjustment module is connected to the data processing module and is used to perform deep learning of the model on the business data cleaned by the data cleaning unit and determine the accuracy of the model to determine the adjustment of the model parameters or the data processing process according to the accuracy of the model.

[0015] Furthermore, the data setting unit determines the processing of the business data according to the type of the associated data, including determining the proportion of sensitive data in the associated data. Under the condition that the proportion of sensitive data is less than or equal to the preset sensitive data proportion, the data setting unit determines that the proportion of structured data and unstructured data in the business data is a first preset proportion.

[0016] Furthermore, the data setting unit determines the processing of the business data according to the type of the associated data, including determining the proportion of sensitive data in the associated data. Under the condition that the proportion of sensitive data is greater than the preset sensitive data proportion, the data setting unit determines that the proportion of structured data and unstructured data in the business data is a second preset proportion.

[0017] Furthermore, the abnormality analysis unit determines the abnormality analysis of the business data based on the proportion of the consistency data, including analyzing all business data in the business data to determine whether there is abnormal data under the condition that the proportion of the consistency data is less than or equal to the preset consistency data proportion.

[0018] Furthermore, the abnormality analysis unit determines the abnormality analysis of the business data according to the proportion of the consistency data, including analyzing key data in the business data to determine whether there is abnormal data under the condition that the proportion of the consistency data is greater than a preset consistency data proportion.

[0019] Furthermore, the data adjustment unit determines the adjustment of the data setting unit according to the importance of the abnormal data, including adjusting the ratio of structured data and unstructured data in the data setting unit according to the comparison result of the importance evaluation value and the preset importance evaluation value.

[0020] Furthermore, the determination of adjusting the ratio of structured data to unstructured data in the data setting unit based on the comparison result of the importance evaluation value and the preset importance evaluation value includes adjusting the ratio of structured data to unstructured data according to an adjustment coefficient under the condition that the importance evaluation value is greater than or equal to the preset importance evaluation value.

[0021] Furthermore, the adjustment module determines the adjustment of the model or the data processing process according to the accuracy of the model, including determining the adjustment of the model or the data processing process according to the comparison result of the difference between the accuracy evaluation value of the model and the preset accuracy evaluation value and the difference between the preset accuracy evaluation values.

[0022] Furthermore, the adjustment of the model or the data processing process determined based on the comparison result of the difference between the accuracy evaluation value of the model and the preset accuracy evaluation value and the preset accuracy evaluation value difference includes adjusting the preset consistency data ratio in the abnormality analysis unit according to the adjustment coefficient under the condition that the difference between the accuracy evaluation value and the preset accuracy evaluation value is less than or equal to the preset accuracy evaluation value difference.

[0023] Furthermore, determining the adjustment of the model or the data processing process based on the comparison result of the difference between the accuracy evaluation value and the preset accuracy evaluation value and the preset accuracy evaluation value difference includes adjusting the parameters of the model under the condition that the difference between the accuracy evaluation value and the preset accuracy evaluation value is greater than the preset accuracy evaluation value difference.

[0024] Compared with the prior art, the present invention has the following advantages: by providing an associated data determination unit to determine whether associated data exists at each business end and to determine the type of associated data, the data setting unit determines the ratio of structured data to unstructured data based on a comparison of the ratio of sensitive data with a preset sensitive data ratio, thereby optimizing data management and improving the efficiency of claims decision-making. Structured data, due to its prescribed data format, is easier to query, analyze, and process, helping to improve the efficiency and accuracy of claims operational decisions; unstructured data contains a richer variety of information, providing strong support for claims operational decisions.

[0025] Furthermore, the present invention improves data accuracy by providing an anomaly analysis unit that determines anomaly analysis of the business data based on a comparison of the consistency data ratio with a preset consistency data ratio. A higher consistency data ratio indicates a higher reliability of the claims data. Under different comparison results of the consistency data ratio with a preset consistency data ratio, different anomaly analysis methods are employed for the business data, thereby improving data accuracy and further improving the accuracy of claims decision-making results.

[0026] Furthermore, the present invention provides a data cleaning unit and cleans the extracted abnormal data and normal data, thereby improving the accuracy of the data and avoiding low model accuracy due to inaccurate data; it also facilitates subsequent data processing and analysis.

[0027] Furthermore, the present invention provides an adjustment module to determine adjustments to the model or the data processing process based on a comparison of the difference between the model's accuracy evaluation value and a preset accuracy evaluation value, and the difference between the preset accuracy evaluation values. If the comparison results differ, the adjustment module selects different adjustment methods, thereby optimizing the utilization of system resources and improving the efficiency of claims decision-making. Furthermore, during the adjustment process, setting a specific adjustment coefficient can avoid over-adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a structural diagram of a digital decision-making system for a claims operation platform based on a large model agent according to an embodiment of the present invention;

[0029] Figure 2 This is a flow chart of the abnormality analysis unit of the digital decision-making system of the claims operation platform based on the large model agent according to an embodiment of the present invention;

[0030] Figure 3 This is a flow chart of a data adjustment unit of a digital decision-making system for a claims operation platform based on a large model agent according to an embodiment of the present invention;

[0031] Figure 4 This is a flow chart of the adjustment module of the digital decision-making system of the claims operation platform based on the large model intelligent agent according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0033] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0034] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside", and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0035] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0036] Example 1

[0037] See also Figure 1-Figure 4 As shown, the digital decision-making system of the claims operation platform based on the large model agent in the embodiment of the present invention includes:

[0038] Data collection module, which is used to collect business data from each business terminal;

[0039] A data processing module, connected to the data acquisition module and used to process the business data, includes:

[0040] a related data determining unit, configured to determine whether related data exists in the business data and determine the type of the related data;

[0041] a data setting unit connected to the associated data determining unit, configured to determine settings for the service data according to a type of the associated data;

[0042] an abnormality analysis unit connected to the data setting unit, configured to determine the consistency of associated data in the business data and determine a proportion of consistent data in the business data, so as to determine abnormality analysis of the business data based on the proportion of consistent data;

[0043] a data cleaning unit connected to the abnormality analysis unit, configured to extract abnormal data and normal data from the business data, and perform cleaning processing on the abnormal data and the normal data;

[0044] a data adjustment unit connected to the data cleaning unit, for determining an adjustment of the ratio of structured data to unstructured data in the business data according to the importance of the abnormal data;

[0045] An adjustment module is connected to the data processing module and is used to perform deep learning of the model on the business data cleaned by the data cleaning unit and determine the accuracy of the model to determine the adjustment of the model parameters or the data processing process according to the accuracy of the model.

[0046] Specifically, the associated data determining unit determines the type of the associated data, including determining that the associated data is sensitive data or non-sensitive data.

[0047] In an embodiment of the present invention, the sensitive data includes but is not limited to the customer's name, ID number, address, telephone number, bank account information, and password; the non-sensitive data includes but is not limited to company information, claim application records, and user operation records.

[0048] Specifically, the data setting unit determines the processing of the business data according to the type of the associated data, including determining the proportion B of sensitive data in the associated data, and determining the proportion D of structured data and unstructured data in the business data based on the comparison result of the proportion B of sensitive data and the preset sensitive data proportion B0.

[0049] In the embodiment of the present invention, the preset value of the sensitive data ratio B0 is 80%.

[0050] Specifically, determining the ratio D of structured data to unstructured data in the business data based on the ratio B of the sensitive data includes:

[0051] If B≤B0, the data setting unit determines that the ratio D of structured data to unstructured data in the business data is a first preset ratio D1;

[0052] If B>B0, the data setting unit determines that the ratio D of the structured data to the unstructured data in the business data is a second preset ratio D2.

[0053] In the embodiment of the present invention, the higher the proportion of the sensitive data, the higher the proportion of the corresponding structured data should be, so the first preset proportion D1 and the second preset proportion D2 should satisfy D1<D2.

[0054] In the embodiment of the present invention, the preset ratio is a value of structured data to unstructured data, wherein the preset ratio includes a first preset ratio and a second preset ratio.

[0055] In an embodiment of the present invention, the value range of the first preset ratio D1 is 0-1, and the value range of the second preset ratio D2 is 2-9; the value of the first preset ratio D1 is 0.5, and the value of the second preset ratio D2 is 5, which can avoid errors caused by values ​​that are too high or too low.

[0056] In this embodiment of the present invention, the ratio of structured data to unstructured data is determined based on a comparison of the sensitive data ratio with a preset sensitive data ratio, thereby optimizing data management and improving the efficiency of claims decision-making. Structured data, due to its prescribed data format, is easier to query, analyze, and process, helping to improve the efficiency and accuracy of claims operational decisions. Unstructured data, on the other hand, contains a richer variety of information, providing strong support for claims operational decisions.

[0057] Specifically, the abnormality analysis unit determines the abnormality analysis of the business data according to the consistency data ratio M, including determining the abnormality analysis of the business data according to a comparison result of the consistency data ratio M and a preset consistency data ratio M0.

[0058] In the embodiment of the present invention, the preset value of the consistency data ratio M0 is 95%.

[0059] Specifically, determining the abnormality analysis of the business data according to the comparison result of the consistency data ratio M and the preset consistency data ratio M0 includes:

[0060] If M≤M0, the abnormality analysis unit analyzes all business data in the business data to determine whether there is abnormal data;

[0061] If M>M0, the abnormality analysis unit analyzes the key data in the business data to determine whether abnormal data exists.

[0062] In an embodiment of the present invention, the key data includes customer information, policy information, accident information, and claim amount.

[0063] Specifically, the data adjustment unit determines the adjustment of the data setting unit according to the importance of the abnormal data, including determining the importance evaluation value P of the abnormal data, and adjusting the ratio D of structured data and unstructured data in the data setting unit according to the comparison result of the importance evaluation value P and the preset importance evaluation value P0.

[0064] In the embodiment of the present invention, the importance evaluation value P is calculated according to the following formula:

[0065] P=ω1×A+ω2×E

[0066] Where P represents the importance evaluation value, ω1 and ω2 represent weight coefficients, A represents the score value of the impact of abnormal data on the business, and E represents the score value of the frequency of abnormal data occurrence.

[0067] In the embodiment of the present invention, the weight coefficients ω1 and ω2 satisfy ω1+ω2=1, wherein the value of ω1 is 0.5 and the value of ω2 is 0.5.

[0068] In the embodiment of the present invention, the score of the impact of abnormal data on business is evaluated based on whether the abnormal data causes system failures or user complaints, and its value range is 0-1. The score of the impact of abnormal data on business is determined according to the following formula:

[0069]

[0070] Where A represents the score of the impact of abnormal data on the business, h represents the number of system failures, and c represents the number of user complaints.

[0071] In the embodiment of the present invention, the scoring value of the frequency of occurrence of abnormal data is determined according to the frequency of occurrence of abnormal data. The higher the frequency of occurrence, the higher the corresponding scoring value. The value range is 0-1. The scoring value of the frequency of occurrence of abnormal data is determined according to the following formula, set:

[0072]

[0073] Among them, E represents the score value of the frequency of abnormal data occurrence, and f represents the frequency of abnormal data occurrence.

[0074] Specifically, the adjusting of the ratio D of structured data to unstructured data in the data setting unit according to the comparison result of the importance evaluation value P and the preset importance evaluation value P0 includes:

[0075] If P≥P0, the data adjustment unit adjusts the ratio D of the structured data to the unstructured data according to the adjustment coefficient X1;

[0076] If P<P0, the data adjustment unit does not adjust the ratio of the structured data to the unstructured data.

[0077] In the embodiment of the present invention, the adjustment coefficient X1 is Wherein ΔP represents the difference between the importance evaluation value P and the preset importance evaluation value P0, and the adjusted ratio of structured data to unstructured data is Dd=D×X1.

[0078] Specifically, the data cleaning unit is used to extract abnormal data and normal data from the business data, and perform cleaning processing on the abnormal data and the normal data.

[0079] In an embodiment of the present invention, the cleaning process of the abnormal data includes changing or removing the abnormal data. For example, when the abnormal information is customer information data, it can be replaced with correct information; when the abnormal information is the claim amount, the claim amount information can be changed based on historical similar cases.

[0080] In the embodiment of the present invention, the cleaning process of the normal data includes removing duplicate content in the data.

[0081] In an embodiment of the present invention, the data processed by the data cleaning unit is input into a model for training, and the accuracy of the model is determined.

[0082] Specifically, the adjustment module determines the adjustment of the model or the data processing process according to the accuracy of the model, including determining the accuracy evaluation value Q of the model, and determining the adjustment of the model or the data processing process according to the comparison result of the difference ΔQ between the accuracy evaluation value Q and the preset accuracy evaluation value Q0 and the preset accuracy evaluation value difference ΔQ0.

[0083] In the embodiment of the present invention, the accuracy evaluation value Q is calculated according to the following formula:

[0084] Q=α×A+β×H

[0085] Among them, Q represents the accuracy evaluation value, A represents the accuracy of the decision result, H represents the proportion of normal data in the business data, and α and β represent weight coefficients.

[0086] In the embodiment of the present invention, the weight coefficients α and β satisfy α+β=1, wherein the value of α is 0.5 and the value of β is 0.5.

[0087] Specifically, determining the adjustment of the model or the data processing process according to the comparison result of the difference ΔQ between the accuracy evaluation value Q and the preset accuracy evaluation value Q0 and the preset accuracy evaluation value difference ΔQ0 includes:

[0088] If ΔQ≤ΔQ0, the adjustment module adjusts the preset consistency data ratio M0 in the abnormality analysis unit according to the adjustment coefficient X2;

[0089] If ΔQ>ΔQ0, the adjustment module adjusts the model.

[0090] In the embodiment of the present invention, the preset accuracy evaluation value Q0 is 0.8, and the preset accuracy evaluation value difference ΔQ0 is 0.1.

[0091] In the embodiment of the present invention, the adjustment coefficient X2 is ΔQ represents the difference between the accuracy evaluation value Q and the preset accuracy evaluation value Q0, and the adjusted preset consistency data ratio M0 is Mm=M0×X2.

[0092] In an embodiment of the present invention, the adjustment module adjusts the model by adjusting the parameters of the model, for example, randomly selecting a parameter combination within a specified range, and evaluating the performance of the model under each parameter combination to find the parameter combination that optimizes the performance.

[0093] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

[0094] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A digital decision-making system for claims operation platform based on large model agent, characterized in that: include: A data collection module, which is used to collect business data from each business terminal; A data processing module, connected to the data acquisition module, for processing the business data, including: a related data determination unit, used to determine whether there is related data in the business data, and to determine the type of the related data; a data setting unit connected to the associated data determining unit, and configured to determine the setting of the ratio of structured data to unstructured data of the business data according to the type of the associated data; an abnormality analysis unit connected to the data setting unit, used to determine the consistency of the associated data in the business data and determine the proportion of consistent data in the business data, so as to determine abnormality analysis of the business data according to the proportion of consistent data; a data cleaning unit connected to the abnormality analysis unit, for extracting abnormal data and normal data from the business data, and performing cleaning processing on the abnormal data and the normal data; A data adjustment unit, connected to the data cleaning unit, for determining an adjustment of the ratio of structured data to unstructured data of the business data according to the importance of the abnormal data; An adjustment module is connected to the data processing module and is used to perform deep learning of the model on the business data cleaned by the data cleaning unit and determine the accuracy of the model, so as to determine the adjustment of the parameters of the model or the adjustment of the data processing process according to the accuracy of the model.

2. According to claim 1, the digital decision-making system for the claims operation platform based on the large model agent is characterized in that: The data setting unit determines the processing of the business data according to the type of the associated data, including determining the proportion of sensitive data in the associated data, so as to determine the proportion of structured data and unstructured data in the business data to be a first preset proportion under the condition that the proportion of sensitive data is less than or equal to the preset sensitive data proportion.

3. According to claim 1, the digital decision-making system for the claims operation platform based on the large model agent is characterized in that: The data setting unit determines, according to the type of the associated data, that processing of the business data includes determining a proportion of sensitive data in the associated data, so as to determine a proportion of structured data to unstructured data in the business data to be a second preset proportion under the condition that the proportion of sensitive data is greater than a preset sensitive data proportion.

4. According to claim 1, the digital decision-making system for the claims operation platform based on the large model agent is characterized in that: The abnormality analysis unit determines the abnormality analysis of the business data according to the proportion of the consistency data, including analyzing all business data in the business data to determine whether there is abnormal data under the condition that the proportion of the consistency data is less than or equal to a preset consistency data proportion.

5. According to claim 1, the digital decision-making system for the claims operation platform based on the large model agent is characterized in that: The abnormality analysis unit determines the abnormality analysis of the business data according to the proportion of the consistency data, including analyzing key data in the business data to determine whether there is abnormal data under the condition that the proportion of the consistency data is greater than a preset consistency data proportion.

6. The digital decision-making system for the claims operation platform based on the large model agent according to claim 1 is characterized in that: The data adjustment unit determines the adjustment of the data setting unit according to the importance of the abnormal data, including adjusting the ratio of structured data to unstructured data in the data setting unit according to a comparison result between the importance evaluation value and a preset importance evaluation value.

7. The digital decision-making system for the claims operation platform based on the large model agent according to claim 6 is characterized in that: The step of determining, based on a comparison result between the importance evaluation value and a preset importance evaluation value, to adjust the ratio of the structured data to the unstructured data in the data setting unit includes adjusting the ratio of the structured data to the unstructured data according to an adjustment coefficient under the condition that the importance evaluation value is greater than or equal to the preset importance evaluation value.

8. The digital decision-making system for the claims operation platform based on the large model agent according to claim 1 is characterized in that: The adjustment module determines the adjustment of the model or the data processing process according to the accuracy of the model, including determining the adjustment of the model or the data processing process according to the difference between the accuracy evaluation value of the model and the preset accuracy evaluation value and the comparison result of the difference between the preset accuracy evaluation values.

9. The digital decision-making system for the claims operation platform based on the large model agent according to claim 8 is characterized in that: The adjustment of the model or the data processing process determined based on the comparison result of the difference between the accuracy evaluation value of the model and the preset accuracy evaluation value and the preset accuracy evaluation value difference includes adjusting the preset consistency data ratio in the abnormal analysis unit according to the adjustment coefficient under the condition that the difference between the accuracy evaluation value and the preset accuracy evaluation value is less than or equal to the preset accuracy evaluation value difference.

10. The digital decision-making system for the claims operation platform based on the large model agent according to claim 8 is characterized in that: Determining the adjustment of the model or the data processing process based on the comparison result of the difference between the accuracy evaluation value and the preset accuracy evaluation value and the preset accuracy evaluation value difference includes adjusting the parameters of the model under the condition that the difference between the accuracy evaluation value and the preset accuracy evaluation value is greater than the preset accuracy evaluation value difference.

Images