Method for evaluating the risk of commercial insurance underwriting and system comprising the same

Abstract

This document discloses a method and system for assessing underwriting risks in commercial insurance. The method utilizes a risk decision engine to provide underwriting risk decision information. This risk decision engine comprises a risk assessment module, a machine learning module, an expert rule module, and a geographic risk analysis module. By integrating these four modules, it provides underwriting risk decision information to improve the accuracy of predicting underwriting risks.

Description

[Technical Field]

[0001] This invention belongs to the field of insurance technology, and in particular to a method for intelligent analysis of underwriting decisions and a system for executing the method. [Previous Technology]

[0002] Commercial insurance projects are complex, and the underwriting and risk assessment process is intricate. Based on the insured object, it can be mainly divided into three aspects: property risk, liability risk, and employee risk. Furthermore, the types of insured objects are diverse. Currently, the underwriting process is mainly done manually, or the underwriting system lacks cross-insurance calculation integration, making the underwriting process time-consuming and inefficient. Moreover, given that commercial insurance is a highly specialized category, a significant amount of time and resources must be invested in the education and training of underwriters to ensure they fully understand the relevant knowledge of commercial insurance underwriting; otherwise, it will be impossible to provide comprehensive protection to meet customer needs.

[0003] In commercial insurance underwriting, which requires a high degree of expertise, the prior art lacks suitable intelligent commercial insurance underwriting risk assessment methods and systems to perform or assist in the process, highlighting the urgent need for improvements in prior art. Therefore, there is a pressing need in the field for improved underwriting methods and systems to address these shortcomings. [Summary of the Invention]

[0004] The present invention is intended to provide a simplified summary of this disclosure so that the reader can have a basic understanding of it. This present invention is not a complete overview of this disclosure, and it is not intended to identify important / critical elements of the embodiments of the invention or to define the scope of the invention.

[0005] This invention relates to a method and system for assessing risks in commercial insurance underwriting. Through a novel process design, this invention can efficiently complete commercial insurance underwriting. By integrating machine learning algorithms, it improves the accuracy of intelligent calculations. Furthermore, through interface settings, it can comprehensively grasp the risks associated with various types of commercial insurance, reducing the risk of errors caused by human intervention.

[0006] This disclosure relates to a commercial insurance underwriting risk assessment method, applicable to a commercial insurance underwriting risk assessment system and a user terminal communication connection, wherein the commercial insurance underwriting risk assessment system includes a customer contribution database and a risk decision engine that are communicationally connected to each other, the method comprising the following steps: (a) providing the risk decision engine, which includes a risk assessment module, a machine learning module, an expert rule module and a geographic risk analysis module electrically coupled to each other, wherein the machine learning module is trained by: providing a first training model formed from a first training dataset, wherein the first training dataset contains multiple insurance application decision information within a first period; providing a second training model formed from a second training dataset, wherein the second training dataset contains multiple insurance application decision information within a second period, wherein the time periods of the second period and the first period partially overlap; Extract at least one thousand of the most recent insurance decision information from the second training dataset, and conduct tests using the first training model and the second training model to obtain an accuracy rate relative to the first training model and the second training model, respectively. If the accuracy rate of the first training model is higher than that of the second training model, the first training model is used for analysis; if the accuracy rate of the second training model is higher than that of the first training model, the second training model is used for analysis. The risk assessment module and the expert rule module each include a first risk calculation rule and a second risk calculation rule. (b) Receive an insurance request information provided by the user terminal, and obtain underwriting risk decision information through the risk decision engine, the risk assessment module, the machine learning module, and the expert rule module. The insurance request information includes information on the insured object and information on an insurance plan.

[0007] According to an embodiment of the present invention, in step (b), a first risk value, a second risk value and a third risk value are obtained after analysis by the risk assessment module, the machine learning module and the expert rule module, and underwriting risk decision information is obtained by comparing the first risk value, the second risk value and the third risk value.

[0008] According to another embodiment of the present invention, the underwriting risk decision information includes an underwriting risk decision risk value, and step (b) further includes the following steps: (b1) When the first risk value is higher than a first risk preset value, the underwriting risk decision risk value is high risk; (b2) When the first risk value is lower than the first risk preset value and the second risk value is lower than a second risk preset value, the risk value of the third risk value is used as the underwriting risk decision risk value; (b3) When the first risk value is lower than the first risk preset value, the second risk value is higher than the second risk preset value, and the third risk value is lower than a third risk preset value, the underwriting risk decision risk value is medium to high risk; (b4) When the first risk value is lower than the first risk preset value, the second risk value is higher than the second risk preset value, and the third risk value is higher than the third risk preset value, the underwriting risk decision risk value is high risk; or (b5) If the first risk value is lower than the first risk preset value, the second risk value is lower than the second risk preset value, and the third risk value is lower than the third risk preset value, then the second risk value is used as the underwriting decision risk value.

[0009] According to one embodiment of the present invention, the first risk calculation rule is to obtain a first risk value by scoring a plurality of risk factors, wherein the risk factors are selected from the group consisting of: insured amount, claims history, product category, insured object, building grade, whether or not an inspection has been conducted, or a combination thereof. Furthermore, the second risk calculation rule includes: scoring a primary insurance policy information to obtain a primary insurance risk value; determining a supplementary insurance risk value for supplementary insurance policy information; and comparing the primary insurance risk value and the supplementary risk value. If the primary insurance risk value is higher than the supplementary risk value, then the primary insurance risk value is used as the third risk value; conversely, if the supplementary risk value is higher than the primary insurance risk value, then the supplementary risk value is used as the third risk value. In addition, the second risk value is obtained through the prediction and analysis of the aforementioned machine learning module, which is trained based on past insurance decision information and a specific method.

[0010] According to one embodiment of the present invention, the underwriting method of the present invention further includes providing a geographic risk analysis module, which includes the following steps: (g-1) using the geographic risk analysis module, based on information of an insured object in the insurance request information, providing geographic information, insurance history information, building information, a natural disaster risk indicator or an adjacent risk information corresponding to the information of the insured object on a map screen; and (g-2) displaying at least one adjacent facility within the area range corresponding to the information of the insured object.

[0011] In an optional embodiment, the area range in step (g-2) is 0 to 1000 meters in diameter. Furthermore, the map image of this invention is composed of a plurality of map layers, wherein each map layer includes an original map layer, an internal data layer, and / or a natural disaster layer information.

[0012] According to a specific embodiment of the present invention, it may further include the following steps: (g-3) receiving a selection request information from the user terminal, wherein the user terminal selects the area range on the map screen as needed; and (g-4) receiving a measurement request information from the user terminal, wherein the user terminal measures the building area of ​​a building within the area range in step (g-2) and / or measures the distance between the target information and an adjacent building on the map screen.

[0013] Furthermore, the method described in any embodiment of the present invention further includes the following steps: (a-1) providing at least one operation interface to the user terminal using a display module, the at least one operation interface displaying the underwriting risk decision information, which includes a risk level, decision recommendations or a risk record list; and (a-2) displaying a map screen and a risk record list on the at least one operation interface using a display module.

[0014] Furthermore, in order to enable users to more effectively assess the risk values ​​between different types of insurance purchased by policyholders, the method of the present invention further includes the following steps: (a-3) using a display module to display multiple types of insurance information corresponding to the information of the insured object and displaying them respectively on the at least one operation interface; and (a-4) the operation interface allows the user to select and interchange different operation interfaces through an operation column.

[0015] In addition, according to other embodiments of the present invention, the present invention further includes the following steps: (a-5) Displaying an additional insurance operation interface using the display module, wherein the additional insurance operation interface displays an additional insurance selection block and an additional insurance application block, wherein the additional insurance application block includes a plurality of additional insurance option keys, and the additional insurance application area includes at least one additional insurance display block, and each additional insurance display block includes an application option key, the name of the insured object, the insured amount or usage information; and (a-6) The user terminal displays or deletes any of the corresponding additional insurance display blocks in grayscale through the additional insurance option key.

[0016] According to an embodiment of the present invention, the method of the present invention further includes the following steps: (e) receiving underwriting decision information from the user terminal, and when the underwriting risk decision information in step (d) is inconsistent with the underwriting decision information, generating a risk decision questionnaire to the user terminal; and (f) receiving feedback information obtained from the user terminal through the risk decision questionnaire, wherein the feedback information serves as a third training data set for the machine learning module.

[0017] Furthermore, another aspect of the present invention relates to a commercial insurance underwriting risk assessment system, which includes a customer contribution database and a risk decision engine that are interconnected. Specifically, the risk decision engine includes four modules: a risk assessment module, a machine learning module, an expert rule module, and a geographic risk analysis module, and is used to execute the method described in any of the above embodiments.

[0018] After reading the following embodiments, those skilled in the art to which this invention pertains will be able to easily understand the basic spirit and other inventive objectives of this invention, as well as the technical means and implementation methods adopted by this invention.

Implementation Method

[0021] To make the description of this disclosure more detailed and complete, illustrative descriptions of embodiments and specific examples of the present invention are provided below; however, this is not the only form of implementing or using the specific examples of the present invention. The embodiments cover features of multiple specific examples and methods and steps for constructing and operating these specific examples, as well as their order. However, other specific examples may also be used to achieve the same or equivalent functions and order of steps.

[0022] Unless otherwise defined in this specification, the scientific and technical terms used herein have the same meaning as understood and commonly used by one of ordinary skill in the art to which this invention pertains. Furthermore, unless conflicting with the context, singular nouns used herein encompass their plural forms, and vice versa.

[0023] To make the description of this disclosure more detailed and complete, illustrative descriptions of embodiments and specific examples of the present invention are provided below; however, this is not the only form of implementing or using the specific examples of the present invention. The embodiments cover features of multiple specific examples and methods and steps for constructing and operating these specific examples, as well as their order. However, other specific examples may also be used to achieve the same or equivalent functions and order of steps.

[0024] It is understood that the "system" typically includes at least some form of storage medium, communication unit, and processing unit. The storage medium includes electrically and non-electrically oriented, removable and non-removable media, which can be used with appropriate methods or techniques to store desired information (such as computer-readable instructions, data structures, application modules, and other data). The storage unit includes, but is not limited to, RAM, ROM, EEPROM, flash memory, or other memory technologies, CD-ROM, digital versatile disc (DVD), or other optical storage, magnetic cartridges, magnetic tapes, magnetic disk storage, and other magnetic storage devices, or any other media capable of storing required information and accessible to the processor. Generally, the communication unit can materialize computer-readable instructions, data, structures, application modules, and other data into various data signals, and can transmit them through any communication medium. By way of example and not limitation, the communication unit includes wired media (such as wired networks or direct wired connections) and wireless media (such as audio waves, infrared rays, radio waves, microwaves, spread spectrum technology, and other wireless media technologies). Furthermore, the communication unit is implemented using a communication chip, such as a component supporting signal transmission for Ethernet, fiber optic communication networks, telecommunications cable networks, Global System for Mobile communication (GSM), Personal Handy-phone System (PHS), Code Division Multiple Access (CDMA) systems, Wideband Code Division Multiple Access (WCDMA) systems, Long Term Evolution (LTE) systems, Worldwide Interoperability for Microwave Access (WiMAX) systems, Wireless Fidelity (Wi-Fi) systems, Bluetooth, or other chips supporting standard communication specifications defined by the Institute of Electrical and Electronics Engineers (IEEE). The processing unit communicates with the communication unit and the storage unit to perform the necessary calculations and functions for commercial insurance underwriting risk assessment methods.In one embodiment of this disclosure, the processing unit may be, for example, a central processing unit (CPU), or other programmable general-purpose or special-purpose microprocessor, digital signal processor (DSP), programmable controller, application-specific integrated circuit (ASIC), or other similar components or combinations thereof, and this disclosure is not limited thereto.

[0025] Furthermore, in this disclosure, the risk decision engine, risk assessment module, machine learning module, expert rule module, and geographic risk analysis module described in this invention are implemented by a processor, memory, and storage device.

[0026] By way of example and not limitation, the “system” refers to a computer device that operates, which may be a desktop computer, server computer, handheld or laptop device, personal digital assistant, multiprocessor system, microprocessor-based system, set-top box, programmable consumer electronics, mobile phone (especially smartphone), network computer, minicomputer, mainframe computer, distributed computing environment including any of the above systems or devices, and similar to them.

[0027] The term "user" as used herein refers to an individual applicable to the commercial insurance underwriting risk assessment method of the present invention, such as an insurance agent. In a preferred embodiment, the user refers to an underwriter.

[0028] To provide better service and to standardize and / or intelligently plan the procedures for commercial insurance underwriting, the present invention provides a method for commercial insurance underwriting and an apparatus for performing the method. The method proposed in this invention systematizes the original cumbersome underwriting process and presents it with novel intelligent algorithms, functional integration, and a specific layout, enabling users to be more efficient in the underwriting process and to make underwriting decisions more consistent.

[0029] Figure 1 is a schematic diagram of a commercial insurance underwriting risk assessment system 100 according to an embodiment of the present invention. The commercial insurance underwriting risk assessment system 100 has a risk decision engine 120 and a customer contribution database 150 electrically coupled to each other. The risk decision engine 120 includes a risk assessment module 122, a machine learning module 124, an expert rule module 126, and a geographic risk analysis module 128. In addition to the machine learning module 124, the risk decision engine 120 of this system uses the risk assessment module 122 and the expert rule module 126, which are based on a first risk calculation rule and a second risk calculation rule, respectively. The geographic risk analysis module 128 can present relevant information about the insured object in a map-based graphical form, thereby more accurately assessing the underwriting risk. Furthermore, the present invention also considers the original policyholder's past insurance records, and combined with the settings of the customer contribution database 150, can provide comprehensive underwriting decisions based on customer attributes.

[0030] Figure 2 illustrates the training method of the machine learning module 124 in the risk decision engine 120 of the present invention. It should be noted that the machine learning module 124 of the present invention is trained through an iterative automatic update mechanism. First, the present invention provides a first training model, which is formed based on a first training dataset, wherein the first training dataset contains multiple insurance application decision information within a first period (step 212). Furthermore, the present invention provides a second training model, which is formed using a second training dataset. Similarly, the second training dataset contains multiple insurance application decision information within a second period (step 214). The first dataset and the second dataset continuously collect multiple insurance application decision information according to the continuation of time, and the first and second periods corresponding to the first dataset and the second dataset partially overlap (step 214). Next, a verification test is performed, extracting at least one thousand of the latest insurance application decision information from the second training dataset, and testing is conducted using the first training model and the second training model to obtain the accuracy relative to the first training model and the second training model, respectively (step 216). If the accuracy of the first training model is higher than that of the second training model, the first training model is used for analysis; if the accuracy of the second training model is higher than that of the first training model, the second training model is used for analysis (step 218), thus maintaining the computational performance of the machine learning module of the present invention in a timely manner. Furthermore, in a non-limiting embodiment, the machine learning module employs a decision tree model. In addition, the first or second period is at least one, two, three, four, five, six, seven, eight, nine, or ten years. In a preferred embodiment, the first or second period is at least three years; in a more preferred embodiment, the first or second period is at least five years.

[0031] Figure 3A illustrates a commercial insurance underwriting risk assessment method according to an embodiment of the present invention, which is executed by the commercial insurance underwriting risk assessment system 100 shown in Figure 1. In the commercial insurance underwriting risk assessment method of the present invention, machine learning is used as a foundation, and the insured object is utilized as the key to underwriting risk assessment. That is, information about the insured object itself, such as geographical information, historical insurance information, building information, natural disaster risk indicators, or adjacent risk information, is referenced to generate insurance risk decision information.

[0032] First, the commercial insurance underwriting risk assessment system 100 of the present invention receives insurance request information input by a user (e.g., an insurance agent) through a user terminal (step 312), wherein the insurance request information includes information about the insured object and insurance plan information. Next, the risk decision engine 120 analyzes the information to obtain underwriting risk decision information and provides it to the user terminal (step 314). It should be noted that the risk decision engine 120 of the present invention, in addition to the machine learning module 124, also includes a risk assessment module 122, an expert rule module 126, and a geographic risk analysis module 128, which respectively calculate the corresponding underwriting risk value through a first risk calculation rule and a second risk calculation rule. The risk decision engine 120 of the present invention performs intelligent risk diagnosis, judging underwriting risk factors according to the insurance request information, for example, calculating the assessed risk value based on relevant risk factors such as the industry of the insured object and the nature of the insured object. Simultaneously, the geographic risk analysis module 128 examines the actual location of the insured object and the risks existing in its surrounding environment.

[0033] In step 314, the risk decision engine 120 obtains a first risk value, a second risk value and a third risk value after analysis by the risk assessment module 122, the machine learning module 124 and the expert rule module 126, and generates underwriting risk decision information by comparing the first risk value, the second risk value and the third risk value.

[0034] Specifically, the first risk calculation rule is to obtain a first risk value by scoring a plurality of risk factors, wherein these risk factors are selected from the following groups: insured amount, claims history, product category, insured object, building grade, whether or not an inspection has been conducted, or a combination thereof. Additionally, it should be noted that the claim information includes information on the insured object and insurance plan information, and the insurance plan information includes, but is not limited to, primary insurance application information and supplementary insurance application information. The second risk calculation rule includes the following steps: scoring a primary insurance application information to obtain a primary insurance risk value; determining a supplementary insurance risk value for a supplementary insurance application information; and comparing the primary insurance risk value and the supplementary risk value. If the primary insurance risk value is higher than the supplementary risk value, the primary insurance risk value is used as the third risk value; conversely, if the supplementary risk value is higher than the primary insurance risk value, the supplementary risk value is used as the third risk value.

[0035] The following is a second risk calculation rule according to an embodiment of the present invention: The second risk calculation rule will evaluate the main insurance and the supplementary insurance to obtain the third risk value. The calculation logic of the main insurance risk value is shown in Table 1 below.

[0036] Table 1 Scoring parameters Business Nature Classification (Initial Score) Initial score (A) Insurance combination Adding and multiplying fractions (B) Building Grade Adding and multiplying fractions (C) Dangerousness of goods Adding or subtracting fractions (D) Connecting iron sheet Adding or subtracting fractions (E) Insurance range Adding or subtracting fractions (F)

[0037] The risk value of the main insurance policy is assessed using the expert rule score: A*(1+B+C)+D+E+F. The risk value of the supplementary insurance policy is determined based on the corresponding level of the sum assured. As mentioned above, the risk value of the main insurance policy and the risk value of the supplementary insurance policy are compared. If the risk value of the main insurance policy is higher than the risk value of the supplementary insurance policy, then the risk value of the main insurance policy is used as the third risk value; conversely, if the risk value of the supplementary insurance policy is higher than the risk value of the main insurance policy, then the risk value of the supplementary insurance policy is used as the third risk value.

[0038] Furthermore, the second risk value is obtained through the prediction and analysis of the aforementioned machine learning module, which is trained based on past insurance decision information and specific methods.

[0039] Furthermore, according to another embodiment of the present invention, step 314 further includes the following steps to determine the underwriting risk decision risk value included in the underwriting risk decision information: (b1) When the first risk value is higher than the first risk preset value, the underwriting risk decision risk value is high risk; (b2) When the first risk value is lower than the first risk preset value and the second risk value is lower than the second risk preset value, the risk value of the third risk value is used as the underwriting risk decision risk value; (b3) When the first risk value is lower than the first risk preset value, the second risk value is higher than the second risk preset value, and the third risk value is lower than the third risk preset value, the underwriting risk decision risk value is medium to high risk; (b4) When the first risk value is lower than the first risk preset value, the second risk value is higher than the second risk preset value, and the third risk value is higher than the third risk preset value, the underwriting risk decision risk value is high risk; or (b5) When the first risk value is lower than the first risk preset value, the second risk value is lower than the second risk preset value, and the third risk value is lower than the third risk preset value, the second risk value is used as the underwriting risk value.

[0040] Those skilled in the art to which this invention pertains should understand that the setting of the risk preset value can be based on actual usage needs, and appropriate risk preset values ​​can be set according to the type of insured case, the amount of insurance, and the insured object.

[0041] Finally, the user terminal determines the underwriting conditions and underwriting result based on the underwriting risk decision information (step 316). In step 316, the dashed lines represent a human step, which is a further utilization of the risk decision information generated by the underwriting method executed by the system of the present invention. Those skilled in the art should understand that this step is not a requirement for limiting the scope of the patent application of the present invention.

[0042] Furthermore, Figure 3B is a flowchart of an underwriting method according to another embodiment of the present invention. The steps in the figure are the same as those in Figure 3A, and will not be repeated here. In this embodiment, the method of the present invention occurs after the user determines the underwriting conditions and underwriting results. The steps include: the commercial insurance underwriting risk assessment system 100 receives underwriting decision information from the user, and when the underwriting risk decision information in step (d) and the user's underwriting decision information (i.e., underwriting conditions and underwriting results) are inconsistent, a risk decision questionnaire is generated for the user (step 318). Furthermore, the commercial insurance underwriting risk assessment system 100 of the present invention receives feedback information obtained from the user through the risk decision questionnaire, wherein the feedback information serves as a third training dataset for a machine learning module (step 320).

[0043] Furthermore, in the underwriting method proposed in this invention, the geographic risk analysis module 128 is one of the key technologies for improving the underwriting method. It presents insurance and risk information related to the information of the insured object and the insurance plan in the information of the insurance request in a more dynamic and comprehensive manner on a map screen.

[0044] Specifically, please refer to Figure 4, which shows the process executed by the geographic risk analysis module 128 in the risk decision engine 120 of the present invention in the underwriting method of the present invention. First, based on the information of the insured object in the claim request information, geographic information, insurance history information, building information, natural disaster risk indicators or adjacent risk information corresponding to the information of the insured object are provided on the map screen (step 412). Please also refer to Figure 5, which shows the map screen 500 displayed by the system display module of the present invention according to an embodiment of the present invention. It should be noted that those skilled in the art should understand that the new commercial insurance underwriting risk assessment system 100 may further include the display module (not shown in the figure) electrically coupled to the risk decision engine 120 and the customer contribution database 150.

[0045] As shown in the figure, the insured object is marked on the map screen 500, and simultaneously, the map screen 500 can display a geographic environment frame 510 (i.e., containing geographic information and natural disaster risk indicators), an adjacent risk frame 520 (i.e., containing adjacent risk information), and a historical data frame 530 (i.e., containing building information and insurance history information). In addition, the geographic risk analysis module 128 can also be connected with currently known commercial software to display a real-scene photo 540 of the insured object on the map screen 500.

[0046] In an optional embodiment, the building information includes building materials, floor height, building type, intended use, or a combination thereof. Additionally, the company information includes a unique identifier, company name, company address, representative's name, approved establishment date, business operations, or a combination thereof.

[0047] Furthermore, the geographic risk analysis module 128 of the present invention further displays at least one adjacent facility within a region relative to the information to be insured (step 414). In an optional embodiment, the region has a diameter of 0 to 1000 meters. As shown in Figure 5, the region 515 can be represented by a frame, and adjacent facilities within the region can be labeled in different ways, such as hazardous facilities, rescue facilities, recreational facilities, or public facilities. Furthermore, the underwriting risk system 100 can receive selection request information from a user terminal, which selects a region on the map screen 500 as needed (step 416), and can also measure the building area of ​​buildings within the region and / or measure the distance between the information to be insured and an adjacent building as needed (step 418). The radius of the region 515 can be adjusted through the ring area exploration interface 560.

[0048] In an optional embodiment, the map screen 500 is composed of a plurality of map layers, wherein the map layers include an original map layer, an internal data layer and / or natural disaster layer information.

[0049] Furthermore, Figures 6A to 6C illustrate the underwriting method of the present invention according to various embodiments of the present invention, presented through an operation interface 600 via a display module (not shown in the figures). Specifically, the underwriting method of the present invention includes providing at least one operation interface 600 to a user terminal after the display module is executed, wherein the operation interface 600 displays underwriting risk decision information 610. In this embodiment, the underwriting risk decision information 610 is mainly presented in three frames, respectively displaying a risk level 612, a decision suggestion 614, or a risk record list 616, wherein the risk level 612 is presented based on the risk value. In optional embodiments, the risk level may be divided into high risk, medium-high risk, medium risk, and / or low risk, but is not limited thereto. Those skilled in the art to which this invention pertains should understand that the definition of risk level can be adapted to correspond different risk preset values ​​to different risk levels according to actual usage needs, and corresponding decision information can be provided according to different risk levels. Table 2 below is an example proposed according to an embodiment of the present invention:

[0050] Table 2 Risk Value Risk level Decision Recommendations 0-50 High risk Strict underwriting 51-60 Medium and high risk On-site inspection 61-80 Medium risk Set a reasonable price 81-100 Low risk Fast review

[0051] The risk record list includes confirmation items, such as items related to risk factors. For example, whether the insured object is a multi-span corrugated iron structure, the nature of its use, and precautions corresponding to the coverage scope. For example, if it is earthquake insurance, it is necessary to confirm whether the insured object is near a fault zone and / or located in a soil liquefaction zone, whether the insured amount is higher than the preset value, and whether there is cumulative insurance.

[0052] According to another embodiment of the present invention, in order to facilitate the user to check the various information items on the risk record list 616, the present invention further displays the map screen 500 and the risk record list 616 on the operation interface 600 in a side-by-side manner, so that the user can simultaneously check the various precautions on the risk record list 616 and confirm them through the map screen 500.

[0053] In addition, the underwriting method of the present invention also provides a cross-insurance underwriting platform. Based on the insurance request information, including information on the insured object and insurance plan information, a plurality of insurance types information are displayed using a display module and displayed on the operation interface respectively. In Figure 6C, the operation interface 600 of the present invention displays a plurality of operation interfaces 610A, 610B, 610C, 610D and 610E. These operation interfaces 610A, 610B, 610C, 610D and 610E are integrated on an operation column 615, allowing the user to select and interchange different operation interfaces 610A, 610B, 610C, 610D or 610E.

[0054] In an optional implementation, the information on multiple types of insurance is selected from the group consisting of: commercial fire insurance, public accident liability insurance, owners' indemnity contract liability insurance, employer's accident liability insurance, product liability insurance, or a combination thereof.

[0055] Commercial insurance underwriting mainly needs to consider three aspects: property risk, liability risk and employee risk. Therefore, this invention integrates the relevant insurance types of each aspect on the same level of review. Through the operation interface settings shown in Figure 6C, users can consider the corresponding risk values ​​and decision suggestions for different types of insurance to complete the underwriting process.

[0056] In addition, in a non-limiting manner, the underwriting method of the present invention can display a corresponding supplementary insurance item based on the information of the insured object and the insurance plan information in the insurance request information, and present it as a supplementary insurance operation interface.

[0057] In the operation interface, the present invention further represents the supplementary insurance items as blocks. The method includes displaying a supplementary insurance operation interface 700 using a display module, wherein the supplementary insurance operation interface 700 displays a supplementary insurance selection block 710 and a supplementary insurance application block 720. The supplementary insurance selection block 710 includes a plurality of supplementary insurance option keys 711. Those skilled in the art should understand that the types of supplementary insurance option keys can be changed according to the insurance type information. For example, taking commercial fire insurance as an example, the supplementary insurance can be explosion insurance, earthquake insurance, typhoon and flood insurance, theft insurance, water damage insurance, rental loss insurance, and business interruption insurance, etc. Furthermore, the present invention features an easy-to-use design on the supplementary insurance operation interface 700. In the supplementary insurance selection block 710, the first and second supplementary insurance options are presented in grayscale, with an "add" icon (e.g., a "+") next to each option. The third and fourth supplementary insurance options are presented in a prominent, non-grayscale format, with a "delete" icon (e.g., an "X") next to each option. The selected supplementary insurance is simultaneously displayed on the supplementary insurance application block 720. The supplementary insurance application block 720 displays the selected supplementary insurance, specifically in the third supplementary insurance display block 712A and the fourth supplementary insurance display block 712B in Figure 7. In addition to displaying the name of the insured item, each insured item also includes an application option button 714. Furthermore, the supplementary insurance application block 720 can also display information related to the supplementary insurance, such as the insured amount or usage information. If the additional insurance is not added, the user can display or delete any corresponding insured item in grayscale using the insurance option button 714. Furthermore, as mentioned above, the display information of each additional insurance in the additional insurance application block 720 is linked to the additional insurance options in the additional insurance selection block 710. Taking Figure 7 as an example, if the third additional insurance is not added, it can be deleted by clicking the third additional insurance option in the additional insurance selection block 710, and the third additional insurance display block 712A in the additional insurance application block 720 will be displayed in grayscale or directly deleted. Therefore, in one embodiment of the present invention, one of the underwriting methods of the present invention is to pre-introduce additional insurance types and pre-set specific additional insurance information as insured items. The user can then close the additional insurance application block using the additional insurance option button 711 in the additional insurance selection block 710 as needed, or re-add the additional insurance using the additional insurance option button 711 as needed.

[0058] In summary, the steps in the various underwriting methods shown in the embodiments of the present invention can be combined or modified. The underwriting scheme flow and flowchart shown in any of the above embodiments are only examples to illustrate the state of the present invention, but are not the only forms of implementing or applying the specific embodiments of the present invention. [Simplified Explanation of the Diagram]

[0019] To make the above and other objects, features, advantages and embodiments of the present invention more apparent and understandable, the accompanying drawings are described as follows: Figure 1 is a schematic diagram of a commercial insurance underwriting risk assessment system 100 according to one embodiment of the present disclosure; Figure 2 is a flowchart of a machine learning module training method according to one embodiment of the present invention; Figure 3A is a schematic diagram of the commercial insurance underwriting risk assessment system 100 performing an underwriting method according to one embodiment of the present disclosure; Figure 3B is a schematic diagram of the commercial insurance underwriting risk assessment system 100 performing an underwriting method according to another embodiment of the present disclosure; Figure 4 is a flowchart of a method performed using the geographic risk analysis module 128 of the present invention according to different embodiments of the present invention; Figure 5 is a map screen 500 presented through a display module after the geographic risk analysis module 128 of the present invention is executed according to a specific embodiment of the present invention; Figures 6A to 6C are schematic diagrams of an operation interface 600 according to one embodiment of the present invention; and Figure 7 is a schematic diagram of an operation interface 700 according to another embodiment of the present invention.

[0020] In accordance with conventional operating practices, the various features and elements in the figures are not drawn to scale. The way they are drawn is to present the specific features and elements related to the present invention in the best possible way. In addition, similar elements / parts are referred to by the same or similar element symbols in different figures.

Claims

1. A commercial insurance underwriting risk assessment method, applicable to a commercial insurance underwriting risk assessment system and a user terminal communication connection, wherein the commercial insurance underwriting risk assessment system includes a customer contribution database and a risk decision engine communication connection between them, the method comprising the following steps: (a) providing the risk decision engine, which includes a risk assessment module, a machine learning module, an expert rule module and a geographic risk analysis module electrically coupled to each other, wherein the machine learning module is trained by the following method: providing a first training model formed from a first training dataset, wherein the first training dataset includes multiple insurance application decision information within a first period; providing a second training model formed from a second training dataset, wherein the second training dataset includes multiple insurance application decision information within a second period, wherein the time periods of the second period and the first period partially overlap; Extract at least one thousand of the most recent insurance decision information from the second training dataset, and test them using the first training model and the second training model to obtain an accuracy rate relative to the first training model and the second training model, respectively. If the accuracy rate of the first training model is higher than that of the second training model, the first training model is used for analysis; if the accuracy rate of the second training model is higher than that of the first training model, the second training model is used for analysis. The risk assessment module and the expert rule module each contain a first risk calculation rule and a second risk calculation rule, respectively. (b) Receive an insurance request information provided by the user terminal, and analyze it through the risk decision engine using the risk assessment module, the machine learning module, and the expert rule module to obtain a first risk value, a second risk value, and a third risk value, respectively. Compare the first risk value, the second risk value, and the third risk value to obtain underwriting risk decision information. The insurance request information includes information about the insured object and information about an insurance plan. The first risk calculation rule is to obtain the first risk value by scoring multiple risk factors; the second risk calculation rule includes: scoring a main insurance policy information to obtain a main insurance risk value, wherein the main insurance risk value = (initial score for business nature classification) × (1 + multiplier score for insurance combination + multiplier score for building grade) + plus / minus score for cargo hazard + plus / minus score for connected sheet metal buildings + plus / minus score for insurance class; determining a supplementary insurance risk value for supplementary insurance policy information, wherein the supplementary insurance is determined according to the corresponding level of the insured amount class; and comparing the main insurance risk value and the supplementary insurance risk value, if the main insurance risk value is higher than the supplementary insurance risk value, then the main insurance risk value is used as the third risk value, otherwise if the supplementary insurance risk value is higher than the main insurance risk value, then the supplementary insurance risk value is used as the third risk value; (g-1) Using the geographic risk analysis module, based on information about an insured object in the insurance request information, it provides geographic information, insurance history information, building information, a natural disaster risk indicator, or adjacent risk information corresponding to the insured object information on a map screen; (g-2) Displays at least one adjacent facility within the area corresponding to one of the insured object information; (a-1) Using a display module, it provides at least one operating interface to the user terminal, the at least one operating interface displaying the underwriting risk decision information, which includes a risk level, decision recommendations, or a risk record list; (a-2) Using the display module, it displays the map screen and the risk record list side-by-side on the at least one operating interface, allowing the user terminal to simultaneously check the precautions on the risk record list and confirm them through the map screen; (e) Receives underwriting decision information from the user terminal, and when the underwriting risk decision information and the underwriting decision information are inconsistent, generates a risk decision questionnaire for the user terminal; and (f) The system receives feedback information obtained from the user through the risk decision questionnaire, which serves as a third training dataset for the machine learning module.

2. The method as described in claim 1, wherein the underwriting risk decision information includes an underwriting risk decision risk value, and step (b) further includes the following steps: (b1) when the first risk value is higher than a first risk preset value, the underwriting risk decision risk value is high risk; (b2) when the first risk value is lower than the first risk preset value and the second risk value is lower than a second risk preset value, the third risk value is used as the underwriting risk decision risk value; (b3) when the first risk value is lower than the first risk preset value, the second risk value is higher than the second risk preset value, and the third risk value is lower than a third risk preset value, the underwriting risk decision risk value is medium to high risk; (b4) when the first risk value is lower than the first risk preset value, the second risk value is higher than the second risk preset value, and the third risk value is higher than the third risk preset value, the underwriting risk decision risk value is high risk; or (b5) If the first risk value is lower than the first risk preset value, the second risk value is lower than the second risk preset value, and the third risk value is lower than the third risk preset value, then the second risk value is used as the underwriting risk decision risk value.

3. The method as described in claim 1, wherein the risk factors are selected from the group consisting of: sum insured, claims history, product category, insured property, building grade, whether or not an inspection has been conducted, or a combination thereof.

4. The method as described in claim 1, wherein the area in step (g-2) has a diameter of 0 to 1000 meters.

5. The method as described in claim 4 further includes the following steps: (g-3) receiving a selection request from the user terminal, wherein the user terminal selects the area range on the map screen as needed; and (g-4) receiving a measurement request from the user terminal, wherein the user terminal measures the building area of ​​a building within the area range in step (g-2) and / or measures the distance between the target information and an adjacent building on the map screen.

6. The method as described in claim 1, wherein the map screen is composed of a plurality of map layers, wherein the map layers include an original map layer, an internal data layer and / or a natural disaster layer information.

7. The method as described in claim 1 further comprises: (a-3) displaying multiple types of insurance information corresponding to the information of the insured object using the display module and displaying them respectively on the at least one operation interface; and (a-4) allowing the user to select and interchange different operation interfaces through an operation column.

8. The method as described in claim 7, wherein the information on the plurality of insurance types is selected from the group consisting of: commercial fire insurance, public liability insurance, owners' indemnity contract liability insurance, employer's liability insurance, product liability insurance, or a combination thereof.

9. The method of claim 7 further includes: (a-5) displaying an additional insurance operation interface using the display module, wherein the additional insurance operation interface displays an additional insurance selection block and an additional insurance application block, wherein the additional insurance application block includes a plurality of additional insurance option keys, and the additional insurance application area includes at least one additional insurance display block, and each additional insurance display block includes an application option key, the name of the insured object, the insured amount or usage information; and (a-6) the client displays or deletes any of the corresponding additional insurance display blocks in grayscale through the additional insurance option key.

10. A commercial insurance underwriting risk assessment system, communicatively connected to a user terminal, comprising: a customer contribution database; and a risk decision engine communicatively connected to the customer contribution database, wherein the risk decision engine includes a risk assessment module, a machine learning module, an expert rule module, and a geographic risk analysis module electrically coupled to each other, and the machine learning module is trained by: providing a first training model formed from a first training dataset, wherein the first training dataset contains multiple insurance application decision information within a first period; and providing a second training model formed from a second training dataset, wherein the second training dataset contains multiple insurance application decision information within a second period, wherein the time periods of the second period and the first period partially overlap. Extract at least one thousand of the most recent insurance decision information from the second training dataset, and conduct tests using the first and second training models to obtain an accuracy rate relative to each model. If the accuracy rate of the first training model is higher than that of the second training model, the first training model is used for analysis; if the accuracy rate of the second training model is higher than that of the first training model, the second training model is used for analysis. The risk assessment module and the expert rule module each include a first risk calculation rule and a second risk calculation rule. The commercial insurance underwriting risk assessment system is executed using any of the methods described in claims 1-9.

Images