Information processing device, information processing method and program for supporting insurance underwriting appraisal

Abstract

To provide a technique allowing an insurance applicant to easily receive an underwriting appraisal with an appropriate application content.SOLUTION: An information processing device for supporting insurance underwriting appraisals, includes: user information acquisition means for acquiring medical insurance information of a target user of an insurance underwriting appraisal from a database for storing information of users collected and managed by a public mechanism; first result acquisition means for acquiring a first determination result of whether or not the insurance can be underwritten by using notification information acquired when the target user applies for the insurance; and output means for outputting a second determination result of whether or not the insurance can be underwritten by using the acquired medical insurance information of the target user. Herein, when the first determination result represents that the insurance cannot be underwritten, the output means inputs the medical insurance information of the target user to a learned risk prediction model to predict a risk of the target user, and outputs the second determination result based on the predicted risk.SELECTED DRAWING: Figure 11

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program for supporting insurance underwriting assessment. [Background technology]

[0002] In recent years, a technology has been known that improves the efficiency of underwriting assessment work by providing appraisers who assess whether or not to accept insurance applications with prediction results from artificial intelligence models and reference information related to assessments (Patent Document 1).

[0003] Patent Document 1 proposes a technology to predict the results of medical assessment using an artificial intelligence model that performs machine learning on insurance application information using application details and assessment results in past cases as learning data. In Patent Document 1, in addition to the assessment results of the medical assessment, noteworthy factors to be used in judging the medical assessment are identified, and similar cases searched for based on the identified noteworthy factors are presented to the human assessor. The technology proposed in Patent Document 1 can make the medical assessment work of the assessor more efficient by presenting similar cases from the past, and can reduce the number of cases for which it is not possible to determine whether or not to accept the insurance, compared to conventional human assessments. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6828209 Summary of the Invention [Problem to be solved by the invention]

[0005] However, it is known that among applicants who are rejected according to the contents of the application notice, there are applicants who are in good health and not at high risk, and it is desirable to appropriately evaluate the risk of the applicant and underwrite the insurance. In Patent Document 1, when the machine appraisal cannot determine whether to reject or not reject the application, a prediction by an artificial intelligence model and reference information are provided to the appraiser, thereby making the human appraisal more efficient, but the above-mentioned viewpoint was not taken into consideration.

[0006] The present invention has been made in consideration of the above problems, and has an object to realize a technique that can appropriately evaluate the risk of an insurance applicant. [Means for solving the problem]

[0007] In order to solve this problem, for example, the information processing device of the present invention comprises: An information processing device for supporting insurance underwriting assessment, A user information acquisition means for acquiring medical insurance information of a user who is a target of an insurance underwriting assessment from a database that stores user information collected and managed by a public system; a first result acquisition means for acquiring a first determination result of whether or not the insurance can be accepted by using notification information acquired when the target user applies for the insurance; and an output means for outputting a second determination result of whether or not the insurance can be accepted, using the acquired medical insurance information of the target user; The output means is characterized in that when the first judgment result indicates that underwriting is not possible, the output means inputs the medical insurance information of the target user into a trained risk prediction model to predict the risk of the target user, and outputs the second judgment result based on the predicted risk. Effect of the Invention

[0008] According to the present invention, it is possible to appropriately evaluate the risk of an insurance applicant. [Brief description of the drawings]

[0009] [Figure 1]FIG. 1 is a diagram showing an example of an underwriting assessment support system according to an embodiment of the present invention. [Diagram 2] FIG. 1 is a diagram for explaining an outline of the operation of the underwriting assessment support system according to the present embodiment. [Diagram 3] FIG. 1 is a block diagram showing an example of a functional configuration of an information processing device according to an embodiment of the present invention; [Figure 4] FIG. 1 is a block diagram showing an example of the functional configuration of a communication device (user terminal) according to an embodiment of the present invention; [Diagram 5] FIG. 2 is a block diagram showing an example of the functional configuration of a communication device of an underwriting insurance company according to the present embodiment; [Figure 6A] FIG. 13 is a diagram showing an example of specific medical checkup information that can be acquired from a database that stores a user's medical insurance information. [Figure 6B] A diagram showing examples of data items such as drug information that can be obtained from a database that stores a user's medical insurance information. [Figure 7] FIG. 13 is a diagram showing an example of a correspondence relationship between notification items in notification information and medical insurance information according to the present embodiment; [Figure 8] FIG. 13 is a diagram showing an example of notification information displayed on a communication device (user terminal) according to the embodiment; [Figure 9] A diagram for explaining the flow of underwriting assessment according to the present embodiment. [Figure 10A] Flowchart (1) showing a series of operations in the underwriting assessment process according to the present embodiment [Figure 10B] Flowchart (2) showing a series of operations in the underwriting assessment process according to the present embodiment [Figure 11] FIG. 1 is a diagram illustrating a risk prediction model according to the present embodiment. [Figure 12] FIG. 1 is a diagram for explaining selection of information to be input to a risk prediction model according to the present embodiment. [Figure 13] FIG. 1 is a diagram for explaining the relationship between multiple types of risks, multiple types of underwriting scores, and underwriting suitability determination according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, the embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the invention according to the claims, and not all combinations of features described in the embodiments are essential to the invention. Two or more features among the multiple features described in the embodiments may be arbitrarily combined. In addition, the same reference numbers are used for the same or similar configurations, and duplicated descriptions are omitted.

[0011] (Configuration of the underwriting assessment support system) First, with reference to FIG. 1, the configuration of an underwriting assessment support system according to an embodiment of the present invention will be described.

[0012] The underwriting assessment support system includes, for example, a communication device 110 used by a user 20 who applies for insurance, an information processing device 100, and a communication device 120 used by a company that underwrites the insurance application (also simply referred to as an underwriting insurance company). Furthermore, with the user's prior consent, the information processing device 100 can access, via a predetermined API, a medical insurance information database 130 that stores the user's medical insurance information collected and managed by a public mechanism. As an example, the medical insurance information database 130 is a Mynaportal database that collects and manages medical insurance information, and information on the user's daily health checkups and medical treatments is collected from medical institutions and stored.

[0013] The information processing device 100 is, for example, a server managed by an underwriting insurance company, or a server managed by a company that provides services related to insurance underwriting assessment to an underwriting insurance company. In the following description, no particular distinction is made between the underwriting insurance company and the company that provides the service, and the description will be given simply as the underwriting insurance company.

[0014] In response to receiving an insurance application from the communication device 110, the information processing device 100 performs processes such as generating a notice described below and determining whether or not the application can be accepted. The information processing device 100 then transmits the final result of the determination of whether or not the application can be accepted to, for example, the communication device 120 of the underwriting insurance company. Depending on the result of the determination of whether or not the application can be accepted, the underwriting insurance company can make a final appraisal of the underwriting of the user's insurance, or (for example, if it is impossible to determine whether or not the application can be accepted) can have an underwriting appraisal conducted by an appraiser with more specialized knowledge.

[0015] Although details will be described later, in this embodiment, a notification form that would conventionally be filled out by a user when applying for insurance is generated in advance by the information processing device 100 and transmitted to the communication device 110, and the contents are checked by the user. The notification form transmitted to the communication device 110 is generated based on medical insurance information acquired by the information processing device 100 from the medical insurance information database 130, and therefore the contents are not dependent on the user's subjectivity or memory.

[0016] For users, the task of simply checking the contents of the declaration form eliminates the need to fill out the declaration form, making it easier to apply for insurance. Also, for underwriting insurance companies, the declaration form based on the above-mentioned objective data can be used to perform underwriting assessments. This reduces the workload of appraisers who, in the past, had to check with applicants about unreliable declaration contents when performing underwriting assessments. Also, even when the contents of digitized declaration forms are processed using a computer, it becomes possible to perform assessments with greater accuracy and accumulate more reliable learning data.

[0017] The communication device 110 is, for example, a user terminal capable of installing and executing an application for performing various insurance procedures provided by an underwriting insurance company. The communication device 110 uses the application to display the received notification information on a display unit and accepts confirmation by the user. The communication device 110 transmits, for example, information indicating that the user has confirmed the notification information to the information processing device 100. Note that, although the above example has been described taking as an example a case in which the communication device 110 executes an application provided by an underwriting insurance company, the communication device 110 may access a website provided by the information processing device 100 to acquire and display the notification information, and transmit the user's confirmation result to the information processing device 100.

[0018] In addition, in this embodiment, a case where the communication device 110 is, for example, a smartphone will be described as an example. However, the communication device 110 is not limited to a smartphone, and may be other terminals capable of displaying and checking the notice, such as a desktop computer, a notebook computer, or a tablet terminal.

[0019] The communication device 120 of the underwriting insurance company is a communication device used by an appraiser who performs, for example, an underwriting appraisal at the underwriting insurance company, and is, for example, a personal computer or a tablet terminal. The communication device 120 of the underwriting insurance company can display, on a display unit, notification information from the user and the result of the underwriting judgment made by the information processing device 100. The appraiser operates the communication device 120 to, for example, finalize a final judgment on the underwriting of insurance based on the judgment result made by the information processing device 100 and the result of his or her own human appraisal.

[0020] (Overview of the operation of the underwriting assessment support system) Next, an outline of the operation of the underwriting assessment support system will be described with reference to Fig. 2. It is assumed that medical insurance information of the user 20 of the communication device 110 is stored in the medical insurance information database 130. This outline of the operation is started, for example, when an application from the communication device 110 for a specific insurance product provided by an underwriting insurance company is accepted by the information processing device 100.

[0021] First, in S201, the communication device 110 grants the information processing device 100 access permission to the medical insurance information database (user information). The method of granting the access permission may be a known method. The access permission may be granted when applying for the above-mentioned predetermined insurance product.

[0022] In S202, the information processing device 100 acquires medical insurance information from the medical insurance information database 130, and generates notification information from the medical insurance information. The generation of the notification information will be described in detail later, but the notification information includes questions regarding whether the user has a medical history, whether they have been examined, treated, or prescribed medication for a specific disease such as a heart or brain disease within the past five years, whether they have been hospitalized, etc. Furthermore, the notification information according to this embodiment includes answers to the questions generated by the information processing device 100. The information processing device 100 transmits the generated notification information to the communication device 110.

[0023] In S203, the communication device 110 displays the notification information on the display unit, and after the user 20 confirms the notification information, accepts an operation indicating the confirmation by the user 20. Upon accepting the user's operation, the communication device 110 transmits information indicating that the notification information has been confirmed to the information processing device 100.

[0024] In S204, in response to receiving information from the communication device 110 indicating that the notification information has been confirmed, the information processing device 100 registers the notification information transmitted to the communication device 110 in S202 in the storage device of the information processing device 100. This registration of the notification information corresponds to obtaining (the notification written by the user 20) via the communication device 110 and registering the data of the notification in the storage device.

[0025] Furthermore, when the information processing device 100 performs an underwriting assessment by applying a rule-based assessment process to the notice information, it further assesses whether or not the notice information can be accepted. The assessment of whether or not the notice information can be accepted is classified into, for example, "not declined" (acceptable), "rejected" (not acceptable), and "unable to assess whether or not the notice information can be accepted." As an example, the rule-based assessment process for the notice information can assess whether or not the notice information can be accepted as "not declined" if none of the questions in the notice information apply, and can assess whether or not the notice information has a medical history as "rejected." In addition, when the case is neither of the above, it can assess whether or not the notice information can be accepted as "unable to assess whether or not the notice information can be accepted." Note that, although an example is described in which the rule-based assessment process is performed using registered notice information to assess whether or not the notice information can be accepted, the assessor may assess whether or not the notice information can be accepted by referring to the notice information.

[0026] If the result of the underwriting judgment using the notification information in S204 is "unacceptable," in S205, the information processing device 100 performs risk prediction using a risk prediction model that uses medical insurance information as an input and judges whether or not to accept the insurance. The risk prediction model is, for example, a statistical model or a machine learning model. Even if a user is rejected based on the judgment based on the notification information, there may be a user who is not so high risk if evaluated from another perspective. For this reason, in this embodiment, the risk of the user is (re)evaluated using the medical insurance information and the risk prediction model, so that it is possible to rescue a user who has been rejected based on the judgment of whether or not to accept the insurance based on the notification information. For the underwriting insurance company, it is also possible to evaluate the risk of the user from multiple angles, and to more appropriately underwrite the insurance.

[0027] If the result of the underwriting judgment using the notification information in S204 is "unable to determine whether or not the insurance can be accepted," in S206, the information processing device 100 performs rule-based processing on the medical examination result information included in the medical insurance information to determine whether or not the insurance can be accepted. The rule-based processing in S206 is not only intended to provide a risk assessment different from that in S204, but is also beneficial in that it can narrow down users who are "unable to determine whether or not the insurance can be accepted." By narrowing down users who are "unable to determine whether or not the insurance can be accepted," it is possible to narrow down the targets of human assessment by experts, thereby making the assessment work more efficient.

[0028] In S207, the information processing device 100 generates notification information for display in which the information included in the notification information is formatted so as to be displayed in a predetermined format (for example, a format for storage on paper or a format commonly used for display) so that the adjuster can easily determine whether or not to accept the claim. The information processing device 100 transmits the generated notification information for display to the communication device 120 of the underwriting insurance company. Note that, in the explanation of this step, an example has been given in which the notification information is the processing target, but other information may be used as long as it is information that the adjuster uses to determine whether or not to accept the claim, and the information processing device 100 may generate the medical checkup information for display using medical insurance information such as the user's medical checkup results.

[0029] In S208, the communication device 120 displays the display notification information received from the information processing device 100 on the display unit. At this time, the appraiser who performs the underwriting assessment refers to the display information displayed on the display unit and determines whether or not to underwrite the loan. In addition, the communication device 120 accepts the assessment result by the appraiser and transmits the accepted assessment result to the information processing device 100.

[0030] In S209, the information processing device 100 outputs the final underwriting result, which is either "not rejected" or "rejected", obtained in any of S204 to S208. The output may be registered in a storage unit of the information processing device 100, or may be displayed on the communication device 120. Also, the result may be transmitted to the communication device 110 as the application result.

[0031] (Configuration of information processing device) Next, a functional configuration example of the information processing device 100 will be described with reference to Fig. 3. Each of the functional blocks described may be integrated or separated, and the functions described may be realized by another block. Also, what is described as hardware may be realized by software, and vice versa.

[0032] The communication unit 301 includes a communication circuit that communicates with various devices via a network. The communication unit 301 receives information processed by the control unit 304 from a communication partner device (e.g., the communication device 110, etc.) and transmits information processed by the control unit 304 to a communication partner device (e.g., the communication device 120). The power supply unit 302 is a power supply that provides power required for the operation of the information processing device 100.

[0033] The storage unit 303 includes a non-volatile storage medium such as a hard disk or a semiconductor memory, and includes various programs executed by the control unit 304 of the information processing device 100, and various data and DBs used by the control unit 304. The various programs include an operating system, a framework, a library, and the like, in addition to a program for executing the underwriting assessment process according to this embodiment. The various data include, for example, setting values ​​of the information processing device 100 and parameter values ​​after learning of each risk prediction model. The storage unit 303 also stores notification information generated and registered for each user, and learning data for each model for learning a risk prediction model.

[0034] The control unit 304 includes a CPU 310, which is a central processing unit, and a RAM 311. The control unit 304 loads a program stored in the storage unit 303 into the RAM 311 and executes it to control the operation of each unit inside the control unit 304 and to control the operation of each unit of the information processing device 100. The control unit 304 also executes an underwriting assessment process, which will be described later.

[0035] The RAM 311 includes a volatile storage medium such as a DRAM, and temporarily stores parameters and processing results for the control unit 304 to execute a program.

[0036] The medical insurance information acquisition unit 320 acquires medical insurance information of the user (user 20) who is the subject of the insurance application from the medical insurance information database 130 via the communication unit 301. The medical insurance information includes, for example, specific medical checkup information 601 shown in Fig. 6A, drug information 611 shown in Fig. 6B, drug information, prescription information, and dispensing information 612, and medical fee notification information 613, and may further include information (not shown) including the notified injury or illness name indicating the existence of a medical history. It may also include information (not shown) including information regarding the user's disability.

[0037] The specific health check information shown in Fig. 6A includes, for example, information on the specific health check taken at a health check institution (measurements and answers in the health check). The specific health check information includes, for example, information in tables such as qualification information, specific health check reception information, health check information basic information, health check information doctor's judgment, health check information questionnaire (specific health check), and specific health check questionnaire (health check for the elderly). Among the specific health check information, health check information questionnaire (specific health check) 602 further includes information such as medication 1 (blood pressure), ... medical history 1 (cerebrovascular), ..., smoking, .... Note that the information in the table of the specific health check information shown in Fig. 6A shows information included in the medical insurance information of the known My Number Portal as an example of this embodiment.

[0038] Moreover, information included in the medical insurance information, such as drug information 611, drug information, prescription information, dispensing information 612, and medical fee notification information 613, has the configuration shown in Fig. 6B. For example, drug information 611 includes information on drugs prescribed by an insurance medical institution, insurance pharmacy, etc. (prescription information and dispensing information are not included). Drug information 611 includes information such as the insurer number, prescription issue date, and examination identification etc. category shown in drug information 614. For example, the examination identification etc. category includes a code corresponding to a category indicating the examination identification etc., that is, a procedure, surgery, imaging diagnosis, basic hospitalization fee, etc.

[0039] The drug information, prescription information, and dispensing information 612 includes, for example, information on drugs prescribed at an insurance medical institution or insurance pharmacy (including prescription information and dispensing information). The medical fee notification information 613 includes, for example, information on medical fees paid at an insurance medical institution or insurance pharmacy. Each piece of information shown in FIG. 6B is also shown as an example of the present embodiment, which is information included in the known My Number Portal medical insurance information. By using the medical insurance information acquired by the medical insurance information acquisition unit 320, the information processing device 100 can acquire, for example, the notified injury or illness name of the user 20, information on examinations and hospitalizations, the measured values ​​of each item in the medical checkup, the presence or absence of a smoking habit, the presence or absence of a physical functional disorder, and the like. Note that in this embodiment, an example is described in which the medical insurance information acquisition unit 320 acquires the above-mentioned medical insurance information from the medical insurance information database 130. However, the medical insurance information acquisition unit 320 may acquire the medical insurance information from the medical insurance information database 130, and may further acquire other information used by the notification information generation unit 321, the risk prediction unit 326, etc. from other databases to use as the medical insurance information used in this embodiment. The other information may be, for example, information (not shown) including a notified injury or illness name indicating the existence of a medical history, information (not shown) including information about the user's disability, or information about the user's medical treatment, but is not limited to these.

[0040] The notification information generating unit 321 generates notification information to be presented to the user 20. For example, as shown in FIG. 7 as an example, the notification information generating unit 321 can associate information 702 obtained from medical insurance information with each item of notification items 701 of the notification information in advance. The notification information generating unit 321 can generate notification information by referring to the medical insurance information according to the predetermined correspondence 700. For example, the notification information generating unit 321 can extract the medical history of the user 20 from the information of the notified injury / illness name, and can extract the presence or absence of examinations or hospitalization for the target disease, etc. from the medical treatment identification, etc. classification. The notification information generating unit 321 can also obtain the presence or absence of physical functional disorders from information not shown, and can obtain the items of abnormalities detected in the medical checkup, the presence or absence of smoking habits, etc. from the specific medical checkup information. In this way, the notification information generating unit 321 can generate notification information without receiving input from the user 20 on the notification form.

[0041] The notification information is transmitted to the communication device 110 and displayed on the display unit. FIG. 8 shows an example of the notification information generated by the information processing device 100 and displayed on the communication device 110. As described above, the notification information includes questions regarding, for example, whether there is a medical history, whether there has been an examination, treatment, medication, etc. for a specific disease such as a heart or brain disease within the past five years, whether there has been hospitalization, etc. Furthermore, the notification information according to this embodiment includes an answer to the question generated by the information processing device 100. For example, in the example shown in FIG. 8, a second question 802 out of multiple questions is displayed on the notification confirmation screen 801. Furthermore, an answer 803 generated by the information processing device 100 is displayed on the notification confirmation screen 801 as an answer to the question 802. In this example, a medical history extracted by the information processing device 100 from the notified injury or illness name of the medical insurance information is displayed as an answer. The notification confirmation screen 801 includes information 804 indicating that the answer has been generated from the medical insurance information. When the user presses the confirmation button 805, the communication device 110 accepts the user's operation and transmits to the information processing device 100 information indicating that the notification information has been confirmed.

[0042] The confirmation information acquisition unit 322 receives information indicating that the notification information has been confirmed, which is transmitted from the communication device 110. The notification information registration unit 323 registers in the storage unit 303 the generated notification information.

[0043] For example, when performing an underwriting assessment by applying a rule-based determination process to the registered notice information, the first underwriting availability acquisition unit 324 determines whether or not to accept the application and acquires one of the following results: not rejected, rejected, or unable to determine whether or not to accept the application. As described above, the appraiser may determine whether or not to accept the application by referring to the notice information, and in this case the first underwriting availability acquisition unit 324 acquires the result of the appraiser's determination of the underwriting assessment.

[0044] The formatting unit 325 performs processing corresponding to the above-mentioned S207. That is, the formatting unit 325 generates display notification information in which information included in the notification information is formatted so that it is displayed in, for example, a format when stored on paper or a format commonly used for display, so that the appraiser can easily determine whether or not to accept the claim. The formatting unit 325 may generate display medical checkup information for medical insurance information such as the user's medical checkup results.

[0045] The risk prediction unit 326 performs risk prediction and judgment of underwriting using a risk prediction model that inputs medical insurance information (e.g., medical checkup information and prescribed drug information). The risk prediction unit 326 performs risk prediction and judgment of underwriting when the judgment result acquired by the first underwriting availability acquisition unit 324 indicates unacceptable. The risk prediction unit 326 inputs the medical insurance information of the user 20 into a risk prediction model to predict the risk of the user 20 over time (e.g., cumulative risk after one year, two years, n years, etc.). The risk over time may be, for example, the probability of hospitalization after a predetermined period of time, the probability of surgery after a predetermined period of time, etc. The risk prediction model is modeled or learned based on data that associates past medical insurance information (e.g., medical checkup information and prescribed drug information) with actual medical procedures (hospitalization, surgery, etc.).

[0046] The safety degree calculation unit 327 converts the risk predicted by the risk prediction unit 326 into an underwriting score. The underwriting score is a score indicating the relative degree of safety within a predetermined group to which the user belongs, for example, a score indicating the degree of safety when compared with users of the same sex and age. The safety degree calculation unit 327 calculates, for example, a deviation value of risk in a group of users of the same sex and age that has been calculated in advance as the score of the user 20. Note that the safety degree calculation unit 327 may use other numerical values ​​indicating deviations or distribution biases in the user group as the score.

[0047] The second underwriting possibility acquisition unit 328 may output either "rejected" or "not rejected (acceptable)" depending on the score calculated by the safety degree calculation unit 327. Details of the risk prediction unit 326, the safety degree calculation unit 327, and the second underwriting possibility acquisition unit 328 will be described later.

[0048] The control unit 304 outputs a final result of the underwriting assessment based on any one of the underwriting adequacy determination acquired by the first underwriting adequacy acquisition unit 324, the underwriting adequacy determination by the second underwriting adequacy acquisition unit 328, and the underwriting adequacy determination received from the communication device 120 in S208. The output may be a registration in a storage unit of the information processing device 100, or a display on the communication device 120. Also, it may be transmitted to the communication device 110 as a result of the application.

[0049] The relationship between the judgment results of a plurality of underwriting assessments will be described with reference to FIG. 9. FIG. 9 shows the operation from the point when the notice information registration unit 323 registers the notice information in the storage unit 303. 901 shows the judgment of underwriting in S204 described above. If the judgment result in 901 is "rejected", the judgment of underwriting assessment is terminated. If the judgment in 901 is "rejected", the underwriting assessment in 902, that is, the judgment of underwriting assessment using the risk prediction model in S205, is executed. On the other hand, if the judgment result in 901 is "unable to judge whether or not underwriting is possible", the underwriting assessment in 903, that is, the underwriting assessment by S206 and S208 is executed. In the underwriting assessment in 902, the application by the user 20 is either rejected or not rejected depending on the judgment result (it may be subject to final confirmation by an assessor separately).

[0050] In addition, in the rule-based judgment for the medical insurance information in 903, for example, each item of the medical checkup result in the medical insurance information is evaluated to judge whether the insurance is acceptable or not as follows: Not accepted, Rejected, or Unacceptable (human assessment is performed) is judged. In the judgment using the medical checkup result, for example, a ranking is determined in advance for each item of the medical checkup result. For example, for the item of systolic blood pressure, a ranking is determined as follows: {Test value 129 or less: A judgment, Test value 130-139: B judgment, Test value 140-159: C judgment, Test value 160 or more: D judgment}. Then, if all items of the medical checkup result are judged as A, it is judged as Not accepted, and if there is even one D judgment, it is judged as Rejected. In cases other than the above, it is judged as Unacceptable (human assessment is performed). With this judgment, the judgment of Not accepted and Rejected becomes the final judgment. Also, in the case of "Unacceptable or Unacceptable", the judgment of the expert appraiser becomes the final judgment result of the underwriting assessment.

[0051] (Configuration of communication device) Furthermore, a configuration example of the communication device 110 will be described with reference to Fig. 4. Fig. 4 shows a functional configuration example of the communication device 110 of this embodiment. Each of the functional blocks described may be integrated or separated, and the functions described may be realized by another block. Also, what is described as hardware may be realized by software, and vice versa.

[0052] The communication unit 401 includes, for example, a communication circuit, and communicates with the information processing device 100 via mobile communication such as LTE, or via wireless communication such as WiFi, to transmit and receive necessary data.

[0053] Operation unit 403 includes buttons and a touch panel provided in communication device 110, and accepts operations on a notice confirmation screen 801 displayed on display unit 405. Note that the GUI of notice confirmation screen 801 may be the GUI of a browser application that displays information on a network, or may be the GUI of an application installed in communication device 110.

[0054] The display unit 405 includes a display panel such as an LCD or an OLED, and displays a GUI for various operations. For example, the display unit 405 displays the above-mentioned notice confirmation screen 801.

[0055] The recording unit 406 includes a non-volatile memory such as a semiconductor memory, and stores set user information, programs executed by the control unit 402, and the like.

[0056] The control unit 402 includes a CPU 410 and a RAM 411, and controls the operation of each functional block in the control unit 402 and each unit in the communication device 110 by the CPU 410 executing a program recorded in the recording unit 406, for example.

[0057] The notice information display unit 412 acquires the notice information transmitted from the information processing device 100 via the communication unit 401, and displays it on the display unit 405. The notice information display unit 412 displays the notice information as the notice confirmation screen 801 described with reference to FIG.

[0058] The confirmation information transmission unit 413 transmits, in response to a user operation on the notice confirmation screen 801, to the information processing device 100, information indicating that the above-mentioned notice information has been confirmed.

[0059] (Configuration of communication equipment of underwriting insurance company) Furthermore, a configuration example of the communication device 120 of the underwriting insurance company will be described with reference to Fig. 5. Fig. 5 shows a functional configuration example of the communication device 120 of this embodiment. Each of the functional blocks described may be integrated or separated, and the functions described may be realized by different blocks. Also, what is described as hardware may be realized by software, and vice versa.

[0060] The communication unit 501 includes, for example, a communication circuit, and communicates with the information processing device 100 via mobile communication such as LTE, or via wireless communication such as WiFi, to transmit and receive necessary data.

[0061] The operation unit 503 includes buttons and a touch panel provided in the communication device 110, and accepts operations by the assessor to display notification information and medical examination results, and input of the results of the personal assessment by the assessor. The display unit 505 includes a display panel such as an LCD or OLED, and displays a GUI for various operations. For example, the display unit 505 displays notification information and medical examination results for the assessor to perform a personal assessment.

[0062] The recording unit 506 includes a non-volatile memory such as a semiconductor memory, and stores set user information, programs executed by the control unit 502, and the like.

[0063] The control unit 502 includes a CPU 510 and a RAM 511, and controls the operation of each functional block in the control unit 502 and each unit in the communication device 110 by the CPU 510 executing a program recorded in the recording unit 506, for example.

[0064] (A series of operations for underwriting assessment processing in information processing device) Next, a series of operations of the underwriting assessment process in the information processing device will be described with reference to Figures 10A and 10B. Each operation described in this series of operations is realized by the control unit 304 loading a program stored in the storage unit 303 into the RAM 311 and executing it, thereby causing each unit of the control unit 304 described above to function.

[0065] In S1001, the medical insurance information acquisition unit 320 acquires the medical insurance information of the target user from the medical insurance information database 130. In S1002, the notification information generation unit 321 extracts information from, for example, predetermined reference information of the medical insurance information described above with reference to Fig. 7, and generates notification information. In S1003, the notification information generation unit 321 transmits the generated notification information to the communication device 110 of the user 20.

[0066] In S1004, the confirmation information acquisition unit 322 determines whether or not information indicating that the notification information has been confirmed by the user has been received from the communication device 110. When the confirmation information acquisition unit 322 receives the information, in S1005, the notification information registration unit 323 registers the confirmed notification information in the storage unit as notification information acquired from the user.

[0067] In S1006, the first underwriting availability acquisition unit 324 acquires the result of the rule-based processing of the registered notification information or the judgment result of the underwriting appraisal by the appraiser. In S1007, the first underwriting availability acquisition unit 324 judges whether the judgment result is non-rejection. If the judgment result is non-rejection, the first underwriting availability acquisition unit 324 ends this series of processing. If the judgment result is not non-rejection, the first underwriting availability acquisition unit 324 proceeds to S1008.

[0068] In S1008, the first underwriting availability acquisition unit 324 further determines whether the determination result is rejection. If the determination result is rejection, the first underwriting availability acquisition unit 324 advances the process to S1013, and if not, advances the process to S1009. In S1009, the control unit 304 acquires the underwriting assessment determination result by rule-based processing of the medical insurance information, and in S1010, determines whether the determination result indicates that it is not possible to determine whether or not the insurance can be accepted. If the determination result indicates that it is not possible to determine whether or not the insurance can be accepted, the control unit 304 advances the process to S1011, and if not (i.e., if the determination result is rejection or not rejection), this series of operations ends.

[0069] In S1011, the formatting unit 325 generates display notification information by formatting the registered notification information into the format of existing notification information, and in S1012, transmits the generated display notification information to the communication device 120 used by the appraiser.

[0070] In S1013, risk prediction using a model for medical insurance information and underwriting processing are executed. After the underwriting processing is executed, the control unit 304 ends this processing.

[0071] (Details of risk prediction and underwriting decision) Further, the risk prediction model according to this embodiment will be described with reference to Fig. 11. The risk prediction unit 326 predicts the user's risk over time for each predetermined time period using multiple risk prediction models, such as the risk prediction model 1102. Specifically, the multiple risk prediction models are configured with individual risk prediction models associated with each predetermined time period (e.g., one year), and output cumulative risks for one year, two years, . . . 10 years, respectively. The set of models for each predetermined time period is configured for multiple types of risk (target risk) (e.g., "risk of hospitalization," "risk of surgery").

[0072] In each risk prediction model, information selected from the medical insurance information 1101 according to each model (for example, information on the history of medical checkups, medication, and diagnosis) is input. FIG. 12 shows how inputs are selected for each model. The medical insurance information 1101 is selected in advance from multiple perspectives, such as the way of dividing years, the counting method, and the granularity of classification, regarding information on medical checkup history, medication history, and medical treatment. Furthermore, input data with high importance is selected according to the relationship with the target risk. Note that the information selected from the medical insurance information 1101 can be determined in advance by experiment or learning. In addition, for example, the information selected may differ between a one-year cumulative risk model and a two-year cumulative risk model in the prediction model 1102 for the risk of hospitalization.

[0073] In this way, by dividing the risk prediction models into predetermined time periods and providing appropriate inputs to each risk prediction model, it is possible to predict risk with high accuracy for each time period.

[0074] Referring again to FIG. 11, by plotting the risks predicted by the individual risk prediction models associated with each predetermined time period (for example, one year), the user's risk over time can be predicted. Specifically, by finding a curve 1103 (function) that fits the output of the plotted multiple risk prediction models, the user's risk over time, including 10 years from now, can be predicted. In this way, the user's risk over time can be predicted with high accuracy.

[0075] FIG. 13 shows the relationship between multiple types of risks (multiple target risks), multiple underwriting scores, and underwriting judgment (underwriting judgment for each insurance product). For example, the hospitalization occurrence risk is used to calculate the hospitalization (lump sum) underwriting score. As described above, in order to calculate the hospitalization occurrence risk, multiple risk prediction models for calculating the hospitalization occurrence risk are used.

[0076] As described above, the safety calculation unit 327 calculates the underwriting score by using the risk calculated by the risk prediction model. At this time, the safety calculation unit 327 calculates, as an example, the underwriting score of the user 20, a deviation value of the risk in a group of users of the same sex and age that has been calculated in advance.

[0077] Furthermore, the safety calculation unit 327 may combine the results of multiple types of risk prediction models to obtain a certain underwriting score. For example, the hospitalization occurrence risk and the hospitalization length risk may be used to calculate a hospitalization (daily benefit) underwriting score. For example, the safety calculation unit 327 calculates the smaller (or larger) of the standard deviation value of the hospitalization occurrence risk and the standard deviation value of the hospitalization length risk, or the weighted average, as the underwriting score, as an example. In this way, by combining the individual results of multiple types of risk prediction models to calculate the underwriting score, the underwriting score can be calculated with high accuracy even when it is difficult to directly and accurately predict a certain underwriting score from one type of risk prediction model.

[0078] The second underwriting possibility acquisition unit 328 judges whether or not the hospitalization lump sum rider can be accepted (outputs "rejected" or "not rejected"), for example, according to the hospitalization (lump sum) underwriting score calculated by the safety degree calculation unit 327. For example, the second underwriting possibility acquisition unit 328 judges to reject when the underwriting score is greater than a threshold value previously determined for each judgment, and outputs "not rejected" when the underwriting score is not greater than a threshold value.

[0079] The above-mentioned underwriting admissibility judgment has been described taking as an example a case where the judgment is based on one underwriting score, but as shown in FIG. 13, the underwriting admissibility judgment (for example, the underwriting admissibility judgment for the main contract (hospitalization / surgery)) may be performed using a combination of multiple underwriting scores. In this case, the second underwriting admissibility acquisition unit 328 may use the lowest (or highest) underwriting score among the multiple underwriting scores, or may use a score obtained by weighted averaging. In other words, if the obtained underwriting score (for example, the lowest underwriting score) is greater than a threshold value previously determined for each judgment, the second underwriting admissibility acquisition unit 328 judges the application to be rejected, and if not, outputs a non-rejection.

[0080] As described above, in this embodiment, the notification information acquired when the target user applies for insurance is used to acquire a first judgment result that judges whether or not the insurance can be accepted. Then, if the first judgment result indicates that the insurance cannot be accepted, the user's medical insurance information acquired from a database that stores user information collected and managed by a public mechanism is input into the trained risk prediction model to predict the user's risk, and a second judgment result is output based on the predicted risk. In this way, even if a user is rejected as a result of the judgment according to the application contents such as a notification, risk prediction is performed using the user's medical insurance information (including, for example, medical examination, medication, and diagnosis history), and the risk of the user can be reevaluated from a different perspective and the risk of the user can be appropriately evaluated. In other words, the risk of the insurance applicant can be appropriately evaluated.

[0081] The invention is not limited to the above-described embodiment, and various modifications and variations are possible within the scope of the gist of the invention. [Explanation of symbols]

[0082] 100...information processing device, 110...communication device, 120...communication device of underwriting insurance company, 130...medical insurance information database, 320...medical insurance information acquisition unit, 324...first underwriting suitability acquisition unit, 326...risk prediction unit, 327...safety degree calculation unit, 328...second underwriting suitability acquisition unit

Claims

1. User information acquisition means for acquiring medical insurance information of a user subject to insurance underwriting assessment, The system includes an output means that inputs the target user's medical insurance information into a trained risk prediction model to predict the target user's risk, and outputs a determination result that determines whether or not the insurance can be underwritten based on the predicted risk. The information processing device is characterized in that the risk prediction model predicts the time-dependent risk of the target user at predetermined intervals using a plurality of risk prediction models, and the plurality of risk prediction models are composed of individual risk prediction models associated with each predetermined interval.

2. The information processing apparatus according to claim 1, characterized in that the output means calculates a score indicating the relative degree of safety within a predetermined group to which the target user belongs, based on the predicted risk of the target user, and outputs the determination result based on a determination using the score.

3. The information processing apparatus according to claim 2, characterized in that the output means predicts the time-series risk of the target user for each of the multiple types of risk using a multiple types of risk prediction model, and calculates the score based on the combination of the multiple types of risks that have been predicted.

4. The information processing apparatus according to claim 2, characterized in that the output means calculates multiple types of scores based on multiple types of risks predicted by multiple types of risk prediction models, and outputs the judgment result using a combination of the multiple types of scores.

5. The information processing device according to claim 1, characterized in that the output means inputs information on the history of health checkups, medications, and diagnoses included in the medical insurance information of the target user into the risk prediction model.

6. The information processing device according to claim 1, characterized in that the user information acquisition means acquires medical insurance information of the target user for insurance underwriting assessment from a database that stores user information collected and managed by a public system.

7. To obtain medical insurance information of a user subject to insurance underwriting assessment, The system inputs the medical insurance information of the target user into a trained risk prediction model to predict the user's risk, and outputs a determination result that determines whether or not the insurance can be underwritten based on the predicted risk. The information processing method is characterized in that the risk prediction model uses a plurality of risk prediction models to predict the time-dependent risk of the target user at predetermined time intervals, and the plurality of risk prediction models are composed of individual risk prediction models associated with each predetermined time interval.

8. A program for causing a computer to function as one of the means of an information processing device according to any one of claims 1 to 6.

Images