Program, information processing unit and information processing method

Abstract

To improve efficiency of processing on an insurance claim.SOLUTION: An information processing unit acquires claim information on an insurance claim of a user, acquires reference information relating to the insurance claim, estimates a possibility of existence of a fraud in the claim information on the basis of the claim information and the reference information, determines whether or not the possibility meets a first condition that indicates a high possibility of existence of a fraud in the claim information, determines whether or not the possibility meets a second condition that indicates a low possibility of existence of a fraud in the claim information, and executes at least one of: processing to send a first notification to a terminal device of a person in charge, if the possibility was determined as not meeting any of the first and second conditions, the first notification prompting execution of a processing related to the insurance claim; and processing to send a second notification to the terminal device of the person in charge, if the possibility was determined as meeting any of the first and second conditions, the second notification including a processing related to the insurance claim and information on the likelihood of a fraud in the claim information.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

[0002] Patent Document 1 describes a technology for determining whether a person who left a job is fraudulently receiving unemployment benefits based on business establishment information, information indicating employment period, and wage information. Patent Document 2 describes a technology related to a nursing care insurance fraudulent claim examination device that can perform examinations over multiple months and retroactive examinations. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-086937 A [Patent Document 2] JP 2003-308393 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, the techniques described in Patent Documents 1 and 2 do not sufficiently improve the efficiency of processing related to insurance claims.

[0005] In particular, the techniques described in Patent Documents 1 and 2 do not allow for sufficient utilization of the human resources of personnel who execute processes related to insurance claims (for example, approval or rejection of insurance payment). In order to utilize the human resources of personnel, it is desirable to have a computer autonomously process insurance claims for which approval or rejection can be determined at a glance, and for personnel to focus on insurance claim processing that requires advanced judgment skills.

[0006] In view of the above problems, an object of the present invention is to improve the efficiency of insurance claim processing. [Means for solving the problem]

[0007] A program according to one embodiment of the present invention causes a computer to function as: a first acquisition means for acquiring claim information related to a user's insurance claim; a second acquisition means for acquiring reference information related to the insurance claim; an estimation means for estimating the possibility of fraud in the claim information based on the claim information and the reference information; a first determination means for determining whether the possibility estimated by the estimation means satisfies a first condition indicating that there is a high possibility of fraud in the claim information; a second determination means for determining whether the possibility estimated by the estimation means satisfies a second condition indicating that there is a low possibility of fraud in the claim information; a first execution means for executing a process of sending a first notification to a person in charge of the insurance claim to prompt the person in charge of the insurance claim to execute a process related to the insurance claim, when it is determined by the first determination means and the second determination means that the possibility estimated by the estimation means satisfies either the first condition or the second condition; and a second execution means for executing at least one of a process related to the insurance claim and a process of sending a second notification including information regarding the possibility of fraud in the claim information to a terminal device of the person in charge, when it is determined by the first determination means and the second determination means that the possibility estimated by the estimation means satisfies either the first condition or the second condition.

[0008] An information processing device according to another embodiment of the present invention includes a first acquisition unit that acquires claim information related to a user's insurance claim, a second acquisition unit that acquires reference information related to the insurance claim, an estimation unit that estimates the possibility of fraud in the claim information based on the claim information and the reference information, a first determination unit that determines whether the possibility estimated by the estimation unit satisfies a first condition indicating that there is a high possibility of fraud in the claim information, a second determination unit that determines whether the possibility estimated by the estimation unit satisfies a second condition indicating that there is a low possibility of fraud in the claim information, a first execution unit that executes a process of sending a first notification to a person in charge of the insurance claim to prompt the person in charge of the insurance claim to execute a process related to the insurance claim, when it is determined by the first determination unit and the second determination unit that the possibility estimated by the estimation unit satisfies either the first condition or the second condition, and a second execution unit that executes at least one of a process related to the insurance claim and a process of sending a second notification including information regarding the possibility of fraud in the claim information to a terminal device of the person in charge, the process including information regarding the possibility of fraud in the claim information.

[0009] An information processing method according to another aspect of the present invention causes a computer to execute a first acquisition step of acquiring claim information related to a user's insurance claim, a second acquisition step of acquiring reference information related to the insurance claim, an estimation step of estimating the possibility of fraud in the claim information based on the claim information and the reference information, a first determination step of determining whether the possibility estimated by the estimation step satisfies a first condition indicating that there is a high possibility of fraud in the claim information, a second determination step of determining whether the possibility estimated by the estimation step satisfies a second condition indicating that there is a low possibility of fraud in the claim information, a first execution step of executing a process of sending a first notification to a person in charge of the insurance claim to prompt the person in charge of the insurance claim to execute a process related to the insurance claim, when it is determined by the first determination step and the second determination step that the possibility estimated by the estimation step satisfies either the first condition or the second condition, and a second execution step of executing at least one of a process related to the insurance claim and a process of sending a second notification including information regarding the possibility of fraud in the claim information to a terminal device of the person in charge. Effect of the Invention

[0010] The present invention can streamline the processing of insurance claims. [Brief description of the drawings]

[0011] [Figure 1] 1 is a diagram showing an overview of a system according to an embodiment of the present invention; [Diagram 2] 1 is a functional block diagram of an information processing device according to an embodiment of the present invention. [Diagram 3] FIG. 2 is a hardware configuration diagram of the information processing device according to the embodiment. [Figure 4] 10 is a flowchart illustrating an example of an operation of the information processing device according to the present embodiment. [Diagram 5] FIG. 2 is a diagram for explaining an example of an operation of the information processing device according to the embodiment. [Figure 6] FIG. 2 is a diagram for explaining an example of an operation of the information processing device according to the embodiment. [Figure 7] FIG. 2 is a diagram for explaining an example of an operation of the information processing device according to the embodiment. [Figure 8] FIG. 2 is a diagram for explaining an example of a display screen of the information processing device according to the embodiment. [Figure 9] FIG. 2 is a diagram for explaining an example of a display screen of the information processing device according to the embodiment. [Figure 10] FIG. 2 is a diagram for explaining an example of a display screen of the information processing device according to the embodiment. [Figure 11] FIG. 2 is a diagram for explaining an example of an operation of the information processing device according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] A preferred embodiment of the present invention (hereinafter, referred to as the present embodiment) will be described with reference to the accompanying drawings. In each drawing, the same reference numerals denote the same or similar configurations.

[0013] In the present invention, "part," "means," "apparatus," and "system" do not simply mean physical means, but also include cases where the functions of the "part," "means," "apparatus," and "system" are realized by software. Also, the functions of one "part," "means," "apparatus," and "system" may be realized by two or more physical means, devices, and software, and the functions of two or more "parts," "means," "apparatus," and "system" may be realized by one physical means, device, and software.

[0014] <1. System Overview> The system according to this embodiment includes an information processing device 1 that estimates the possibility of fraud in the billing information 138 based on billing information 138 related to an insurance claim of an insured person and reference information 140 related to the insurance claim. The information processing device 1 executes a first process when the possibility of fraud in the billing information 138 is high or low, and executes a second process otherwise.

[0015] The first process is typically at least one of a process related to an insurance claim executed on behalf of the person in charge and a process of sending a notification to the person in charge terminal device 3 including information regarding the possibility of fraud in the billing information 138.

[0016] The second process is typically a process of transmitting a notification to the person in charge terminal device 3 to prompt the person in charge to execute a process related to an insurance claim.

[0017] An overview of the operation of the system according to this embodiment will be described with reference to Fig. 1. The system according to this embodiment includes an information processing device 1, an insured person terminal device 2, and a person in charge terminal device 3.

[0018] First, the information processing device 1 acquires the billing information 138 from the insured person terminal device 2 (S1). The billing information 138 includes various information for the insured person to request compensation from the insurance company.

[0019] Next, the information processing device 1 acquires the reference information 140 from the external storage device 4 (S2). The reference information 140 is information that the information processing device 1 refers to in order to estimate the possibility that fraud exists in the billing information 138.

[0020] Next, the information processing device 1 generates an assessment of the possibility that fraud exists in the billing information 138 based on the billing information 138 and the reference information 140 (S3).

[0021] Next, the information processing device 1 determines whether the evaluation generated in S3 satisfies either a first condition or a second condition (S4). The first condition is typically a condition that applies when there is a high possibility that fraud exists in the billing information 138. On the other hand, the second condition is typically a condition that applies when there is a low possibility that fraud exists in the billing information 138.

[0022] The information processing device 1 executes a first process (S5) when the billing information 138 is highly likely to contain fraud (i.e., when the evaluation satisfies a first condition) or when the billing information 138 is low likely to contain fraud (i.e., when the evaluation satisfies a second condition). As described above, an example of executing the first process is when the information processing device 1 executes a process related to an insurance claim on behalf of a person in charge.

[0023] On the other hand, when the information processing device 1 cannot determine whether the possibility of fraud in the claim information 138 is high or low (i.e., the evaluation does not meet either the first condition or the second condition), it executes the second process (S6). As described above, an example of executing the second process is when the information processing device 1 sends a notification to the person in charge terminal device 3 to prompt the person in charge to execute a process related to the insurance claim.

[0024] To summarize the above operations, the information processing device 1 executes the first process when there is obvious fraud in the billing information 138 or when there is no obvious fraud in the billing information 138 and no judgment by the person in charge is required, and executes the second process when neither of these cases apply (i.e., when judgment by the person in charge is required). In this way, the information processing device 1 can assist the person in processing the insurance claim and make the processing related to the insurance claim more efficient.

[0025] <2. System Functional Configuration> An example of the functional configuration of the system according to this embodiment will be described with reference to FIG.

[0026] The system according to this embodiment includes an information processing device 1, an insured person terminal device 2, a person in charge terminal device 3, an external storage device 4, and a communication network 5. The information processing device 1 is capable of communicating with the insured person terminal device 2, the person in charge terminal device 3, and the external storage device 4 via the communication network 5. Note that the system may include at least some of the insured person terminal device 2, the person in charge terminal device 3, and the external storage device 4 in multiple units.

[0027] [Information processing device 1] The information processing device 1 includes a control unit 10, a storage unit 14, and a network interface unit 18. The control unit 10, the storage unit 14, and the network interface unit 18 are electrically connected via a bus 16. In the following, the information processing device 1 will be described as a device owned by an insurance company.

[0028] [Control unit 10] The control unit 10 functions as an acquisition means 100, an estimation means 102, a determination means 104, a calculation means 106, an execution means 110, a management means 112, a reception means 113, a display means 114 and a selection means 116 by executing various programs stored in the memory unit 14.

[0029] [Acquisition means 100] The acquisition means 100 acquires various information from the insured person terminal device 2, the person in charge terminal device 3, and the external storage device 4. In this embodiment, acquiring information includes making the information processable in the control unit 10. Acquiring information includes, for example, receiving the information from another device and reading the information from the storage unit 14.

[0030] The acquisition means 100 acquires billing information 138 related to an insurance claim of an insured person (which may also be referred to as a user). The acquisition means 100 may acquire the billing information 138 from any of the insured person terminal device 2, the person in charge terminal device 3, and the external storage device 4. The acquisition means 100 may also be referred to as a billing information acquisition means or a first acquisition means 100a.

[0031] The claim information 138 includes various information for the insured to claim compensation from the insurance company. The claim information 138 includes information on a claim ID, a claim date, the insured's name, the date of damage occurrence, the damaged item, the claim amount, the estimated amount required to repair the damage, the state of the damage, the cause of the damage, the location of the damage, an image of the damaged area, and a repair company (i.e., a business that repairs the damage caused to the insured's property), etc. The claim information 138 is typically information input by the insured, but at least a part of the claim information 138 may be information input by a person in charge, or at least a part of the claim information 138 may be information automatically set by the information processing device 1. The claim information 138 may be information generated by performing optical character recognition (which may also be called OCR) on a document written or created by the insured.

[0032] There is a possibility that fraud exists in the billing information 138. A state in which fraud exists in the billing information 138 is typically a state in which the billing information 138 includes false information. Specifically, the billing information 138 may include information that the insured person has falsely input with the aim of fraudulently receiving compensation. An example of an insured person fraudulently receiving compensation is when the insured person falsely claims insurance money by claiming that a loss has occurred, even though the loss has not actually occurred.

[0033] A state in which fraud exists in billing information 138 is, for example, a state in which there is a large discrepancy between the information contained in billing information 138 and information that is objectively considered correct. Information that is objectively considered correct includes information regarding the actual situation of the occurrence of damage related to the insurance claim (site information 140a described below) and information regarding statistics of one or more other insurance claims (statistical information 140b described below), etc.

[0034] Another situation in which there may be fraud in billing information 138 is one in which an insurance claim related to billing information 138, if approved, would result in someone receiving a benefit that they should not have received.

[0035] Moreover, the acquisition means 100 acquires reference information 140 related to the insurance claim. The acquisition means 100 may acquire the reference information 140 from any of the insured person terminal device 2, the person in charge terminal device, and the external storage device 4. The acquisition means 100 may also be referred to as a reference information acquisition means or a second acquisition means 100b.

[0036] The reference information 140 is information used to estimate the possibility of fraud in the billing information 138 by comparing it with the billing information 138. The reference information 140 can also be said to be information that serves as a standard for estimating the possibility of fraud. The reference information 140 includes, for example, various information held by the insurance company (for example, information related to past insurance claims and information related to the relationship between the insured and other insured persons), information related to the weather, information related to road conditions, information obtained from social network services, and information reported in the news, etc. The reference information 140 can also be said to be standard information.

[0037] The reference information 140 may include site information 140a relating to the actual situation of the occurrence of the damage related to the insurance claim. The site information 140a includes, for example, information on the location of the site where the damage occurred, the weather at the site where the damage occurred, the movement of the damaged object from the site, and a satellite image of the damaged object. The site information 140a may be, for example, information acquired by the information processing device 1 from a device attached to the damaged object (for example, a car navigation system attached to a car).

[0038] The reference information 140 may also include statistical information 140b relating to the statistics of one or more other insurance claims. The statistical information 140b includes, for example, information on the average amount and standard deviation of the claims amount of insurance claims accepted by the insurance company in the past, the average time between the damage occurrence date and the claim date and the standard deviation, the presence or absence of related insurance claims (hereinafter referred to as "related insurance claims"), and the performance of the repair company. The statistical information 140b may also include information on whether or not there was fraud in each of the claim information 138 relating to one or more other insurance claims. The statistical information 140b is not limited to information obtained by performing statistical processing on some information, and may be, for example, multiple records relating to past insurance claims. The statistical information 140b may also be referred to as historical information regarding past insurance claims.

[0039] [Estimation means 102] The estimation means 102 estimates the possibility that fraud exists in the billing information 138 based on the billing information 138 and the reference information 140. The estimation means 102 typically estimates that there is a high possibility that fraud exists in the billing information 138 when there is a large difference between the billing information 138 and the reference information 140, and estimates that there is a low possibility that fraud exists in the billing information 138 when there is a small difference between the billing information 138 and the reference information 140. It can also be said that the estimation means 102 estimates that the possibility that fraud exists in billing information 138 that differs greatly from the reference information 140 is higher than the possibility that fraud exists in billing information 138 that differs little from the reference information 140.

[0040] When the process related to the insurance claim includes a first process and a second process executed after the first process, and the claim information 138 includes the first claim information acquired by the time of executing the first process and the second claim information acquired by the time of executing the second process, the estimation means 102 may include a first estimation means for estimating the possibility that fraud exists in the claim information 138 based on the first claim information and the reference information 140, and a second estimation means for estimating the possibility that fraud exists in the claim information 138 based on the second claim information and the reference information 140. That is, the estimation means 102 may estimate the possibility that fraud exists in the claim information 138 at each of two or more different time points. A specific example will be described later.

[0041] Furthermore, the estimation means 102 may estimate the possibility that fraud exists in the billing information 138 based on a machine learning model that has been trained on the statistical information 140b. A specific example will be described later.

[0042] Furthermore, the estimation means 102 may estimate the possibility that fraud exists in the billing information 138 based on information about the performance of the repair company contained in the statistical information 140b. A specific example will be described later.

[0043] Furthermore, when the insurance claim includes a first insurance claim and a second insurance claim made after the first insurance claim, and the statistical information 140b includes information on whether or not there was fraud in the claim information 138 related to the first insurance claim, the estimation means 102 may estimate the possibility that there is fraud in the claim information 138 related to the second insurance claim, further based on the relationship between the first insurance claim and the second insurance claim. The relationship between the first insurance claim and the second insurance claim includes, for example, the relationship between the insured related to the first insurance claim and the insured related to the second insurance claim (e.g., identity, kinship, closeness of address, closeness of age, etc.), the relationship between the time when the first insurance claim was made and the time when the second insurance claim was made, the identity of the damage related to the first insurance claim and the damage related to the second insurance claim, etc.

[0044] [Determination means 104] The determination means 104 determines whether or not the possibility estimated by the estimation means 102 satisfies a first condition indicating that there is a high possibility that fraud exists in the billing information 138. In other words, the determination means 104 determines that the first condition exists when it is estimated that there is a high possibility that fraud exists in the billing information 138. The determination means 104 that determines that there is a high possibility that fraud exists in the billing information 138 can also be called first determination means 104a, black case determination means, or high risk case determination means.

[0045] The first condition may be that a value corresponding to the possibility that fraud exists in the billing information 138, calculated by the calculation means 106 described below, is equal to or greater than a first threshold value.

[0046] The determination means 104 determines whether or not the possibility estimated by the estimation means 102 satisfies a second condition indicating a low possibility of fraud existing in the billing information 138. In other words, the determination means 104 determines that the second condition is met when it is estimated that there is a low possibility of fraud existing in the billing information 138. The determination means 104 that determines that there is a low possibility of fraud existing in the billing information 138 can also be called the second determination means 104b, a white case determination means, or a low-risk case determination means.

[0047] The second condition may be that a value corresponding to the possibility that fraud exists in the billing information 138, calculated by the calculation means 106 described below, is equal to or greater than a second threshold value.

[0048] The first condition and the second condition are conditions that are not satisfied simultaneously for any calculation result by the calculation means 106. That is, typically, the calculation result by the calculation means 106 does not satisfy both the first condition and the second condition.

[0049] [Calculation means 106] The calculation means 106 calculates a value corresponding to the possibility that fraud exists in the billing information 138 estimated by the estimation means 102. The calculation means 106 may calculate a larger value the more likely it is that fraud exists in the billing information 138, and may calculate a smaller value the more likely it is that fraud exists in the billing information 138. It can also be said that the value calculated by the calculation means 106 based on billing information 138 that is more likely to contain fraud is larger than the value calculated by the calculation means 106 based on billing information 138 that is less likely to contain fraud. The value is typically a numerical value. The value can also be said to be an evaluation.

[0050] [Means of Execution 110] The execution means 110 includes a first execution means 110a and a second execution means 110b.

[0051] -First execution means 110a- When the first determination means 104a and the second determination means 104b determine that the possibility estimated by the estimation means 102 does not satisfy either the first condition or the second condition, the first execution means 110a transmits a notification (hereinafter referred to as a "first notification") to the terminal device 3 of the person in charge of the insurance claim, prompting the execution of processing related to the insurance claim. The first execution means 100a can also be called a first transmission means for transmitting the first notification.

[0052] Furthermore, when the person in charge is either the first person in charge or a second person in charge having a role different from the first person in charge, the first execution means 110a may transmit a notification to the person in charge terminal device 3 of at least one of the first person in charge and the second person in charge, selected according to the possibility estimated by the estimation means 102. In other words, the first execution means 110a may transmit a notification to the person in charge terminal device 3 of a different person in charge according to the possibility that fraud exists in the billing information 138.

[0053] -Second execution means 110b- The second execution means 110b executes a second process when the first determination means 104a and the second determination means 104b determine that the possibility estimated by the estimation means 102 satisfies either the first condition or the second condition. Specifically, when the possibility of fraud in the billing information 138 is high or low, the second execution means 110b executes at least one of a process related to an insurance claim and a process of sending a notice (hereinafter referred to as a "second notice") including information regarding the possibility of fraud in the billing information 138 to the person in charge terminal device 3. It can also be said that the second execution means 110b includes a second transmission means for sending the second notice.

[0054] The process related to the insurance claim executed by the second execution means 110b is typically a process related to approval or rejection of the insurance claim. The process related to the insurance claim executed by the second execution means 110b may be a process executed on behalf of the person in charge. In other words, the process to be executed by the person in charge may be reduced by the second execution means 110b executing the process related to the insurance claim.

[0055] A claim denial typically involves a decision not to grant coverage for the claim. If a claim is denied, the insured and the agent are prevented from proceeding further with the claim.

[0056] The processing related to the rejection of an insurance claim may include, in addition to the processing of rejecting the insurance claim, accepting input of information regarding permission for rejection from the person in charge terminal device 3, returning at least a part of the procedure related to the insurance claim, and sending information regarding a notice of rejection to the insured person terminal device 2, etc.

[0057] The process of approving a claim typically involves issuing or deciding to issue coverage for the claim. If the claim is approved, the insured and the agent do not need to proceed further with the claim.

[0058] In addition to the process of approving the insurance claim, the process related to the approval of the insurance claim may also include accepting input of information regarding permission for approval from the person in charge terminal device 3, progressing at least part of the procedure related to the insurance claim, and sending information regarding a notice of approval to the insured person terminal device 2.

[0059] When the second execution means 110b executes the process related to the insurance claim, the second execution means 110b may execute the process related to the insurance claim without sending a notification to the person in charge terminal device 3 (i.e., without receiving any particular input from the person in charge). This allows the process related to the insurance claim to be executed without receiving any intervention from the person in charge, so that the information processing device 1 can improve the efficiency of the process related to the insurance claim.

[0060] When the second execution means 110b executes the process related to the insurance claim, the second execution means 110b may not execute the process related to the insurance claim until the reception means 113, which will be described later, receives input of information regarding permission for the process related to the insurance claim from the person in charge. This allows the information processing device 1 to autonomously execute the process related to the insurance claim after ensuring an opportunity for the person in charge to double-check.

[0061] [Management means 112] The management means 112 is a management means 112 that manages the status of the insurance claim in association with the billing information 138. The status of the insurance claim includes an incomplete status indicating that the processing related to the insurance claim is incomplete, and a completed status indicating that the processing related to the insurance claim is completed. When the second execution means 110b performs processing related to the denial of the insurance claim or processing related to the approval of the insurance claim, the management means 112 may change the status of the insurance claim to a completed status. This allows the person in charge to easily know which processing related to the insurance claim he or she needs to execute (regardless of the information processing device 1), and makes the processing related to the insurance claim more efficient.

[0062] [Method of reception 113] The reception unit 113 receives, from the person in charge, information regarding permission for the processing related to the insurance claim to be executed by the second execution unit 110b. A specific example will be described later.

[0063] [Display means 114] The display means 114 displays information related to the possibility estimated by the estimation means 102. An example of the display screen will be described later.

[0064] [Selection Means 116] The selection means 116 selects the first and second persons in charge according to the possibility estimated by the estimation means 102. In response to this, the first execution means 110a transmits a notification to the first or second person in charge according to the selection by the selection means 116. Note that the second person in charge may be, for example, a person who has more work experience in processing insurance claims than the first person in charge. Alternatively, for example, the second person in charge may be a person who is the superior of the first person in charge.

[0065] The selection means 116 may select the second person in charge when the value calculated by the calculation unit 106 is equal to or greater than the third threshold and equal to or less than the fourth threshold, and may select the first person in charge when the value is equal to or greater than the second threshold and equal to or less than the third threshold, or equal to or greater than the fourth threshold and equal to or less than the first threshold. The magnitude relationship between the first threshold to the fourth threshold is as follows.

[0066] 2nd threshold < 3rd threshold < 4th threshold < 1st threshold

[0067] In other words, the first execution means 110a may send a notification to the second person in charge when it is difficult to determine the possibility that fraud exists in the billing information 138 (for example, when the probability of fraud existence is estimated to be 50%), and may send a notification to the first person in charge when it is relatively easy to determine the possibility that fraud exists in the billing information 138 (for example, when the probability of fraud existence is estimated to be 70%).

[0068] [Storage section 14] The memory unit 14 stores various programs executed by the control unit 10 as well as various information used for the operation of the information processing device 1. The memory unit 14 stores, for example, billing information 138, reference information 140, and information regarding the status managed by the management means 112.

[0069] [Network Interface Section 18] The network interface unit 18 realizes communication with the insured person terminal device 2, the person in charge terminal device 3, and the external storage device 4 via the communication network 5. The communication by the network interface unit 18 may be realized in accordance with various communication standards.

[0070] [Insured person terminal device 2] The insured person terminal device 2 is a terminal device used by the insured person. The insured person terminal device 2 is, for example, an electronic device such as a personal computer, a smartphone, or a tablet terminal. By using the insured person terminal device 2, the insured person can transmit billing information 138 to the insurance company (i.e., the information processing device 1).

[0071] [Staff terminal device 3] The person in charge terminal device 3 is a terminal device used by the person in charge. The person in charge terminal device 3 is, for example, an electronic device such as a personal computer, a smartphone, or a tablet terminal. The person in charge can use the person in charge terminal device 3 to execute processing related to insurance claims.

[0072] [External storage device 4] The external storage device 4 is an external device that the information processing device 1 accesses to obtain various information. The external storage device 4 may be, for example, a device installed in a data center. The external storage device 4 may store billing information 138 and reference information 140. The external storage device 4 may receive information from the information processing device 1, the insured person terminal device 2, and the person in charge terminal device 3, and store the information.

[0073] The external storage device 4 may store information about external services. Specifically, the information processing device 1, the insured person terminal device 2, and the person in charge terminal device 3 are typically included in a system owned by an insurance company, while the external storage device 4 may be, for example, a device of an SNS provider.

[0074] The external storage device 4 is not limited to being a standalone device, and may be a storage unit included in any device.

[0075] <3. Hardware configuration of information processing device 1> 3, an example of a hardware configuration in which the above-described information processing device 1 is realized by a computer 30 will be described. Note that the functions of each device can also be realized by dividing them into a plurality of devices.

[0076] As shown in FIG. 3, the computer 30 includes a processor 300 , a storage device 302 , an input I / F 304 , a data I / F 306 , a communication I / F 308 , and a display device 310 .

[0077] The processor 300 controls various processes in the computer 30 by executing programs stored in the storage device 302. For example, each functional unit included in the control unit 10 of the information processing device 1 can be realized by the processor 300 executing the programs stored in the storage device 302.

[0078] The storage device 302 is a storage medium such as a random access memory (RAM), etc. The RAM temporarily stores the program code of the program executed by the processor 300 and data required when the program is executed.

[0079] The storage device 302 may also be a non-volatile storage medium such as a hard disk drive (HDD) or a flash memory. The storage device 302 stores an operating system and various programs for implementing the above-mentioned configurations. The storage medium storing the various programs may be a non-transitory computer readable medium. In addition, the storage device 302 may store a table for registering various information such as business information, and a DB for managing the table. Such programs and data are loaded into the storage device 302 as necessary, and are referred to by the processor 300.

[0080] The input I / F 304 is a device for receiving input from a user. Specific examples of the input I / F 304 include a camera, a button, a microphone, a keyboard, a mouse, a touch panel, various sensors, a wearable device, etc. The input I / F 304 may be connected to the computer 30 via an interface such as a Universal Serial Bus (USB).

[0081] The data I / F 306 is a device for inputting data from outside the computer 30. A specific example of the data I / F 306 is a drive device for reading data stored in various storage media. The data I / F 306 may be provided outside the computer 30. In that case, the data I / F 306 is connected to the computer 30 via an interface such as a USB.

[0082] The communication I / F 308 is a device for performing data communication with an external device of the computer 30 via the communication network 5 in a wired or wireless manner. The communication I / F 308 may be provided outside the computer 30. In that case, the communication I / F 308 is connected to the computer 30 via an interface such as a USB.

[0083] The display device 310 is a device for displaying various types of information. Specific examples of the display device 310 include a liquid crystal display, an organic EL (Electro-Luminescence) display, and a display of a wearable device. The display device 310 may be provided outside the computer 30. In that case, the display device 310 is connected to the computer 30 via, for example, a display cable. In addition, when a touch panel is adopted as the input I / F 304, the display device 310 can be configured integrally with the input I / F 304.

[0084] The components of the information processing device 1 described in the above embodiment are assumed to realize a predetermined process in cooperation with other hardware by the processor 300 executing a program stored in the storage device 302. In other words, these components are assumed to be software or firmware, and also corresponding hardware, and in both concepts, they are also described as "functions," "means," "parts," "processing circuits," "units," or "modules," and can be interpreted as each of these.

[0085] <4. Operation of the information processing device 1> An example of the operation of the information processing device 1 will be described with reference to Fig. 4. Fig. 4 is a flowchart showing the operation from when the information processing device 1 acquires the billing information 138 to when it causes the execution means 110 to function.

[0086] First, the information processing device 1 acquires the billing information 138 by the acquiring means 100 (S100).

[0087] An example of billing information 138 acquired by information processing device 1 in S100 will be described with reference to Fig. 5. Billing information 138 is information in which a value is associated with each of a plurality of items. In billing information 138, values ​​are associated with each of a plurality of items as follows.

[0088] "Claim ID" ... "100-100001", "Claim date" ... "January 21, 2023", "Insured person's name" ... "Hoken Taro", "Date of damage" ... "January 20, 2023", "Damaged item" ... "Automobile", "Claim amount" ... "1.2 million yen", "Damage status" ... "Unable to move", "Cause of damage" ... "Lightning strike", "Location of damage" ... "Coordinates [X 1 ,Y 1 ]," "Image of damaged area" ... refer to drawing," "Repair company" ... "△△ factory"

[0089] At least some of the values ​​in the claim information 138 may be automatically set by the information processing device 1 or the like, without being input by the insured person. In this example, the values ​​associated with the items "Claim ID", "Claim date", and "Name of insured person" are assumed to be values ​​automatically set by the information processing device 1. Also, the values ​​of the other items are assumed to be input by the insured person himself. The column "Possibility of fraud" will be described later.

[0090] Next, the information processing device 1 sets the status associated with the billing information 138 acquired in S100 to "incomplete" by the management means 112 (S102). That is, the status is set to a state in which the process related to the insurance claim of the billing information 138 can be progressed by at least one of the information processing device 1 and the person in charge.

[0091] Next, the information processing device 1, by using the estimation means 102, refers to the reference information 140 and estimates the possibility that fraud exists in the billing information 138 (S104).

[0092] 5 to 7, an example of a method for estimating the possibility that fraud exists in the billing information 138 based on the billing information 138 and the reference information 140 will be described. Figures 6 and 7 are diagrams showing examples of site information 140a and statistical information 140b, respectively.

[0093] 6 is a diagram showing an example of the site information 140a. In the site information 140a, values ​​are associated with each of a plurality of items as follows.

[0094] "Time when the fault was detected" ... "January 20, 2023, 12:00", "Movement of damaged objects from the site" ... "Yes (coordinates [X 2 ,Y 2 ] → coordinate [X 3 ,Y 3 ]), "Arrange a tow truck" ... "None", "Weather at the site where the damage occurred" ... "Cloudy", "Location of the site where the damage occurred" ... Coordinates [X 2 ,Y 2 ]," "Satellite image of the damage"...see drawing

[0095] 7 is a diagram showing an example of the statistical information 140b. In the statistical information 140b, values ​​are associated with each of a plurality of items as follows.

[0096] "Claim amount" ... "Average: 200,000 yen, standard deviation: 100,000 yen", "Time from occurrence of damage to insurance claim" ... "Average: 2 days, standard deviation: 1 day", "Related insurance claims (other claims by the same insured)" ... "None", "Related insurance claims (claims by people related to the insured)" ... "Yes (same damage, same cause)", "Repair company (□□ factory)" ... "Excellent", "Repair company (△△ factory)" ... "Caution", "Repair company (○○ factory)" ... "Average"

[0097] Of the billing information 138, there is a possibility that there is fraud in the values ​​entered by the insured person. In the column of "Possibility of fraud" in the table of FIG. 5, the possibility of fraud for each value is indicated in three stages: "high," "medium," and "low." Note that for values ​​that are automatically set by the information processing device 1, there is no possibility of fraud being present, so "- (automatic setting)" is indicated. An example of a method for the estimation means 102 to estimate the possibility of fraud for each value will be described below.

[0098] First, according to billing information 138 in Fig. 5, the value "January 20, 2023" is associated with the "Date of Damage" field. According to on-site information 140a in Fig. 6, the value "January 20, 2023, 12:00" is associated with the "Time of Detection of Failure" field. In this case, since the contents of billing information 138 and on-site information 140a match, it is estimated that there is a low possibility that fraud exists in the value associated with the "Date of Damage" field.

[0099] According to the billing information 138 in Fig. 5, the value "automobile" is associated with the "damaged item" item. According to the site information 140a in Fig. 6, the value of an image of an automobile with a malfunction is associated with the "satellite image of damaged item" item. In this case, since the contents of the billing information 138 and the site information 140a match, it is assumed that the value associated with the "damaged item" item is unlikely to include fraud.

[0100] According to the billing information 138 in FIG. 5, the value "1.2 million yen" is associated with the "billing amount" item. According to the statistical information 140b in FIG. 7, the values ​​"average: 200,000 yen, standard deviation: 100,000 yen" are associated with the "billing amount" item. In other words, if we assume that the "billing amounts" of multiple past insurance claims follow a normal distribution, the "billing amount" of the billing information 138 is significantly higher than the average. In this case, it is estimated that there is a medium possibility that fraud exists in the value associated with the "billing amount" item of the billing information 138.

[0101] According to the billing information 138 in FIG. 5, the item "damage status" is associated with a value of "unable to move". According to the site information 140a in FIG. 6, the item "movement of damaged item from site" is associated with a value of "yes (coordinates [X 2 ,Y 2 ] → coordinate [X 3 ,Y 36, the value "arrangement of a tow truck" is associated with the field "nothing." In this case, since the contents of the billing information 138 and the field information 140a are inconsistent (i.e., the insured has not arranged for a tow truck, but has moved a car that has been declared as "undriveable"), it is estimated that there is a high possibility that there is fraud in the value associated with the field "damage status."

[0102] According to billing information 138 in Fig. 5, the value "lightning" is associated with the item "cause of damage." According to site information 140a in Fig. 6, the value "cloudy" is associated with the item "weather at the site where the damage occurred." In this case, since the contents of billing information 138 and site information 140a are inconsistent, it is presumed that there is a high possibility that fraud exists in the value associated with the item "cause of damage."

[0103] According to the billing information 138 in Figure 5, the item "Location of damage" includes "Coordinates [X 1 ,Y 1 According to the site information 140a in FIG. 6, the item "Location of the site where the damage occurred" has the value "Coordinates [X 2 ,Y 2 In this case, since the contents of the billing information 138 and the site information 140a are inconsistent, it is estimated that the value associated with the item "Location of damage occurrence" is highly likely to indicate fraud.

[0104] According to the billing information 138 in Fig. 5, an image of the right headlight of a car that is broken is associated as a value with the "image of the damaged area" field. According to the on-site information 140a in Fig. 6, an image of the left headlight of a car that is broken is associated as a value with the "satellite image of the damaged item" field. In this case, since the contents of the billing information 138 and the on-site information 140a are inconsistent, it is presumed that there is a high possibility that fraud exists in the value associated with the "image of the damaged area" field.

[0105] According to the billing information 138 in Fig. 5, the value "△△ Factory" is associated with the "Repair Company" item. According to the statistical information 140b in Fig. 6, the value "Caution Required" is associated with the "Repair Company (△△ Factory)" item. In this case, although the billing information 138 itself is not false, it is estimated that there is a high possibility that fraud exists in the value associated with the "Repair Company" item, based on the fact that many of the insurance claims in the past in which "△△ Factory" has been in charge of repairs have been fraudulent.

[0106] The above describes a method for estimating the possibility that fraud exists in each value of the billing information 138, but the present invention is not limited to this. For example, the possibility that fraud exists in the billing information 138 may be estimated based on other insurance claims related to the insurance claim related to the billing information 138. For example, according to the statistical information 140b, the value "Yes (same damage, same cause)" is associated with the item "related insurance claim (claim by related person of the insured)". In other words, the insurance claim related to the billing information 138 may have the same content as an insurance claim made in the past by a related person of the insured (for example, a relative), and it is estimated that there is a high possibility that fraud exists in the billing information 138.

[0107] As described above, the estimation means 102 estimates the possibility that fraud exists in the billing information 138 based on the billing information 138 and the reference information 140 (including the site information 140a and the statistical information 140b).

[0108] Next, the information processing device 1 calculates, by the calculation means 106, a value (hereinafter referred to as a "fraud score") corresponding to the possibility estimated by the estimation means 102 (S106). In this embodiment, the fraud score calculated by the calculation means 106 is described as α (where α is an integer between 0 and 100). In this embodiment, the fraud score is a numerical value between 0 and 100, inclusive, with a larger numerical value corresponding to a higher possibility that fraud exists in the billing information 138 and a smaller numerical value corresponding to a lower possibility that fraud exists in the billing information 138.

[0109] Next, the information processing device 1 determines whether or not the possibility estimated by the estimation means 102 satisfies a first condition by the first determination means 104a (S108). In this embodiment, the first condition is that the fraud score is equal to or greater than a first threshold. That is, the first determination means 104a determines whether or not α, which is the fraud score calculated in S106, is equal to or greater than the first threshold.

[0110] If α, which is the fraud score calculated in S106, is equal to or greater than the first threshold (S108 YES), the information processing device 1 executes a process of transmitting a second notification to the person in charge terminal device 3 by the second execution means 110b (S110). That is, the information processing device 1 transmits information to the person in charge that there is a high possibility that fraud exists in the billing information 138 and that caution is required when executing processing related to the insurance claim. This makes it easier for the person in charge to determine the possibility that fraud exists in the billing information 138, and allows the person in charge to execute processing related to the insurance claim more efficiently.

[0111] Furthermore, if the fraud score α calculated in S106 is lower than the first threshold (NO in S108), the information processing device 1 determines, by the second determination means 104b, whether or not the possibility estimated by the estimation means 102 satisfies a second condition (S114). In this embodiment, the second condition is that the fraud score is equal to or lower than a second threshold lower than the first threshold. That is, the second determination means 104b determines whether the fraud score α calculated in S106 is equal to or lower than the second threshold.

[0112] If the fraud score α calculated in S106 is equal to or less than the second threshold (YES in S114), the information processing device 1 causes the second execution means 110b to execute a process for approving the insurance claim (S116). In this embodiment, the second execution means 110b executes a process for approving the insurance claim (for example, settlement of the insurance claim or payment of the insurance money) without sending a notification to the person in charge terminal device 3.

[0113] Next, the information processing device 1 changes the status associated with the billing information 138 to “Complete” by the management means 112 (S118). That is, the processing related to the insurance claim of the billing information 138 is set to a state in which neither the information processing device 1 nor the person in charge needs to proceed.

[0114] Furthermore, if α, which is the fraud score calculated in S106, is greater than the second threshold (S114 YES), the information processing device 1 transmits a first notification to the person in charge terminal device 3 by the first execution means 110a (S120). The first notification is typically a notification that prompts the person in charge of the insurance claim to execute the process related to the insurance claim. More specifically, the first notification includes information that it cannot be said that there is a high or low possibility that fraud exists in the claim information 138, and that a decision or intervention by the person in charge is necessary to proceed with the process related to the insurance claim.

[0115] In summary, a program according to one embodiment of the present invention includes a computer 30 of an information processing device 1, a first acquisition means 100a for acquiring billing information 138 related to a user's insurance claim, a second acquisition means 100b for acquiring reference information 140 related to the insurance claim, an estimation means 102 for estimating the possibility of fraud in the billing information 138 based on the billing information 138 and the reference information 140, a first determination means 104a for determining whether the possibility estimated by the estimation means 102 satisfies a first condition indicating that there is a high possibility of fraud in the billing information 138, a second determination means 104b for determining whether the possibility estimated by the estimation means 102 satisfies a second condition indicating that there is a low possibility of fraud in the billing information 138, and a third determination means 104c for determining whether the possibility estimated by the estimation means 102 satisfies a The device functions as a first execution means 110a that executes a process of sending a first notification to the terminal device 3 of the person in charge of the insurance claim, when it is determined by the first determination means 104a and the second determination means 104b that the possibility estimated by the estimation means 102 does not satisfy either the first condition or the second condition, to encourage the person in charge of the insurance claim to execute processing related to the insurance claim, and a second execution means 110b that executes at least one of a process related to the insurance claim and a process of sending a second notification to the terminal device 3 of the person in charge, the process including information regarding the possibility of fraud in the claim information 138, when it is determined by the first determination means 104a and the second determination means 104b that the possibility estimated by the estimation means 102 satisfies either the first condition or the second condition.

[0116] <5. Effects> The information processing device 1 can improve the efficiency of the processing related to insurance claims. Specifically, the information processing device 1 executes a process to assist the processing related to insurance claims by the person in charge in accordance with the possibility of fraud in the billing information 138. More specifically, the information processing device 1 executes the following operations (1)-(5) in accordance with the possibility of fraud in the billing information 138.

[0117] (1) If there is a high probability that there is fraud in the billing information 138, then processing is performed to deny the insurance claim. According to this operation, the information processing device 1 can autonomously execute the process of rejecting insurance claims that are obviously fraudulent without the need for a judgment from a person in charge. This allows the person in charge to devote their efforts to insurance claims that require human judgment.

[0118] (2) If there is a high possibility that fraud exists in the billing information 138, a process of transmitting a second notification indicating that there is a high possibility that fraud exists in the billing information 138 is executed. This operation reduces the labor required for the person in charge to investigate and determine whether or not there is fraud in the insurance claim. Also, the person in charge can actually confirm whether the judgment result is appropriate before processing the insurance claim. This makes it possible to reduce the labor required by the person in charge while allowing responsible human judgment.

[0119] In addition, since autonomously rejecting an insurance claim by the information processing device 1 as in (1) above may result in an unreasonable outcome for the insured, from the viewpoint of responsibly executing processing related to the insurance claim, the above operation (2) is preferable to the above operation (1).

[0120] (3) If there is a low probability that fraud exists in the billing information 138, processing is performed regarding approval of the insurance claim. According to this operation, for insurance claims that are obviously free of fraud without the need for a judgment from a person in charge, the information processing device 1 can autonomously execute the process for approval. This allows the person in charge to devote their efforts to insurance claims that require human judgment.

[0121] (4) If the likelihood of fraud in the billing information 138 is low, a process is executed to send a second notification indicating that the likelihood of fraud in the billing information 138 is low. This operation reduces the labor required for the person in charge to investigate and determine whether or not there is fraud in the insurance claim. Also, the person in charge can actually confirm whether the judgment result is appropriate before processing the insurance claim. This makes it possible to reduce the labor required by the person in charge while allowing responsible human judgment.

[0122] In addition, since autonomously approving an insurance claim by the information processing device 1 as in (3) above rarely results in an unreasonable outcome for the insured person, from the viewpoint of reducing the workload of the person in charge, the above operation (3) is preferable to the above operation (4).

[0123] (5) If the possibility of fraud in the billing information 138 is neither high nor low, a process is executed to send a first notification to the person in charge to prompt the person in charge to carry out processing related to the insurance claim. This action allows agents to focus their efforts on claims that require more judgment.

[0124] The above (1)-(5) can be combined as appropriate depending on the requirements of the insurance claim system.

[0125] The reference information 140 may include scene information 140a relating to the actual circumstances of the occurrence of the claim damage, and may also include statistical information 140b relating to one or more other claim statistics.

[0126] This allows the information processing device 1 to estimate with higher accuracy the possibility that fraud exists in the billing information 138, using objectively correct information as a standard.

[0127] A program according to one embodiment of the present invention further causes the computer 30 of the information processing device 1 to function as a calculation means 106 that calculates a fraud score corresponding to the possibility estimated by the estimation means 102, and the first condition may be that the fraud score is greater than or equal to a first threshold, and the second condition may be that the fraud score is less than or equal to a second threshold that is lower than the first threshold.

[0128] This allows the information processing device 1 to quantitatively evaluate the possibility that fraud exists in the billing information 138. Furthermore, by using the fraud score in the notification sent by the first execution means 110a or on the display screen displayed by the display means 114, the person in charge can intuitively grasp the possibility that fraud exists in the billing information 138.

[0129] A program according to one embodiment of the present invention further causes the computer 30 to function as a reception means 113 that receives input from a person in charge regarding processing related to an insurance claim, and the second execution means 110b may execute processing related to an insurance claim without the reception means 113 receiving input.

[0130] This allows the information processing device 1 to execute processing related to insurance claims without bothering the person in charge.

[0131] The claims information 138 includes information about businesses recovering damages caused to the insured's property, and the statistical information 140b includes information about the performance of the businesses, and the estimation means 102 may further estimate the possibility of fraud in the claims information 138 based on the information about the performance. The information about the performance of the businesses may be manually recorded by a person in charge.

[0132] This allows insurance claims involving sensitive repairers to be presumed to be fraudulent, regardless of the insured's own intent to commit fraud.

[0133] The insurance claim includes a first insurance claim and a second insurance claim made after the first insurance claim, the statistical information 140b includes information regarding whether or not there was fraud in the claim information 138 related to the first insurance claim, and the estimation means 102 may estimate the possibility of fraud in the claim information 138 related to the second insurance claim further based on the relationship between the first insurance claim and the second insurance claim.

[0134] This allows for the presumption of possible fraud in insurance claims involving other insureds that require caution, regardless of the insured's own intent to commit fraud.

[0135] A program according to an embodiment of the present invention may further cause the computer 30 of the information processing device 1 to function as a management means 112 that manages the status of an insurance claim in association with the billing information 138, the status including an incomplete status indicating that the processing related to the insurance claim is incomplete and a completed status indicating that the processing related to the insurance claim is completed. Furthermore, the management means 112 may change the status to a completed status when the second execution means 110b performs a process related to rejection or a process related to approval.

[0136] This enables the information processing device 1 to manage the status of each of a plurality of insurance claims and to control which insurance claims should be processed by the person in charge.

[0137] The claim information 138 may be generated by performing optical character recognition on a document completed or created by the insured.

[0138] This improves the convenience for the insured person from sending the claim information 138 to receiving the compensation. Also, since it is no longer necessary for the person in charge to manually input documents filled or created by the insured person into the information processing device 1, the person in charge can focus on processing the insurance claim.

[0139] <6. Display screen example> An example of a display screen of the information processing device 1 will be described with reference to Fig. 8 to Fig. 10. The display means 114 may display various information on a display device 310 provided in a device other than the information processing device 1. The display means 114 may display various information on a display device 310 provided in the person in charge terminal device 3, for example. The information processing device 1 may transmit display information for displaying various information.

[0140] 8-9 are examples of display screens for the second notification (i.e., a notification including information regarding the possibility of fraud in the billing information 138). FIG. 10 is an example of a display screen for the first notification (i.e., a notification encouraging the person in charge to execute processing related to the insurance claim). By operating these screens, the person in charge can have the information processing device 1 execute processing related to the insurance claim, or can execute processing related to the insurance claim himself / herself.

[0141] Fig. 8 is an example of a display screen of the second notification when it is determined that the fraud score is equal to or greater than the first threshold value. The display screen of Fig. 8 displays a fraud score display area d10, a dialogue display area d12, a rejection permission button d14, an additional information input button d16, a message transmission button d18, and a billing information detail display table d20.

[0142] The fraud score display area d10 displays the fraud score calculated by the calculation means 106. The fraud score is an example of information regarding the possibility of fraud being present in the billing information 138. In the display screen of FIG. 8, "fraud score 85" is displayed in the fraud score display area d10.

[0143] The dialog display area d12 displays an overview of the operation required of the person in charge. In the display screen of FIG. 8, the dialog display area d12 displays "This claim has a high fraud score. Do you want to allow the system to reject it?" In this way, the second notification may include information urging the person in charge to execute processing related to the insurance claim, in addition to information regarding the possibility of fraud in the claim information 138.

[0144] The rejection permission button d14 is a button that, when pressed by the person in charge, causes the second execution means 110b to execute a process related to the rejection of the insurance claim. The reception means 113 receives input of information related to the permission of the process related to the insurance claim from the person in charge via the rejection permission button d14.

[0145] The input additional information button d16 is a button for the person in charge to use a function for adding additional information to the claim information 138. The send message button d18 is a button for the person in charge to use a function for sending a message from the person in charge to the insured person. The claim information detail display table d20 is a table in which the contents of the claim information 138 can be confirmed. If the person in charge feels that the fraud score calculated by the information processing device 1 is not valid, he or she can use the input additional information button d16 and the send message button d18 to proceed with the processing related to the insurance claim (without rejecting the insurance claim).

[0146] Fig. 9 is an example of a display screen of the second notification when it is determined that the fraud score is equal to or less than the second threshold. The display screen of Fig. 9 displays a fraud score display area d10, a dialogue display area d12, an approval permission button d22, an additional information input button d16, a message transmission button d18, and a billing information detail display table d20. Of these, the fraud score display area d10, the dialogue display area d12, the additional information input button d16, the message transmission button d18, and the billing information detail display table d20 are the same as those of Fig. 8, so their explanations will be omitted.

[0147] The approval permission button d22 is a button that, when pressed by the person in charge, causes the second execution means 110b to execute a process related to the approval of the insurance claim. The reception means 113 receives input of information related to the permission of the process related to the insurance claim from the person in charge via the approval permission button d22.

[0148] Fig. 10 is an example of a display screen of the first notification when it is determined that the fraud score is lower than the first threshold and higher than the second threshold. The display screen of Fig. 10 displays a fraud score display area d10, a dialogue display area d12, an additional information input button d16, a message send button d18, and a claim information detail display table d20. The person in charge can proceed with the processing related to the insurance claim by using the additional information input button d16 and the message send button d18.

[0149] The dialogue display area d12 in FIG. 10 displays "This is a claim with a slightly high fraud score. Please take necessary measures and proceed with the process." This dialogue is an example of information that prompts the person in charge to execute the process related to the insurance claim. The display screen in FIG. 10 also displays the fraud score display area d10, similar to the display screens in FIGS. 8-9. In this way, the first notification may include information regarding the possibility of fraud in the claim information 138, in addition to the information that prompts the person in charge to execute the process related to the insurance claim.

[0150] <7. Variations> [Variation 1] In the above embodiment, the possibility that fraud exists in the billing information 138 is estimated by comparing the billing information 138 with the statistical information 140b as shown in Fig. 7, but this is not limited to the above. Specifically, when the statistical information 140b includes information on whether or not fraud exists in each of the billing information 138 related to one or more other insurance claims, the estimation means 102 may estimate the possibility that fraud exists in the billing information 138 based on a machine learning model trained on the statistical information 140b. In other words, the machine learning model is a model generated by learning the relationship between the billing information 138 related to one or more other insurance claims acquired in the past and information on whether or not fraud exists in each of the billing information 138.

[0151] According to such a machine learning model, the possibility that fraud exists in the newly acquired billing information 138 can be estimated based on the history of the past billing information 138. For example, a billing information vector V consisting of the values ​​of each item of the new billing information 138 can be calculated as follows: 1 and a billing information vector V consisting of the values ​​of each item of the past billing information 138 2 If the billing information 138 are similar and fraud was present in the past billing information 138, it can be estimated that there is a high possibility that fraud is also present in the new billing information 138.

[0152] [Variation 2] The process related to the insurance claim includes a first process and a second process executed after the first process, the claim information 138 includes the first claim information acquired by the time of executing the first process and the second claim information acquired by the time of executing the second process, and the estimation means 102 may include a first estimation means for estimating the possibility of fraud in the claim information 138 based on the first claim information and the reference information 140, and a second estimation means for estimating the possibility of fraud in the claim information 138 based on the second claim information and the reference information 140. That is, when the claim information 138 changes over time (for example, when a person in charge adds additional information to the claim information 138 and when new information related to the insurance claim is acquired from the insured person), the information processing device 1 may estimate the possibility of fraud in the claim information 138 at each time point based on the claim information 138 at each time point.

[0153] A specific example will be described with reference to Fig. 11 and Fig. 12. Fig. 11 is a diagram for explaining an example of a method for estimating the possibility that fraud exists in the claim information 138 at each of multiple points in time after the information processing device 1 starts processing related to the insurance claim.

[0154] First, the information processing device 1 receives an accident report from the insured person (S200) and starts processing related to the insurance claim (S202). At this point, there is not enough information to estimate the possibility that fraud exists in the billing information 138, and the information processing device 1 does not estimate the possibility that fraud exists in the billing information 138.

[0155] Next, the information processing device 1 acquires the billing information 138 from the insured person terminal device 2 during the period p1 (S204). Then, the information processing device 1 executes the first process (S206). At this time, the information processing device 1 calculates a fraud score using the estimation means 102 and the calculation means 106 based on the billing information 138 acquired during the period p1, and determines whether the fraud score satisfies either the first condition or the second condition using the determination means 104. In this example, the fraud score is calculated to be "50", and does not satisfy either the first condition or the second condition. That is, the information processing device 1 transmits a notification to the person in charge terminal device 3 using the first execution means 110a.

[0156] The person in charge who has received the notification sends a message to the insured person terminal device 2 by operating a display screen such as that shown in Fig. 10. In this example, the person in charge requests the insured person to send additional information (S208).

[0157] Next, the information processing device 1 acquires additional information of the billing information 138 from the insured person terminal device 2 during the period p2 (S210). The additional information includes information that was not included in the billing information 138 acquired by the information processing device 1 in S204. That is, the information processing device 1 can calculate a fraud score with higher accuracy by using the additional information.

[0158] Next, the information processing device 1 executes the second process (S212). At this time, the information processing device 1 calculates a fraud score by the estimation means 102 and the calculation means 106 based on the billing information 138 (including additional information) acquired during the period p1-period p2, and determines whether the fraud score satisfies either the first condition or the second condition by the determination means 104. In this example, the fraud score is calculated to be "35," and does not satisfy either the first condition or the second condition. That is, the information processing device 1 transmits a notification to the person in charge terminal device 3 by the first execution means 110a.

[0159] The person in charge who has received the notification sends a message to the insured person terminal device 2 by operating a display screen such as that shown in Fig. 10. In this example, the person in charge requests the insured person to send further additional information (S214).

[0160] Next, the information processing device 1 acquires additional information of the billing information 138 from the insured person terminal device 2 during period p3 (S216). The additional information includes information that was not included in the billing information 138 and the additional information acquired by the information processing device 1 in S204 and S210. That is, the information processing device 1 can calculate the fraud score with higher accuracy by using the additional information acquired in S216.

[0161] Next, the information processing device 1 executes the third process (S218). At this time, the information processing device 1 calculates a fraud score using the estimation means 102 and the calculation means 106 based on the billing information 138 (including additional information) acquired during the period p1 to p3, and determines whether the fraud score satisfies either the first condition or the second condition using the determination means 104. In this example, the fraud score is calculated to be "10", which satisfies the second condition. In this example, the information processing device 1 executes a process related to the approval of the insurance claim using the second execution means 110b.

[0162] Due to the acquisition of additional information, the billing information 138 changes over time, and the judgment result by the judgment means 104 may also change. Therefore, by adopting the configuration described in this example, the person in charge can focus on work that requires advanced judgment skills. Specifically, in this example, the person in charge had to intervene because the possibility of fraud in the billing information 138 could not be said to be low or high during the period p1-p2, but at the end of the period p3 when sufficient information had been collected, the information processing device 1 automatically approved the insurance claim. In other words, the person in charge does not need to execute the third process or subsequent processes, and can proceed with processing related to other insurance claims.

[0163] <9. Other embodiments>

[0164] The information processing device 1 may be a device consisting of multiple devices. The information processing device 1 may also be a part of another device. The program described in the above embodiment may also cause multiple devices to operate in cooperation with each other.

[0165] The technical matters described in the above embodiments can be rearranged, combined, or partially extracted for use based on the knowledge of those skilled in the art.

[0166] In the above embodiment, the damage is an automobile, and the insurance claim is related to an automobile accident, but this is not limited to this. For example, the damage may be any object such as real estate. In addition, the insurance claim may be related to fire, flood, wind damage, etc.

[0167] The following embodiments are also included in the present invention.

[0168] [Appendix 1] A program according to one aspect of the present invention includes a computer 30, a first acquisition means 100a for acquiring billing information 138 related to a user's insurance claim, a second acquisition means 100b for acquiring reference information 140 related to the insurance claim, an estimation means 102 for estimating the possibility of fraud in the billing information 138 based on the billing information 138 and the reference information 140, a first determination means 104a for determining whether the possibility estimated by the estimation means 102 satisfies a first condition indicating that there is a high possibility of fraud in the billing information 138, a second determination means 104b for determining whether the possibility estimated by the estimation means 102 satisfies a second condition indicating that there is a low possibility of fraud in the billing information 138, and the first determination means 104b. a first execution means 110a that, when it is determined by the first determination means 104a and the second determination means 104b that the possibility estimated by the estimation means 102 does not fall under either the first condition or the second condition, executes a process of sending a first notification to the terminal device 3 of the person in charge of the insurance claim to prompt the person in charge to execute a process related to the insurance claim; and a second execution means 110b that, when it is determined by the first determination means 104a and the second determination means 104b that the possibility estimated by the estimation means 102 falls under either the first condition or the second condition, executes at least one of a process related to the insurance claim and a process of sending a second notification to the terminal device 3 of the person in charge, the second notification including information regarding the possibility of fraud in the billing information 138.

[0169] [Appendix 2] In the program of Appendix 1, the second execution means 110b executes processing related to an insurance claim when the possibility estimated by the estimation means 102 satisfies a second condition, and the processing related to the insurance claim executed by the second execution means 110b may include processing related to approval of the insurance claim.

[0170] [Appendix 3] In the program of Appendix 1 or Appendix 2, the second execution means 110b executes processing related to an insurance claim when the possibility estimated by the estimation means 102 satisfies the first condition, and the processing related to the insurance claim executed by the second execution means 110b may include processing related to rejection of the insurance claim.

[0171] [Appendix 4] Any one of the programs in Supplementary Note 1 to Supplementary Note 3 may further cause the computer 30 to function as a reception means 113 that receives input from a person in charge regarding processing related to an insurance claim, and the second execution means 110b may execute processing related to an insurance claim without the reception means 113 receiving input.

[0172] [Appendix 5] The program of any one of Supplementary Note 1 to Supplementary Note 4 may further cause the computer 30 to function as a receiving means 113 that receives, from a person in charge, an input regarding permission for processing related to the insurance claim to be executed by the second execution means 110b.

[0173] [Appendix 6] In the program of any one of Supplements 1 to 5, the reference information 140 may include statistical information 140b related to one or more other claim statistics.

[0174] [Appendix 7] In the program of Appendix 6, the statistical information 140b includes information regarding whether or not fraud has existed in each of the billing information 138 relating to one or more other insurance claims, and the estimation means 102 may estimate the possibility of fraud existing in the billing information 138 based on a machine learning model trained on the statistical information 140b.

[0175] [Appendix 8] In the program of Appendix 6, the claims information 138 includes information about businesses recovering damage caused to the insured's property, the statistical information 140b includes information about the business's performance, and the estimation means 102 may further estimate the possibility of fraud in the claims information 138 based on the information about the performance.

[0176] [Appendix 9] In the program of Appendix 6, the insurance claim includes a first insurance claim and a second insurance claim made after the first insurance claim, the statistical information 140b includes information regarding whether or not there was fraud in the claim information 138 related to the first insurance claim, and the estimation means 102 may estimate the possibility of fraud in the claim information 138 related to the second insurance claim further based on the relationship between the first insurance claim and the second insurance claim.

[0177] [Appendix 10] In the program of any one of Supplements 1 to 9, the reference information 140 may include on-site information 140a relating to the actual circumstances of the occurrence of the damage related to the insurance claim.

[0178] [Appendix 11] The program of any one of Supplementary Notes 1 to 10 may further cause the computer 30 to function as a calculation means 106 for calculating a value corresponding to the likelihood estimated by the estimation means 102, and the first condition may be that the value is greater than or equal to a first threshold, and the second condition may be that the value is less than or equal to a second threshold that is lower than the first threshold.

[0179] [Appendix 12] Any one of the programs of Appendices 1 to 11 further causes the computer 30 to function as a management means 112 that associates and manages the status of an insurance claim with the billing information 138, the status including an incomplete status indicating that the processing related to the insurance claim is incomplete and a completed status indicating that the processing related to the insurance claim has been completed, and the management means 112 may change the status to a completed status when the second execution means 110b has performed the processing related to the insurance claim.

[0180] [Appendix 13] In the program of any one of Appendixes 1 to 12, the billing information 138 may be generated by performing optical character recognition on a document written or created by a user.

[0181] [Appendix 14] In the program of any one of Supplementary Notes 1 to 13, the processing related to insurance claims includes a first processing and a second processing executed after the first processing, the billing information 138 includes first billing information 138 acquired by the time the first processing is executed and second billing information 138 acquired by the time the second processing is executed, and the estimation means 102 may include a first estimation means 102 that estimates the possibility of fraud in the billing information 138 based on the first billing information 138 and the reference information 140, and a second estimation means 102 that estimates the possibility of fraud in the billing information 138 based on the second billing information 138 and the reference information 140.

[0182] [Appendix 15] In the program of any one of Supplementary Notes 1 to 14, the persons in charge include a first person in charge and a second person in charge having a role different from the first person in charge, and the program further causes the computer 30 to function as a selection means 116 that selects the first person in charge and the second person in charge according to the possibility estimated by the estimation means 102, and the first execution means 110a may execute a process of sending a first notification to the terminal device 3 of at least one of the first person in charge and the second person in charge based on the selection by the selection means 116.

[0183] [Appendix 16] An information processing device 1 according to another aspect of the present invention includes a first acquisition unit that acquires billing information 138 related to a user's insurance claim, a second acquisition unit that acquires reference information 140 related to the insurance claim, an estimation unit that estimates the possibility of fraud in the billing information 138 based on the billing information 138 and the reference information 140, a first determination unit that determines whether the possibility estimated by the estimation unit satisfies a first condition indicating that there is a high possibility of fraud in the billing information 138, a second determination unit that determines whether the possibility estimated by the estimation unit satisfies a second condition indicating that there is a low possibility of fraud in the billing information 138, and a third determination unit that determines whether the possibility estimated by the estimation unit satisfies a second condition indicating that there is a low possibility of fraud in the billing information 138. The device is equipped with a first execution unit that executes a process of sending a first notification to a person in charge of the insurance claim to prompt the person in charge of the insurance claim to execute a process related to the insurance claim when the first determination unit and the second determination unit determine that the possibility estimated by the estimation unit does not meet either the first condition or the second condition, and a second execution unit that executes at least one of a process related to the insurance claim and a process of sending a second notification to the person in charge's terminal device 3, the process including information regarding the possibility of fraud in the claim information 138, when the first determination unit and the second determination unit determine that the possibility estimated by the estimation unit meets either the first condition or the second condition.

[0184] [Appendix 17] An information processing method according to another aspect of the present invention includes, in a computer 30, a first acquisition step of acquiring billing information 138 related to a user's insurance claim, a second acquisition step of acquiring reference information 140 related to the insurance claim, an estimation step of estimating a possibility that fraud exists in the billing information 138 based on the billing information 138 and the reference information 140, a first determination step of determining whether the possibility estimated by the estimation step satisfies a first condition indicating that there is a high possibility that fraud exists in the billing information 138, and a second determination step of determining whether the possibility estimated by the estimation step satisfies a second condition indicating that there is a low possibility that fraud exists in the billing information 138. a first execution step of executing a process of sending a first notification to the person in charge of the insurance claim to prompt the person in charge of the insurance claim to execute a process related to the insurance claim when it is determined by the first judgment step and the second judgment step that the possibility estimated by the estimation step does not satisfy either the first condition or the second condition; and a second execution step of executing at least one of a process related to the insurance claim and a process of sending a second notification including information regarding the possibility of fraud in the claim information 138 to the terminal device 3 of the person in charge when it is determined by the first judgment step and the second judgment step that the possibility estimated by the estimation step satisfies either the first condition or the second condition. [Explanation of symbols]

[0185] 1...information processing device, 30...computer, 100...acquisition means, 100a...first acquisition means, 100b...second acquisition means, 102...estimation means, 104...determination means, 104a...first determination means, 104b...second determination means, 106...calculation means, 110...execution means, 110a...first execution means, 110b...second execution means, 112...management means, 113...reception means, 114...display means, 116...selection means, 138...billing information, 140...reference information, 140a...site information, 140b...statistical information

Claims

1. Computer, A first acquisition means for acquiring billing information related to a user's insurance claim; A second acquisition means for acquiring reference information related to the insurance claim; an estimation means for estimating the possibility that fraud exists in the billing information based on the billing information and the reference information; a first determination means for determining whether or not the possibility estimated by the estimation means satisfies a first condition indicating that there is a high possibility that fraud exists in the billing information; a second determination means for determining whether or not the possibility estimated by the estimation means satisfies a second condition indicating that the billing information is unlikely to include fraud; a first execution means for executing a process of sending a first notification to a person in charge of the insurance claim to prompt the person in charge of the insurance claim to execute a process related to the insurance claim when the first determination means and the second determination means determine that the possibility estimated by the estimation means does not satisfy either the first condition or the second condition; a second execution means for executing at least one of a process related to the insurance claim and a process of sending a second notification including information regarding the possibility of fraud in the claim information to a terminal device of the person in charge when the first determination means and the second determination means determine that the possibility estimated by the estimation means satisfies either the first condition or the second condition; A program that functions as a

2. 2. The program according to claim 1, wherein the second execution means executes processing related to the insurance claim when the possibility estimated by the estimation means satisfies the second condition, and the processing related to the insurance claim executed by the second execution means includes processing related to approval of the insurance claim.

3. 2. The program according to claim 1, wherein the second execution means executes processing related to the insurance claim when the possibility estimated by the estimation means satisfies the first condition, and the processing related to the insurance claim executed by the second execution means includes processing related to rejection of the insurance claim.

4. The computer, and further functioning as a reception unit for receiving input related to the processing of the insurance claim from the person in charge; The program according to claim 1 , wherein the second execution means executes the process related to the insurance claim without the reception means receiving an input.

5. The computer, 2 . The program according to claim 1 , further functioning as a reception unit that receives, from the person in charge, an input regarding permission for the process related to the insurance claim to be executed by the second execution unit.

6. The program of claim 1 , wherein the reference information comprises statistical information regarding one or more other claims statistics.

7. the statistical information includes information regarding whether or not fraud was present in each of the claim information relating to the one or more other insurance claims; The program according to claim 6 , wherein the estimation means estimates the possibility that fraud exists in the billing information based on a machine learning model that has been trained on the statistical information.

8. The claim information includes information regarding an entity recovering damage caused to the insured's property; The statistical information includes information regarding the business's performance, 7. The program according to claim 6, wherein the estimation means estimates the possibility that fraud exists in the billing information further based on information related to the performance.

9. The insurance claim includes a first insurance claim and a second insurance claim made after the first insurance claim, The statistical information includes information regarding whether or not fraud was present in the claim information related to the first insurance claim, The program according to claim 6 , wherein the estimation means estimates a possibility that fraud exists in the claim information related to the second insurance claim, further based on a relationship between the first insurance claim and the second insurance claim.

10. The program according to claim 1 , wherein the reference information includes on-site information regarding an actual situation of occurrence of the damage related to the insurance claim.

11. The computer, a calculation means for calculating a value corresponding to the likelihood estimated by the estimation means; the first condition is that the value is equal to or greater than a first threshold; The program according to claim 1 , wherein the second condition is that the value is equal to or less than a second threshold value that is lower than the first threshold value.

12. The computer, a management means for managing the insurance claim status in association with the claim information, the status including an incomplete status indicating that the processing related to the insurance claim is incomplete and a completed status indicating that the processing related to the insurance claim is completed; 10. The program according to claim 1, wherein the management means changes the status to the completion status when the second execution means has performed the process related to the insurance claim.

13. The program of claim 1 , wherein the billing information is generated by performing optical character recognition on a document written or created by the user.

14. The process related to the insurance claim includes a first process and a second process executed after the first process, the billing information includes first billing information acquired by the time the first process is executed and second billing information acquired by the time the second process is executed; The program according to any one of claims 1 to 9, wherein the estimation means includes a first estimation means for estimating the possibility of fraud in the billing information based on the first billing information and the reference information, and a second estimation means for estimating the possibility of fraud in the billing information based on the second billing information and the reference information.

15. The person in charge includes a first person in charge and a second person in charge having a role different from that of the first person in charge, The computer, a selection unit that selects the first person in charge and the second person in charge according to the possibility estimated by the estimation unit; 10. The program according to claim 1, wherein the first execution means executes a process of sending the first notification to a terminal device of at least one of the first person in charge and the second person in charge based on the selection by the selection means.

16. A first acquisition unit that acquires billing information related to an insurance claim of a user; A second acquisition unit that acquires reference information related to the insurance claim; an estimation unit that estimates a possibility that fraud exists in the billing information based on the billing information and the reference information; a first determination unit that determines whether or not the possibility estimated by the estimation unit satisfies a first condition indicating that there is a high possibility that fraud exists in the billing information; a second determination unit that determines whether or not the possibility estimated by the estimation unit satisfies a second condition indicating that the billing information is unlikely to include fraud; a first execution unit that executes a process of sending a first notification to a person in charge of the insurance claim to prompt the person in charge of the insurance claim to execute a process related to the insurance claim when the first determination unit and the second determination unit determine that the possibility estimated by the estimation unit does not satisfy either the first condition or the second condition; a second execution unit that executes at least one of a process related to the insurance claim and a process of sending a second notification including information regarding the possibility of fraud in the claim information to a terminal device of the person in charge when the first determination unit and the second determination unit determine that the possibility estimated by the estimation unit satisfies either the first condition or the second condition; An information processing device comprising:

17. On the computer, A first acquisition step of acquiring claim information related to a user's insurance claim; A second obtaining step of obtaining reference information related to the insurance claim; an estimation step of estimating a possibility that fraud exists in the billing information based on the billing information and the reference information; a first determination step of determining whether or not the possibility estimated by the estimation step satisfies a first condition indicating that there is a high possibility that fraud exists in the billing information; a second determination step of determining whether or not the possibility estimated by the estimation step satisfies a second condition indicating that the billing information is unlikely to include fraud; a first execution step of executing a process of sending a first notification to a person in charge of the insurance claim to prompt the person in charge of the insurance claim to execute a process related to the insurance claim when it is determined by the first determination step and the second determination step that the possibility estimated by the estimation step does not satisfy either the first condition or the second condition; a second execution step of executing at least one of a process related to the insurance claim and a process of sending a second notification including information regarding the possibility of fraud in the claim information to a terminal device of the person in charge when it is determined by the first determination step and the second determination step that the possibility estimated by the estimation step satisfies either the first condition or the second condition; An information processing method.

Images