JP7144495B2 - Payment exclusion possibility determination device, payment exclusion possibility determination system, and payment exclusion possibility determination method - Google Patents

Description

The present invention relates to a payment exclusion possibility determination device, a payment exclusion possibility determination system, and a payment exclusion possibility determination method.

There are cases where a customer makes an insurance claim by mistake (erroneous claim) or a customer makes an insurance claim fraudulently (fraudulent claim). There is a claim that Therefore, an insurance company has an assessment department that assesses whether or not a claim is eligible for insurance payment.

Japanese Patent Application Laid-Open No. 2003-30441

For claims that are not covered by insurance claims, especially fraudulent claims, it is necessary to check multiple factors in a complex manner, and it is difficult to conduct uniform manual checks. Therefore, the decision as to whether or not a claim is eligible for insurance payment depends largely on the experience of the person in charge, and it is difficult to improve the efficiency of the assessment process.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a payment exclusion possibility determination device, a payment exclusion possibility determination system, and a payment exclusion possibility determination method that can streamline the assessment work for claims for payment of insurance claims.

A device for determining the possibility of non-payment according to the present invention includes: a first acquisition unit that acquires insurance claim information from a customer; for each combination of a plurality of predetermined items and the values registered in each of the plurality of predetermined items, with respect to the total number of records corresponding to the combination in the past insurance claims Statistical data that stores the ratio of the number of records for which insurance claims have not been paid out of the number of records corresponding to the combination is set for the items included in the claim information for the claims for insurance claims. values, values set for items included in the information described in the document, and values set for items included in the customer's insurance contract information, the statistical data is A value for each of the plurality of predetermined items provided is obtained, a weighting factor set for each of the plurality of predetermined items, the plurality of predetermined items and the obtained predetermined items A value representing the possibility that the insurance claim will not be covered by the insurance payment, using the product of the ratio of the statistical data corresponding to the combination of the values for each of the plurality of prescribed items, and and a determination unit that calculates

The above configuration may further include a determination unit that determines, from among a plurality of departments, a department that handles the insurance claim based on the value representing the possibility.

In the above configuration, the second acquisition unit may acquire the information written on the document by converting the image of the document into text using AI-OCR.

In the above configuration, the plurality of predetermined items may be items included in any one of the billing information, the contract information, and the information written on the document.

In the above configuration, the plurality of predetermined items may be items using a plurality of items included in the past insurance claim information, contract information, and information described in documents.

In the above configuration, the non-payment possibility determination device may include a statistical data generation unit that generates the statistical data.

A system for determining the possibility of non-payment according to the present invention includes: a reception unit that receives input of insurance claim information from a customer; a first acquisition unit that acquires an image of a document related to the insurance claim; a first device having a transmission section for transmitting information and an image of the document; a second acquisition section for acquiring information described in the document from the image of the document; and a plurality of predetermined items. and the values registered in each of the plurality of predetermined items, out of the number of records corresponding to the combination of the total number of records corresponding to the combination in the past insurance claims With reference to statistical data that stores the percentage of the number of records that have not been paid, for the insurance claim, the value set for the item included in the claim information, included in the information described in the document For each of the plurality of predetermined items included in the statistical data, using the values set for the items and the values set for the items included in the customer's insurance contract information a weighting factor set for each of the plurality of predetermined items, and a value for each of the plurality of predetermined items and the obtained values for each of the plurality of predetermined items; a determination unit that calculates a value representing the possibility that the insurance claim will not be covered by the insurance payment, using the respective products of the ratio of the statistical data corresponding to the combination and a device;

A method for determining the possibility of non-payment according to the present invention includes a step of acquiring information on a claim for insurance money from a customer, a step of acquiring information described in a document related to the insurance claim, and number of records corresponding to said combination against all records corresponding to said combination in past insurance claims Statistical data that stores the percentage of records for which insurance claims have not been paid is referenced, and with regard to the claims for insurance claims, the values set for the items included in the claim information and the values described in the documents Using the values set for the items included in the information and the values set for the items included in the customer's insurance contract information, the plurality of predetermined values included in the statistical data obtaining a value for each of the items, weighting coefficients set for each of the plurality of predetermined items, and each of the plurality of predetermined items and the obtained plurality of predetermined items; calculating a value representing the probability that the insurance claim will not be covered by an insurance payment, using the product of each of the percentage of the statistical data corresponding to the combination with the value of executes.

Advantageous Effects of Invention According to the present invention, it is possible to improve the efficiency of assessment work for claims for payment of insurance claims.

FIG. 1 is a block diagram of a non-payment possibility determination system according to one embodiment. FIG. 2A is a functional block diagram of the customer terminal, and FIG. 2B is a diagram showing the hardware configuration of the customer terminal. FIG. 3A is a functional block diagram of the determination server, and FIG. 3B is a diagram showing the hardware configuration of the determination server. FIG. 4 is a diagram showing an example of an insurance payment management table. FIG. 5 is a sequence diagram showing the process until the determination server acquires the billing information and the document image. 6A shows an example of a login screen, FIG. 6B shows an example of a top page, and FIG. 6C shows an example of an insurance claim information input screen. FIG. 4 is a diagram showing; FIG. 7A is a diagram showing an example of a document photographing screen, and FIG. 7B is a diagram showing an example of a reception completion screen. FIG. 8A is a diagram showing an example of a determination DB, and FIG. 8B is a diagram showing an example of a weighting factor DB. FIG. 9 is a flowchart illustrating an example of processing executed by a determination server; FIG. 10 is a diagram showing an example of document description information. FIG. 11 is a diagram showing examples of determination items.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[Embodiment]
FIG. 1 is a block diagram of a non-payment possibility determination system 100 according to one embodiment. The non-payment possibility determination system 100 is a system that determines the possibility that, when a customer makes an insurance claim, the claim is subject to insurance payment.

As shown in FIG. 1, the payment exclusion possibility determination system 100 includes a customer terminal 60, a contract information management server 20, a determination server 30, an insurance claim payment management server 40, and the like.

The contract information management server 20, determination server 30, and insurance payment management server 40 are connected via a network NW2 such as an in-house LAN.

Also, the determination server 30 is connected to the customer terminal 60 via the network NW1. The network NW1 is, for example, a telecommunications line such as the Internet.

(Customer terminal 60)
The customer terminal 60 is a smart phone, a tablet terminal, a personal computer, or the like owned by a customer of an insurance company (or an insurance agency). When a customer makes a claim for payment of an insurance claim, the customer inputs information necessary for claiming payment of the insurance claim to the customer terminal 60 and transmits the information to the determination server 30 .

FIG. 2A is a functional block diagram of the customer terminal 60. As shown in FIG. As shown in FIG. 2A, the customer terminal 60 functions as a display control section 61, a communication section 62, an image acquisition section 63, and the like.

FIG. 2B is a diagram showing the hardware configuration of the customer terminal 60. As shown in FIG. As illustrated in FIG. 2B, the customer terminal 60 includes a CPU (Central Processing Unit) 601, a RAM (Random Access Memory) 602, a storage device 603, a communication device 604, an input device 605, a display device 606, and an imaging device. 607 and the like. Each of these devices is connected by a bus or the like. A CPU 601 is a central processing unit. A RAM 602 is a volatile memory that temporarily stores programs executed by the CPU 601 and data processed by the CPU 601 . Storage device 603 is a non-volatile storage device. As the storage device 603, for example, a ROM (Read Only Memory), a solid state drive (SSD) such as a flash memory, a hard disk driven by a hard disk drive, or the like can be used. The functions of each part of the customer terminal 60 are implemented by the CPU 601 executing the programs stored in the storage device 603 . The function of each part of the customer terminal 60 may be configured by a dedicated circuit or the like. A communication device 604 is an interface to the network NW1. The input device 605 is a device for inputting information, such as a keyboard, mouse, and touch panel. The display device 606 is a device for displaying information, such as a liquid crystal display. The image pickup device 607 is a device for photographing an object to be photographed, and is, for example, a camera. Note that the customer terminal 60 does not have to be equipped with the imaging device 607 .

(Contract information management server 20)
The contract information management server 20 is, for example, a computer such as a server, and is installed in an insurance company or in a cloud environment contracted by the insurance company. The contract information management server 20 stores insurance contract information of each customer. Contract information includes, for example, insurance policy number, contract type, indemnification rider, insurance period, contract date, past claim history, first year contract date, relationship between insured person and policyholder, gender of insured person, insured person Includes information such as the age of the insurer, the agency that handled the policy, the address of the policyholder, and the address of the insured.

(Determination server 30)
The determination server 30 is, for example, a computer such as a server, and is installed in an insurance company or a cloud environment contracted by the insurance company. When the determination server 30 receives an insurance payment request from the customer terminal 60, the determination server 30 determines the possibility that the request will not be covered by the insurance payment.

FIG. 3A is a functional block diagram of the determination server 30. As shown in FIG. As shown in FIG. 3A, the determination server 30 includes a billing information receiving unit 31, an image information processing unit 32, a payment non-payment possibility determination unit 35, a corresponding department determination unit 36, a determination DB storage unit 37, and a determination It functions as a DB creating unit 38 and the like.

FIG. 3B is a diagram showing the hardware configuration of the determination server 30. As shown in FIG. As shown in FIG. 3B, the determination server 30 includes a CPU 301, a RAM 302, a storage device 303, a communication device 304, an input device 305, and the like. Each of these devices is connected by a bus or the like. A CPU 301 is a central processing unit. A RAM 302 is a volatile memory that temporarily stores programs executed by the CPU 301 and data processed by the CPU 301 . Storage device 303 is a non-volatile storage device. As the storage device 303, for example, a ROM, a solid state drive (SSD) such as a flash memory, a hard disk driven by a hard disk drive, or the like can be used. The CPU 301 executes the programs stored in the storage device 303 to implement the functions of each section of the determination server 30 . The function of each part of the determination server 30 may be configured by a dedicated circuit or the like. Communication device 304 is an interface to network NW1 and network NW2. The input device 305 is a device for inputting information, such as a keyboard and a mouse.

(Insurance payment management server 40)
The insurance claim payment management server 40 is, for example, a computer such as a server, and is installed within an insurance company or in a cloud environment contracted by the insurance company. The insurance payment management server 40 stores an insurance payment management table for managing insurance payment claim information from customers. For example, as shown in FIG. 4, the insurance claim payment management table includes claim information entered by the customer from the customer terminal 60, information described in documents submitted at the time of claiming insurance claims (document description information), and contract information. including. In addition, in the insurance claim payment management table, the results of assessment by a specialized assessment department as to whether or not each insurance claim is subject to insurance payment, and whether or not the insurance claim is to be paid are stored. Actual information such as the insurance amount actually paid in the case is also stored.

≪Billing Information Acquisition Process≫
First, in the payment exclusion possibility determination system 100 according to the present embodiment, the process until the determination server 30 acquires the claim information for insurance payment and the image of the document necessary for claiming the insurance is described in the sequence of FIG. Description will be made with reference to the drawings.

In the process of FIG. 5, a customer who makes an insurance claim accesses the web page for customers of the insurance company using the customer terminal 60 (step S11). The display control unit 61 of the customer terminal 60 displays the login screen of the web page for the customer on the display device 606 (step S13).

FIG. 6A is a diagram showing an example of a login screen. As shown in FIG. 6A, the login screen has login ID and password (PW) as input items.

The customer enters a login ID and a password on the login screen shown in FIG. 6A (step S15). When the customer presses the send button B1 on the login screen shown in FIG. 6A, the communication unit 62 of the customer terminal 60 transmits the entered login ID and password to the determination server 30 (step S17).

Upon receiving the login ID and password, the determination server 30 executes authentication processing to determine whether or not the combination of the login ID and password is correct (step S19).

If the combination of the received login ID and password is correct, the determination server 30 transmits the top page for the customer to the customer terminal 60 (step S21). The display control unit 61 of the customer terminal 60 displays the top page on the display device 606 (step S23). FIG. 6B shows an example of the top page. As shown in FIG. 6B, the top page allows selection of processing that can be executed on the web page for the customer. For example, on the top page shown in FIG. 6B, confirmation or correction of policyholder information, confirmation of compensation details, and insurance claims can be selected.

When making an insurance claim, the customer selects an insurance claim by, for example, tapping with a finger or clicking with a mouse on "Insurance Claim" on the top page shown in FIG. 6B (step S25).

The display control unit 61 of the customer terminal 60 displays, for example, a billing information input screen as shown in FIG. 6C on the display device 606 (step S27). The customer inputs the information necessary for claiming the insurance money on the claim information input screen (step S29). For example, as shown in FIG. 6(C), the customer responds to questions about hospitalization status, date of onset of illness, symptoms, etc. displayed on the billing information input screen, thereby providing information necessary for claiming insurance (claim information).

In addition, the customer photographs a document necessary for claiming insurance money, such as a receipt issued by a medical institution (step S31). For example, when the "Next" button B2 is pressed on the billing information input screen shown in FIG. 6(C), a photographed document screen is displayed on the display device 606 of the customer terminal 60 as shown in FIG. 7(A). . When the customer presses the button B3 on the screen shown in FIG. 7A, the image acquisition unit 63 takes a picture of the document with the imaging device 607 and acquires the image of the document. If the customer terminal 60 does not have the imaging device 607, the image acquisition unit 63 captures the image of the document with a camera separate from the customer terminal 60 via the USB cable connected to the camera. It may be obtained by the above method and transmitted to the determination server 30 .

When the customer operates the customer terminal 60 to transmit the billing information and the document image, the communication section 62 transmits the billing information and the document image to the determination server 30 (step S33). Upon receiving the billing information and the document image, the determination server 30 transmits a reception completion screen to the customer terminal 60 (step S35). The display control unit 61 of the customer terminal 60 causes the display device 606 to display a reception completion screen as shown in FIG. 7B (step S37).

Through the above processing, the determination server 30 can acquire insurance claim information and document images.

《Process of judging the possibility of non-payment》
Based on the billing information and the document image obtained as described above, the customer's insurance contract information, the judgment DB (Data Base) and the weighting coefficient DB stored in the judgment DB storage unit 37, the judgment server 30 determines the possibility that the claim for payment received from the customer terminal 60 is not subject to insurance payment (possibility of non-payment). The determination DB and the weighting factor DB are databases created for the purpose of quantifying the know-how of persons in charge of assessment work.

Before describing the processing executed by the determination server 30, the determination DB and the weighting coefficient DB used to determine the possibility of payment exclusion will be described.

FIG. 8A is a diagram showing an example of a determination DB. The determination DB is statistical data created by the determination DB creation unit 38 based on the insurance payment record (see FIG. 4) for claims for payment of insurance claims. As shown in FIG. 8A, the determination DB includes fields for determination item, item value, number of non-target records, number of all records, and non-target rate.

An item name used to determine the possibility of non-payment is registered in the determination item field. The item name registered in the judgment item field is the item name of the item included in the insurance claim information, the insurance contract information, and the information described in the document submitted at the time of insurance claim (for example, FIG. 8A). ), or a definition of a value calculated from the item values of multiple items (for example, the item "first year contract date" in the contract information It may be "elapsed period from first year contract date to illness onset date" which can be calculated from the value and the item value of "disease onset date" of the billing information. Judgment items may be items that are frequently used or emphasized by an assessment department that assesses whether or not a claim from a customer is subject to insurance payment.

A value for a judgment item is registered in the item value field.

The total number of records field registers the number of records registered in the insurance claim payment management table in which item values are registered in the items corresponding to the judgment items. For example, when the payment exclusion possibility determination system 100 is introduced, the determination DB creation unit 38, for all records registered in the insurance claim payment management table, for each combination of determination items and item values, Count the number of records and register it in the "Number of all records" item. After the introduction of the non-payment possibility determination system 100, the determination DB creating unit 38 performs the same processing on newly registered records at predetermined intervals (for example, every day or every week). , to update the data in the judgment DB. As a result, it becomes possible to determine the possibility of non-payment based on the determination DB that reflects the new information.

In the field of the number of non-covered records, the number of records for which the insurance payment was not paid out of the records in which the item value is registered in the item corresponding to the judgment item in the insurance payment management table is registered. For example, when the payment exclusion possibility determination system 100 is introduced, the determination DB creation unit 38, for all records registered in the insurance claim payment management table, for each combination of determination items and item values, Count the number of records for which insurance claims have not been paid, and register them in the "Number of non-covered records" item. After the introduction of the non-payment possibility determination system 100, the determination DB creating unit 38 performs the same processing on newly registered records at predetermined intervals (for example, every day or every week). , to update the data in the judgment DB. This makes it possible to determine the possibility of exclusion based on the determination DB that reflects new information. As described above, the insurance claim payment management table reflects the result of assessment by the person in charge of the assessment department that specializes in assessment as to whether or not each claim is subject to insurance payment. Therefore, by using the data of the insurance claim payment management table, it is possible to obtain highly reliable statistical data for determining the possibility of not being covered by insurance claims.

A ratio of the number of non-target records to the total number of records (a value obtained by dividing the number of non-target records by the total number of records) is registered in the field of non-target rate.

FIG. 8B is a diagram showing an example of the weighting factor DB. The weighting factor DB has fields for determination items and weighting factors. A determination item of the determination DB is registered in the determination item field. In the weighting factor field, a weighting factor for each determination item is registered, as shown in FIG. 8B. The value of the weighting factor is set according to the degree to which the determination item affects the determination of whether or not the claim is out of the scope of payment (determination of liability). For example, the name of the disease and the clinical department are items that directly affect the determination of responsibility, so a large weighting factor is set. On the other hand, the agency does not directly affect the judgment of liability, but is an auxiliary item for the judgment, so a relatively small weighting factor is set.

Using the determination DB and the weighting factor DB, the non-payment possibility determination unit 35 determines the possibility that the claim for payment of insurance money will be excluded from the payment of insurance money.

FIG. 9 is a flowchart showing an example of processing executed by the determination server 30. As shown in FIG. The process of FIG. 9 is started, for example, when the billing information receiving section 31 receives the billing information and the document image.

As shown in FIG. 9, when the request information receiving section 31 receives the request information and the document image, the image information processing section 32 converts the received document image into text by AI-OCR (Artificial Intelligence-Optical Character Recognition). (Step S53). The image information processing unit 32 extracts the contents described for the predetermined items from the text information, and as shown in FIG. Output. The information (items) read from the document image by AI-OCR is adjusted according to the content of insurance coverage.

Next, the non-payment possibility determination unit 35 acquires information used to determine the possibility of non-payment from the billing information, document description information, and contract information (step S55). Specifically, the non-payment possibility determination unit 35 acquires information registered for items corresponding to determination items in the determination DB from billing information, document description information, and contract information. In addition, the non-payment possibility determination unit 35 acquires information necessary for calculating the values defined in the determination items of the determination DB from the billing information, document description information, and contract information.

For example, when "disease name" is registered in the determination item of the determination DB, the non-payment possibility determination unit 35 acquires information registered for "disease name" from the billing information. In addition, for example, when "examination department" is registered in the determination item of the judgment DB, the possibility determination unit 35 determines information registered for "examination department" from the document description information (for example, neurosurgery). In addition, for example, when "agency" is registered in the judgment item of the judgment DB, the non-payment possibility judgment unit 35 acquires the information of the agency that used the insurance contract from the contract information. .

Next, the non-payment possibility determination unit 35 determines the possibility that the claim is not covered by insurance (step S61). Specifically, the probability that the claim will not be covered by insurance claims (non-covered probability) is calculated.

Here, a method of calculating the non-payable probability will be described. First, the non-payment possibility determination unit 35 acquires the non-payment rate set for the same item value as the information extracted in step S61 from the determination DB (see FIG. 8A). For example, "brain surgery" is registered for "examination department" in document information, "cerebral infarction" is registered for "disease name" in billing information, and "agency" is registered in contract information Assume that "Agency C" is registered in the In this case, the non-payment possibility determination unit 22 acquires the respective non-target rates of "10%", "10%" and "33%" from the determination DB.

The non-payment possibility determination unit 35 multiplies the obtained non-target ratio by the weighting factor set for each determination item in the weighting factor DB (see FIG. 8B), and obtains the sum of the sums as the sum of the weighting factors. The value (weighted average value) divided by 2 is calculated as the non-payment probability.

For example, as shown in FIG. 8B, it is assumed that the weighting factor for "examination department" is 5, the weighting factor for "disease name" is 5, and the weighting factor for "agency" is 1. In this case, the non-payment probability is
(10% x 5 + 10% x 5 + 33% x 1)/11 = 12%
becomes.

Next, the corresponding department determination unit 36 determines a department that will respond to the customer's claim for payment (assess the claim) based on the non-payment probability (step S63). Specifically, for example, when the probability of non-payment is low, that is, for a claim that is likely to be covered by insurance, the corresponding department is determined to be a department that simply assesses the claim. On the other hand, if the overall non-coverage rate is high, there is a high possibility that the policy will not be covered by insurance, so a department with a high level of expertise will be selected as the department to deal with this. As a result, claims that are likely to be covered by insurance will be processed by the department that assesses claims in a simplified manner, shortening the time it takes for insurance claims to be paid to customers. It can lead to improvement of satisfaction. On the other hand, for claims that are likely to be excluded from insurance payments, having highly specialized departments assess claims can prevent claims from being paid for fraudulent or erroneous claims. .

As described in detail above, according to the present embodiment, the determination server 30 includes the claim information receiving unit 31 that acquires insurance claim information from a customer, and Based on the image information processing unit 32 that acquires the information obtained from the claim, the claim information, the information described in the document, the customer's insurance contract information, and the past insurance claim and the insurance payment record for the past insurance claim and a non-payment possibility determination unit 35 that determines the possibility that the insurance claim is not subject to insurance payment, based on the determination DB created as described above. As a result, it is considered that the determination DB created based on past insurance claims and payment results reflects the know-how of the person in charge at the time of assessment. Therefore, by using the determination DB, it is possible to accurately determine the possibility of non-payment, and to improve the efficiency of the insurance payment assessment work.

Further, in the present embodiment, the determination server 30 determines the department that handles the claim based on the possibility that the insurance claim will not be covered by the insurance claim (probability of non-payment). As a result, for claims that are likely to be subject to insurance payment, for example, the department that assesses the claim can handle it in a simplified manner, shortening the time it takes for the insurance payment to be made to the customer. can lead to improved customer satisfaction. On the other hand, it is possible to prevent insurance claims from being paid for fraudulent or erroneous claims by having highly specialized departments deal with claims that are likely to be excluded from insurance claims.

Further, in the present embodiment, the image information processing unit 32 obtains information written on the document by converting the image of the document into text by AI-OCR. This eliminates the need for the customer or the operator of the insurance company to register the information written in the document, thereby preventing the insurance claim information from being erroneous due to an input error or the like.

In addition, in this embodiment, the non-payment possibility determination unit 35 uses the claim information, the information described in the document, and the customer's insurance contract information to set the insurance claim from the customer in the determination DB. Obtain the value for the judgment item that is set, obtain the non-target rate corresponding to the value obtained from the judgment DB, and use the weighting factor set for each of the judgment items and the obtained non-target rate to determine the likelihood that a claim will be ineligible for payment. As a result, it is possible to change the weight of items that directly affect the judgment of liability and items that do not directly affect the judgment, , items that are problematic in fraudulent claims) can be reflected in the weighting factor, so the possibility of being excluded from payment of insurance claims can be determined more accurately.

In the above-described embodiment, "examination department", "disease name", and "agency" are used as the determination items of the determination DB, but the items are not limited to these. For example, the items shown in FIG. 11 can be used as determination items. Here, No. in FIG. 1 contract type to No. The item values for the 9 insured person address items can be extracted from the contract information. No. The item value for the item 10 "Is the date of hospitalization within the insurance period of the contract information?" When included, "within period" can be registered in the item value field, and "out of period" can be registered when not included.

Also, No. 11 "number of days elapsed from the first year contract date to the date of admission" and No. The item value for 12 "Number of days elapsed from the first year contract date to the onset of illness" is the date of admission and the date of onset of illness included in the billing information (the date of admission and the date of onset of illness entered by the customer on the billing information input screen), and the contract information You can register a value calculated from the date of the first year contract.

No. 13 "site symptoms (injuries)" ~ No. Item values for the 15 “disease name” items can be obtained from the billing information.

No. If the name of the document information matches the insured person's information in the contract information, register "name match" as the item value for item 16 "name and insured person match". If not, "name mismatch" can be registered.

No. The item value for item 17 "Whether the period of hospitalization is within the insurance period of the contract information?" can be registered, and if it is not included, "out of period" can be registered.

No. 18 "Hospitalization days"-No. The item value for the item 21 "Patient's burden ratio" can be obtained from the information described in the document. No. The item value for the item 22 "Description of scores for specialized psychiatric therapy" can be "description" if data exists in the document description information, and "no description" if data does not exist.

In this way, items to be checked by the person in charge of assessment work when assessing a claim and items to be noted may be set as the assessment items. Also, the weighting factor for the determination item may be set based on how important the person in charge of assessment tasks regards the determination item. As a result, it is possible to determine the possibility that the claim will not be subject to payment based on the information reflecting the experience and know-how of the person in charge of assessment.

In the above embodiment, the non-payment possibility determination unit 35 determines the possibility that the customer's claim will not be covered by the insurance payment. , is synonymous with the possibility of being covered by an insurance claim. Therefore, the non-payable possibility determination unit 35 may determine the possibility that the claim is subject to insurance payment. Further, when the non-payment possibility determination unit 35 determines the possibility of being a payment target, in the determination DB, the field of "number of non-target records" is set to "number of target records" and the field of "non-target rate" is may be used as the "target rate". In this case, the determination DB creation unit 38 may store the number of records in which insurance claims have been paid in response to insurance claims from the insurance claim payment management table in the "number of target records" field. Also, the ratio of the number of target records to the total number of records may be stored in the "target rate" field.

In the above embodiment, the image information processing unit 32 converts the document image received from the customer terminal 60 of the customer into text by AI-OCR, thereby acquiring the information described in the document. It is not something that can be done. For example, the customer may provide the determination server 30 with information written on the document by inputting information written on the document at the customer terminal 60, and separately send the document as proof to the insurance company. Alternatively, the information described in the document may be provided to the judgment server 30 by the operator of the insurance company inputting the information described in the document based on the document sent by the customer.

In addition, in the above embodiment, a case has been described in which the document related to an insurance claim is a receipt for medical expenses. etc., but not limited to these as long as they are necessary for assessing whether or not to pay an insurance claim.

Also, the items included in the contract information and claim information and the items included in the insurance claim payment management table shown in the above embodiment are merely examples, and other items may be included.

Further, in the above embodiment, the determination DB creating unit 38 of the determination server 30 creates the determination DB, but the present invention is not limited to this. For example, the insurance payment management server 40 may create the determination DB.

Although the preferred embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications, Change is possible.

30 determination server 31 billing information receiving unit 32 image information processing unit 35 non-payment possibility determination unit 36 corresponding department determination unit 37 determination DB storage unit 38 determination DB creation unit 60 customer terminal 100 non-payment possibility determination system

Claims (8)

a first acquisition unit that acquires insurance claim information from a customer;
a second acquisition unit that acquires information described in documents related to the insurance claim;
For each combination of a plurality of predetermined items and the values registered in each of the plurality of predetermined items, the number of records corresponding to the combination in the past insurance claims corresponds to the above combination. referencing statistical data that stores the percentage of records for which claims were not paid out of the number of records;
With respect to the insurance claim, values set for items included in the claim information, values set for items included in the information described in the document, and the customer's insurance contract obtaining values for each of the plurality of predetermined items included in the statistical data using values set for the items included in the information;
weighting factors set for each of the plurality of predetermined items; a determination unit that calculates a value representing the possibility that the insurance claim will not be covered by the insurance payment, using the respective products of the ratio of the statistical data and
Possibility determination device for non-payment. a decision unit that decides, from among a plurality of departments, a department corresponding to the insurance claim based on the value representing the possibility;
The non-payment possibility determination device according to claim 1. 3. The non-payment possibility determination device according to claim 1, wherein said second acquisition unit acquires information described in said document by converting an image of said document into text by AI-OCR. 4. The item according to any one of claims 1 to 3 , wherein the plurality of predetermined items are items included in any one of the billing information, the contract information, and the information described in the document. Possibility determination device for non-payment. 5. Any one of claims 1 to 4 , wherein the plurality of predetermined items are items using a plurality of items included in the past insurance claim information, contract information, and information described in documents. or 1 item. The non-payment possibility determination device according to any one of claims 1 to 5 , further comprising a statistical data generation unit that generates the statistical data. a reception unit that receives input of insurance claim information from a customer; a first acquisition unit that acquires an image of a document related to the insurance claim; and a transmission unit that transmits the claim information and the image of the document. a first device having
For each combination of a second acquisition unit that acquires information described in the document from the image of the document, a plurality of predetermined items, and values registered in each of the plurality of predetermined items , with reference to statistical data storing the ratio of the number of records for which insurance claims were not paid out of the number of records corresponding to said combination to the total number of records corresponding to said combination in past insurance claims; , the values set for the items included in the billing information, the values set for the items included in the information described in the document, and the items included in the customer's insurance contract information using the values set for the statistical data to obtain the values for each of the plurality of predetermined items included in the statistical data, and the weights set for each of the plurality of predetermined items Using the product of each of the coefficient and the ratio of the statistical data corresponding to the combination of the plurality of predetermined items and the acquired values for each of the plurality of predetermined items, the insurance a determination unit that calculates a value representing the likelihood that the claim will not be covered by an insurance claim ;
A non-payment possibility determination system comprising: obtaining insurance claim information from a customer;
obtaining information contained in documents related to the insurance claim;
For each combination of a plurality of predetermined items and the values registered in each of the plurality of predetermined items, the number of records corresponding to the combination in the past insurance claims corresponds to the above combination. Statistical data storing the ratio of the number of records for which insurance claims were not paid out of the number of records is referred to, and with regard to the claims for insurance claims, the values set for the items included in the claim information are stored in the documents. Using the values set for the items included in the described information and the values set for the items included in the customer's insurance contract information, the predetermined data included in the statistical data a weighting factor set for each of the plurality of predetermined items, the plurality of predetermined items, and the obtained plurality of predetermined items calculating a value representing the likelihood that the insurance claim will not be covered by insurance, using the product of each of the percentage of the statistical data corresponding to the combination of the values for each of the items; ,
A computer-executed non-payment possibility determination method.

Images