Information processing device, support method, and support program

Abstract

To facilitate the disclosure of information required in insurance contracts. [Solution] The information processing device comprises an acquisition unit that acquires disclosure information indicating candidate items that may fall under the disclosure items in an insurance contract, and an extraction unit that extracts disclosure items related to the candidate items from a document containing the disclosure items for insurance, using an extraction model. Furthermore, this information processing device can also support the decision-making of applicants when declaring disclosure items.

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus, a support method, and a support program.

Background Art

[0002] When concluding an insurance contract, an applicant for insurance is required to disclose the disclosure items determined in advance for each insurance. For example, in the case of life insurance, medical insurance, etc., it is often required to disclose the presence or absence of specific injuries or illnesses in the recent few years. If a contract is concluded without disclosing the facts corresponding to the disclosure items, or if a disclosure of content different from the facts is made, the insurance contract may be cancelled as a violation of the disclosure obligation, and the insurance money or benefit payment may not be received. Therefore, it is important to accurately disclose all the necessary disclosure items without omission.

[0003] As a technology for supporting the disclosure for insurance contracts, for example, there is a disclosure information providing system described in Patent Document 1. In the disclosure information providing system described in Patent Document 1, a disease history determination master in which diseases that require disclosure when joining insurance are associated with information on medical treatment acts or pharmaceuticals corresponding to those diseases is prepared in advance. Then, in this disclosure information providing system, using the disease history determination master and the information on medical treatment acts and pharmaceuticals described in the receipt issued by a medical institution, the information on medical treatment acts and pharmaceuticals corresponding to predetermined diseases that require disclosure is extracted.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] The notification information provision system described in Patent Document 1 is based on the premise that a medical history judgment master can be prepared in advance and that necessary information can be obtained from medical claims. However, in reality, it is not easy to prepare a medical history judgment master for each of the many different insurance products, and it is not always possible to obtain the necessary information from medical claims.

[0006] Therefore, currently, most insurance applicants have to go through the cumbersome process of comparing their own medical history, which might fall under the disclosure items of the insurance they wish to purchase, with the disclosure items of the insurance they wish to purchase, in order to confirm whether each item needs to be disclosed. The illustrative purpose of this disclosure is to provide technology that can help facilitate the disclosure of disclosure items in insurance contracts. [Means for solving the problem]

[0007] An information processing device relating to an exemplary aspect of this disclosure includes: an acquisition means for acquiring disclosure information indicating candidate items that may fall under disclosure items that an applicant for an insurance contract must disclose at the time of the contract; and an extraction means for extracting related disclosure items, which are disclosure items related to the candidate items, from a document containing the insurance disclosure items, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document related to the data.

[0008] Other information processing devices relating to illustrative aspects of this disclosure include: acquisition means for acquiring disclosure information indicating candidate items that may fall under disclosure items that an applicant for an insurance contract must disclose at the time of the contract; and determination means for determining whether or not the candidate items fall under the disclosure items using a language model that has been trained on natural language.

[0009] In an exemplary aspect of this disclosure, the support method involves at least one processor performing an acquisition process to acquire disclosure information indicating candidate items that may fall under disclosure items that an applicant for an insurance contract must disclose at the time of the contract, and an extraction process to extract related disclosure items that are disclosure items related to the candidate items from a document containing the insurance disclosure items, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document related to the data.

[0010] The support program relating to an illustrative aspect of this disclosure causes a computer to function as an acquisition means for acquiring disclosure information indicating candidate items that may fall under disclosure items that an insurance contract applicant must disclose at the time of the contract, and as an extraction means for extracting related disclosure items, which are disclosure items related to the candidate items, from a document containing the insurance disclosure items, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document related to the data. [Effects of the Invention]

[0011] One illustrative aspect of this disclosure is that it can help facilitate the disclosure of information in insurance contracts. [Brief explanation of the drawing]

[0012] [Figure 1] This is a block diagram showing the configuration of the information processing device related to this disclosure. [Figure 2] This flowchart shows the flow of support methods related to this disclosure. [Figure 3] This is a block diagram showing the configuration of other information processing devices related to this disclosure. [Figure 4] This figure shows an example of extracting relevant notices. [Figure 5] This figure shows an example of how to determine whether a candidate item falls under the category of related notification items. [Figure 6]This figure shows an example of a UI (User Interface) screen that accepts correction instructions in natural language. [Figure 7] This figure shows an example of a UI screen presented to prompt users to check for missing information. [Figure 8] This figure shows an example of a UI screen that accepts input for a description of the history information. [Figure 9] Figure 3 is a flowchart showing the processing flow executed by the information processing device. [Figure 10] This is a block diagram showing the configuration of an information processing device related to a reference example. [Figure 11] This is a block diagram showing the configuration of a computer that functions as an information processing device related to this disclosure. [Modes for carrying out the invention]

[0013] The following are examples of embodiments of the present invention. However, the present invention is not limited to the exemplary embodiments shown below, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining some or all of the technologies (things or methods) employed in each of the exemplary embodiments shown below may also be included in the scope of the present invention. Furthermore, embodiments obtained by appropriately omitting some of the technologies employed in each of the exemplary embodiments shown below may also be included in the scope of the present invention. In addition, the effects mentioned in each of the exemplary embodiments shown below are examples of effects that can be expected in that exemplary embodiment and do not define the scope of the present invention. That is, embodiments that do not produce the effects mentioned in each of the exemplary embodiments shown below may also be included in the scope of the present invention.

[0014] [First Exemplary Embodiment] A first exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. This exemplary embodiment is a basic form for each of the exemplary embodiments described later. Note that the scope of application of each technology adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technology adopted in this exemplary embodiment can also be adopted in other exemplary embodiments included in the present disclosure as long as there are no particular technical obstacles. In addition, each technology shown in the drawings referred to for explaining this exemplary embodiment can also be adopted in other exemplary embodiments included in the present disclosure as long as there are no particular technical obstacles.

[0015] (Configuration of Information Processing Apparatus 1) The configuration of the information processing apparatus 1 according to this exemplary embodiment will be described with reference to FIG. 1. FIG. 1 is a block diagram showing the configuration of the information processing apparatus 1. As shown in FIG. 1, the information processing apparatus 1 includes an acquisition unit 101 and an extraction unit 102.

[0016] The acquisition unit 101 acquires notification information indicating candidate items that may correspond to notification items, which are items to be notified by an applicant for an insurance contract at the time of the contract.

[0017] Here, the above-mentioned "insurance" may be any insurance for which notification items are defined. For example, the above-mentioned insurance may be insurance related to the life and health of the insured, such as life insurance, pension insurance, medical insurance, etc., or insurance related to the owned property, such as automobile insurance and fire insurance.

[0018] In addition, the above-mentioned "notification items" are items subject to the notification obligation. Generally, matters that are objectively recognized as matters that the insurer would not conclude the contract if it knew the circumstances, or at least would not conclude the contract under the same conditions, are the notification items. For example, for insurance related to the life and health of the insured, such as life insurance, the current health status and past medical history of the insured are the notification items.

[0019] Furthermore, the "notification information" mentioned above only needs to indicate at least one candidate item that may fall under the category of notification items. For example, the acquisition unit 101 may acquire as notification information text data listing items that the applicant considers to be potentially subject to notification items.

[0020] Furthermore, for example, the acquisition unit 101 may acquire the applicant's medical history information as disclosure information. The history information only needs to show the applicant's history that may be related to the disclosure items. For example, the acquisition unit 101 may acquire history information such as the applicant's medical history (which may include their current health status), whether they regularly visit hospitals, their prescription history, the results of health checkups, the results of medical examinations by doctors, the results of medical tests, and their history of receiving public assistance.

[0021] The method for obtaining notification information is optional. Furthermore, the acquisition unit 101 may acquire multiple types of notification information. For example, the acquisition unit 101 may acquire both text data entered by the applicant and historical information recorded in a predetermined database (for example, a database that records various information related to medical insurance, etc., associated with personal identification information assigned to each citizen) as notification information.

[0022] The extraction unit 102 uses an extraction model trained on machine learning to take a document and data pair as input and output the portion of the document related to the data in question. It extracts disclosure items related to the candidate items shown in the disclosure information acquired by the acquisition unit 101 from a document containing insurance disclosure items. Since the disclosure items extracted by the extraction unit 102 are disclosure items related to the candidate items, these disclosure items will be referred to as related disclosure items below.

[0023] The "extraction model" described above can be any model that can be used to extract relevant disclosure items from a document containing insurance disclosure items. For example, a model that divides a document containing disclosure items into predetermined units of text (e.g., each sentence or each disclosure item), calculates a score indicating the similarity of the content to candidate items for each of the resulting texts, and outputs texts whose calculated scores are above a predetermined threshold can be used as the extraction model described above. Such a model can be generated, for example, by machine learning using training data that associates the above scores for text and data pairs with ground truth data. The data paired with the text is typically text data, but other formats of data such as image data can also be used.

[0024] Furthermore, the "extraction model" described above may be a general-purpose model that can be used for purposes other than extracting relevant notices, or it may be a general-purpose model that has been fine-tuned for the extraction of relevant notices. Also, the extraction model may be provided by the information processing device 1, or it may be provided by another device. In the latter case, the extraction unit 102 utilizes the extraction model via another device that provides the extraction model.

[0025] Furthermore, the "document containing the insurance disclosure items" mentioned above only needs to contain at least some of the disclosure items for the insurance that the applicant intends to contract. For example, the application form, explanatory document, or contract terms and conditions of the insurance can be used as a document containing the insurance disclosure items.

[0026] As described above, the information processing device 1 according to this exemplary embodiment is configured to include: an acquisition unit 101 that acquires disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract; and an extraction unit 102 that uses an extraction model trained on machine learning to take a set of document and data as input and output the portion of the document related to the data, to extract related disclosure items that are disclosure items related to the candidate items from a document that contains the insurance disclosure items.

[0027] The relevant disclosure items extracted in the manner described above are related to the candidate items, and therefore can be said to be items that are highly likely to be disclosed by the applicant among the disclosure items described in the document containing the disclosure items. Furthermore, the disclosure information necessary for extracting relevant disclosure items only needs to indicate candidate items that may fall under the disclosure items, and may, for example, indicate candidate items that the insurance applicant is not certain fall under the disclosure items.

[0028] In other words, according to the above configuration, it is possible to extract relevant disclosure items that are highly likely to be items that the insurance applicant should disclose, based on candidate items that the insurance applicant may not be certain are subject to disclosure. Therefore, the information processing device 1 has the effect of supporting the easy disclosure of disclosure items in insurance contracts. Furthermore, the information processing device 1 can also support the applicant's decision-making when declaring disclosure items, that is, the decision of whether or not to disclose candidate items as disclosure items.

[0029] Furthermore, how the extracted relevant disclosure items are used to facilitate disclosure is at the discretion of the user. For example, the information processing device 1 may present the extracted relevant disclosure items to the applicant as reference information. In this case, the applicant can decide whether or not to disclose the candidate items by referring to the presented relevant disclosure items. Alternatively, for example, as described in Exemplary Embodiment 2 below, the device may determine whether or not the above candidate items fall under the category of relevant candidate items, and present the candidate items determined to fall under the category as items that the applicant should disclose.

[0030] (Support Program) The functions of the information processing device 1 described above can also be implemented by a program. The support program according to this exemplary embodiment is an insurance contract application support program, which causes a computer to function as an acquisition means for acquiring disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and as an extraction means for extracting related disclosure items, which are disclosure items related to the candidate items, from a document containing the insurance disclosure items, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document related to the data. This support program has the effect of making it easier to disclose disclosure items in an insurance contract.

[0031] (Flow of support methods) The flow of the support method according to this exemplary embodiment will be explained with reference to Figure 2. Figure 2 is a flowchart showing the flow of the support method. Note that the entity executing each step in this support method may be a processor provided in the information processing device 1, a processor provided in another device, or the entity executing each step may be a processor provided in a different device.

[0032] In S1 (acquisition process), at least one processor acquires disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract.

[0033] In S2 (extraction process), at least one processor uses an extraction model trained on machine learning to take a document and data pair as input and output the portion of the document that is relevant to the data in question, to extract related disclosure items from the document containing the insurance disclosure items, which are disclosure items related to the candidate items shown in the disclosure information obtained in S1.

[0034] As described above, the support method according to this exemplary embodiment is a method for supporting an application for an insurance contract, and employs a configuration in which at least one processor performs an acquisition process to acquire disclosure information indicating candidate items that may fall under the disclosure items that an insurance applicant must disclose at the time of the contract, and an extraction process to extract related disclosure items that are disclosure items related to the candidate items from a document containing the insurance disclosure items, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document related to the data. This support method has the effect of making it easier to disclose disclosure items in an insurance contract.

[0035] [Second exemplary embodiment] A second exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. Components having the same function as those described in the above-described exemplary embodiment are denoted by the same reference numerals, and their descriptions are omitted as appropriate. The scope of application of each technology adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technology adopted in this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure, to the extent that no particular technical problems arise. Furthermore, each technology shown in the drawings referenced to describe this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure, to the extent that no particular technical problems arise.

[0036] (Configuration of Information Processing Device 1A) The configuration of the information processing device 1A according to this exemplary embodiment will be described with reference to Figure 3. Figure 3 is a block diagram showing the configuration of the information processing device 1A. The information processing device 1A is a device equipped with a function to support the application for an insurance contract. The information processing device 1A may be a local device used by individual users, or it may be a server that provides insurance contract application support services to multiple users.

[0037] As shown in the figure, the information processing device 1A includes a control unit 10A that controls all parts of the information processing device 1A, and a storage unit 11A that stores various data used by the information processing device 1A. The information processing device 1A also includes a communication unit 12A for the information processing device 1A to communicate with other devices, an input unit 13A that receives input to the information processing device 1A, and an output unit 14A for the information processing device 1A to output data. The control unit 10A includes an acquisition unit 101A, an extraction unit 102A, a determination unit 103A, a presentation control unit 104A, a reception unit 105A, and an input support unit 106A.

[0038] The acquisition unit 101A acquires disclosure information that indicates candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, similar to the acquisition unit 101 in exemplary embodiment 1. The acquisition unit 101A may also acquire various other information that needs to be entered when entering into an insurance contract (for example, applicant attribute information such as name and date of birth).

[0039] Similar to the extraction unit 102 in Exemplary Embodiment 1, the extraction unit 102A uses an extraction model trained on machine learning to take a document and data pair as input and output the portion of the document related to the data, to extract related disclosure items, which are disclosure items related to candidate items, from a document containing insurance disclosure items. Hereinafter, the extraction model used by the extraction unit 102A will be referred to as extraction model M1.

[0040] The determination unit 103A determines whether the candidate items shown in the notification information acquired by the acquisition unit 101A correspond to notification items, using a language model trained on natural language. More specifically, the determination unit 103A determines whether the candidate items shown in the notification information acquired by the acquisition unit 101A correspond to the related notification items extracted by the extraction unit 102A.

[0041] However, the extraction of relevant disclosure items by the extraction unit 102A is not a necessary prerequisite for the determination by the determination unit 103A. For example, the determination unit 103A may perform a process to determine whether the candidate items shown in the disclosure information acquired by the acquisition unit 101A correspond to the disclosure items of the insurance that the applicant intends to contract, for each of the disclosure items of the insurance. Compared to the case where relevant disclosure items are extracted, the number of determinations performed by the determination unit 103A will increase, but even if such a process is adopted, it is still possible to identify the items that the applicant should disclose.

[0042] Here, machine learning natural language means, more specifically, learning the arrangement of its constituent elements (such as words) in natural language sentences, and the arrangement of sentences in a document. Examples of language models that have learned natural language include BERT (Bidirectional Encoder Representations from Transformers), RoBERTa (Robustly optimized BERT approach), and ELECTRA (Efficiently Learning an Encoder that Classifies Token Replacements Accurately). In the following, the language model used by the determination unit 103A will be referred to as language model M2.

[0043] In this exemplary embodiment, we describe an example in which the language model M2 accepts input in the form of a text prompt written in natural language and outputs a response in natural language. However, the language model M2 may also be a model that can accept input in the form of data other than text data, such as images.

[0044] The language model M2 may be a general-purpose language model that can be used for purposes other than inferring whether a candidate item corresponds to a notice item, or it may be a general-purpose language model that has been fine-tuned for inferring whether a candidate item corresponds to a notice item. Furthermore, the language model M2 may be provided by the information processing device 1A or by another device. In the latter case, the determination unit 103A utilizes the language model M2 via another device that is equipped with the language model M2.

[0045] The presentation control unit 104A presents various information related to supporting insurance contract applications. For example, the presentation control unit 104A presents candidate items that the determination unit 103A has determined to be related disclosure items as items that the applicant should disclose. Also, for example, the presentation control unit 104A presents the inference results output by the language model M2. As will be described in detail later, the presentation control unit 104A may present the above inference results along with the basis for the inference. The method of presenting information is arbitrary. For example, the presentation control unit 104A may present the information by having the output unit 14A output the information, or by having the communication unit 12A output the information to another device. Furthermore, the information can be presented in any manner, such as display, printing, voice, or a combination thereof.

[0046] The reception unit 105A receives various instructions related to assisting with insurance contract applications. For example, the reception unit 105A receives instructions to modify the results of inference by the language model M2. The method of receiving instructions is arbitrary. For example, the reception unit 105A may receive instructions input via the input unit 13A, or it may receive instructions from other devices via the communication unit 12A.

[0047] The input support unit 106A assists the applicant in entering various items that they are required to enter when proceeding with the insurance contract. For example, the input support unit 106A enters the candidate items that the determination unit 103A has determined to be related disclosure items into the disclosure item input field in the insurance input form that the applicant intends to contract.

[0048] It should be noted that the judgment result of the judgment unit 103A is not necessarily correct, so it is preferable to present the candidate items to the applicant and have them check for any errors before entering them into the input form. Furthermore, it is not mandatory to provide the input support unit 106A. If the input support unit 106A is not provided, the applicant can refer to the candidate items presented by the presentation control unit 104A (candidate items that have been determined to be related to the notification items) when entering them into the input form or filling out the application form.

[0049] As described above, the information processing device 1A includes an acquisition unit 101A that acquires disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and an extraction unit 102A that uses an extraction model M1 that has been trained to take a document and data set as input and output the portion of the document related to the data to extract related disclosure items that are disclosure items related to the candidate items from a document that contains the insurance disclosure items.Therefore, the information processing device 1A has the effect of being able to support the easy disclosure of disclosure items in an insurance contract, similar to the information processing device 1.

[0050] Furthermore, as described above, the information processing device 1A includes a determination unit 103A that determines whether a candidate item falls under the category of related notification items using a language model M2 that has been trained on natural language, and a presentation control unit 104A that presents the candidate item that the determination unit 103A has determined to fall under the category of related notification items as an item that the applicant should be notified about. As a result, in addition to the effects of the information processing device 1, the effect of making the applicant aware of the candidate items to be notified and enabling them to make the notification smoothly can be obtained.

[0051] Furthermore, the determination by the determination unit 103A can be omitted. If the determination by the determination unit 103A is omitted, the presentation control unit 104A only needs to present the relevant notification items extracted by the extraction unit 102A. In this case, the applicant only needs to compare the presented relevant notification items with the candidate items and decide whether or not to announce the candidate items. This eliminates the need for the applicant to search for notification items related to the candidate items from the document that defines the notification items, thereby simplifying the announcement process.

[0052] Furthermore, as described above, the information processing device 1A includes an acquisition unit 101A that acquires disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and a determination unit 103A that determines whether or not the candidate items shown in the disclosure information acquired by the acquisition unit 101A fall under the disclosure items using a language model that has been trained on natural language. With this information processing device 1A, an objective determination result can be obtained as to whether or not the candidate items fall under the disclosure items. Therefore, the effect is obtained that it is possible to support the easy disclosure of disclosure items in insurance contracts.

[0053] (Example of extracted related notices) An example of extracting relevant disclosure items by the information processing device 1A will be explained with reference to Figure 4. Figure 4 is a diagram showing an example of extracting relevant disclosure items. In the example in Figure 4, the applicant for the insurance contract is inputting disclosure information 401 into the information processing device 1A. The input disclosure information 401 is acquired by the acquisition unit 101A provided in the information processing device 1A.

[0054] The notification information 401 is a list of several items that the applicant believes may fall under the category of items to be notified, i.e., the candidate items mentioned above. The notification information 401 also asks whether these candidate items are subject to notification, or in other words, whether these items fall under the category of items to be notified. The applicant may input the notification information 401, for example, via the input unit 13A, or via the communication unit 12A using their own terminal device. The applicant may also input the notification information 401 as text data or as voice data. In the latter case, the acquisition unit 101A can acquire the notification information 401 in text format by having the input voice data recognized by the information processing device 1A or other voice recognition device.

[0055] Furthermore, in the example shown in Figure 4, the information processing device 1A (more specifically, the acquisition unit 101A) acquires document 402, which contains the disclosure items for the insurance that the applicant is trying to contract, from database D. If the system is configured to acquire document 402 from database D, then documents containing the disclosure items for various types of insurance should be recorded in database D in association with the identification information of that insurance. Then, the identification information of the insurance that the applicant is trying to contract should be input into the information processing device 1A. As a result, the acquisition unit 101A can acquire document 402, which contains the disclosure items for the insurance that the applicant is trying to contract, from database D. Note that the method of acquiring document 402 is arbitrary and is not limited to the example described above. For example, the applicant may input document 402 together with the disclosure information 401 into the information processing device 1A, or document 402 may be stored in the information processing device 1A in advance.

[0056] Next, the information processing device 1A (more specifically, the extraction unit 102A) inputs the notification information 401 and document 402 acquired by the acquisition unit 101A as described above into the extraction model M1. As a result, the related notification items 403 are output from the extraction model M1.

[0057] Related disclosure item 403 is a disclosure item that relates to at least one of the multiple candidate items shown in disclosure information 401, which are among the disclosure items specified in document 402. Specifically, related disclosure item 403 refers to the disclosure item "(1) Within the last five years, I have been hospitalized due to cerebral hemorrhage, myocardial infarction, or heart failure," which relates to the candidate items shown in disclosure information 401, namely "I was hospitalized last year due to a subarachnoid hemorrhage" and "I had a myocardial infarction 20 years ago."

[0058] As will be explained in detail below, the determination unit 103A determines whether the candidate items shown in the notification information 401 correspond to the related notification items 403 extracted in the manner described above. Alternatively, as described above, the presentation control unit 104A may present the extracted related notification items 403 to the applicant. In this case, the presentation control unit 104A may either extract and present the portion containing the related notification items 403 from document 402, or present document 402 with the portion containing the related notification items 403 highlighted. The highlighting should be done in a manner that allows for the distinction between the highlighted and unhighlighted portions. For example, the presentation control unit 104A may highlight the portion containing the related notification items 403 in document 402 by changing the background color of the highlighted portion, changing the font of the highlighted portion, etc.

[0059] (Example of determining whether a candidate item falls under the category of related notification items) Figure 5 shows an example of determining whether a candidate item falls under the category of related notification items. As described above, this determination is performed by the determination unit 103A. The language model M2 is also used for this determination. Therefore, the determination unit 103A generates a prompt describing the content of the instructions for the language model M2 and inputs it to the language model M2.

[0060] In the example shown in Figure 5, the information processing device 1A (more specifically, the determination unit 103A) inputs prompt 501 to the language model M2. Prompt 501 includes the related notification item 403 shown in Figure 4 and each candidate item shown in the notification information 401 shown in Figure 4, and instructs the device to infer the relationship between the related notification item 403 and the candidate item.

[0061] More specifically, each candidate item shown in disclosure information 401 is described in the "Medical History" section of prompt 501. Additionally, related disclosure item 403 is described in the "Disclosure Items" section of prompt 501. Furthermore, prompt 501 includes the statement, "You are a physician and are currently conducting an insurance underwriting assessment." While including such a statement is not mandatory, it is expected to improve inference accuracy.

[0062] Furthermore, prompt 501 instructs the system to check whether it can be inferred that the medical history falls under the disclosure items. The wording of the prompt can be changed as appropriate within the range to obtain the desired inference result. For example, the determination unit 103A may generate prompts with different inference instructions depending on the target candidate items, the type of insurance, the language model used, etc.

[0063] Furthermore, prompt 501 includes an "answer format" and instructs the user to answer using this format. By specifying the answer format in this way, it becomes possible to obtain inference results in the desired format. This answer format also includes an item called "basis for inference." By including such an item, the language model M2 can output the basis for the inference along with the inference result. Note that the instruction to output the basis for the inference may also be given by including a sentence such as "Please answer with the basis along with the inference result" in the prompt.

[0064] Furthermore, prompt 501 may include sentences that specify the output conditions. For example, prompt 501 may include sentences such as, "For every sentence in the medical history, it is necessary to infer whether or not it corresponds to a disclosure item," or "The number of elements included in the answer format must match the number of sentences in the medical history." This makes it possible to improve the inference accuracy of the language model M2.

[0065] In prompt 501, everything except the content of "medical history" and "disclosure items" is standard. For this reason, the parts of prompt 501 other than the content of "medical history" and "disclosure items" may be stored as a standard template in the storage unit 11A or the like. This allows the determination unit 103A to generate prompt 501 by inputting the candidate items shown in the disclosure information 401 acquired by the acquisition unit 101A and the related disclosure items 403 extracted by the extraction unit 102A into the above template.

[0066] The inference result 502 shown in Figure 5 is an example of an inference result obtained by inputting prompt 501 into language model M2. Inference result 502 shows the result of inferring the relationship between related notification items and candidate items in the format of the response format shown in prompt 501. Specifically, inference result 502 shows the inference result that for each candidate item shown in notification information 401, the candidate item "corresponds to notification items" or "does not correspond to notification items". In addition, inference result 502 shows the basis for the inference along with the inference result for each candidate item.

[0067] Note that the prompt 501 in Figure 5 is a prompt for determining whether multiple candidate items fall under the related notification item 403 at once. However, prompts may be used to determine whether each candidate item falls under the related notification item 403 individually. In this case, the determination unit 103A only needs to input the candidate item and the related notification item 403 into the language model M2 for each candidate item and determine whether the candidate item falls under the related notification item 403.

[0068] Furthermore, if multiple related notices are extracted, the determination unit 103A may generate a prompt that includes multiple related notices and multiple candidate items, instructing the determination unit 103A to infer the relationship between each candidate item and each related notice (for example, whether or not it corresponds to a related notice, and if so, which related notice it corresponds to). Alternatively, in this case, the determination unit 103A may generate a prompt similar to prompt 501 in Figure 5 for each related notice and infer the relationship between each related notice and each candidate item. Alternatively, the determination unit 103A may input one of the multiple related notices and one of the candidate items into the language model M2 and perform the process of determining whether the candidate item corresponds to the related notice for each of the multiple candidate items. By performing this process for each related notice, the determination unit 103A can determine which related notice each candidate item corresponds to (or does not correspond to any of the related notices).

[0069] The format in which the inference results are output can be specified by a prompt. For example, the determination unit 103A may generate a prompt instructing the user to answer with one of three options: that the item is a notice item, neutral, or that the item is not a notice item. In this case, when the determination unit 103A outputs a response indicating that a certain candidate item belongs to a certain notice item, it should determine that the candidate item is a notice item. The processing to be performed when the answer is neutral can be predetermined. For example, when the presentation control unit 104A receives a response of neutral to a prompt asking whether a certain candidate item is a related notice item, it may present the applicant with combinations of those candidate items and related notice items, and ask the applicant to input whether the candidate item is a related notice item. Alternatively, for example, the determination unit 103A may generate a prompt instructing the user to output a numerical value (for example, a number between 0 and 1) indicating the likelihood of the item being a related notice item. In this case, the determination unit 103A only needs to determine that a candidate item corresponds to a related notification item if the numerical value output for a combination of a candidate item and a related notification item is greater than or equal to a predetermined threshold.

[0070] Furthermore, Figure 5 shows an example screen 503, which is an example of a UI screen for presenting the inference results 502. In example screen 503, the inference result of whether each candidate item corresponds to a related notification item is presented along with the reasoning behind the inference. The presentation control unit 104A can generate and present a UI screen like example screen 503 using the various information shown in the inference results 502. In this way, the presentation control unit 104A may present the candidate items that the determination unit 103A has determined to correspond to related notification items to the applicant as items that the applicant should notify. In addition, the presentation control unit 104A may present the reasoning behind the inference output by the language model M2 to the applicant as a basis for judging whether the inference result is appropriate or not.

[0071] Furthermore, if the output unit 14A has a function to display and output an image, the presentation control unit 104A may cause the output unit 14A to display the UI screen. Alternatively, the presentation control unit 104A may cause the UI screen to be displayed on an external display device of the information processing device 1A (for example, a display device on the terminal device used by the applicant) via the communication unit 12A.

[0072] Furthermore, screen example 503 is designed to accept instructions for correcting the inference results. Specifically, screen example 503 displays text prompting the user to select the "Correct" button if there is an error in the inference result, and also displays the corresponding button (software key) for correcting the inference result. This button should be displayed for each of the inference results for multiple candidate items.

[0073] When the user selects the "Edit" button, the reception unit 105A receives the operation and notifies the judgment unit 103A accordingly. Upon receiving this notification, the judgment unit 103A changes the judgment result for the candidate item in question. Specifically, if the judgment unit 103A receives an instruction to edit a candidate item that it has inferred to "belongs to a notice item," it changes the judgment result for that candidate item to "does not belong to a notice item." Similarly, if the judgment unit 103A receives an instruction to edit a candidate item that it has inferred to "does not belong to a notice item," it changes the judgment result for that candidate item to "belongs to a notice item."

[0074] As described above, the determination unit 103A may generate a prompt 501 that includes the relevant notification item and the candidate item, instructing it to infer the relationship between the relevant notification item and the candidate item, and then determine whether the candidate item corresponds to the relevant notification item based on the inference result 502, which is the output obtained by inputting the generated prompt 501 to the language model M2. This makes it possible to appropriately determine whether the candidate item corresponds to the relevant notification item.

[0075] (Regarding the presentation of inference results and re-inference) The reception unit 105A may accept correction instructions that express the corrections to the inference results in natural language. This will be explained with reference to Figure 6. Figure 6 is a diagram showing an example of a UI screen that accepts correction instructions in natural language. In addition to the example screen 601, which is an example of a UI screen that accepts correction instructions in natural language, Figure 6 also shows a prompt 602 that instructs re-inference based on the correction instructions in natural language, and a re-inference result 603 obtained by inputting the prompt 602 into the language model M2.

[0076] Screen example 601 displays the inference results of language model M2 and the notification items (related notification items) that were inferred. It also prompts the user to check if there are any errors in the inference results and asks them to input the details of any errors and instruct the system to re-infer. Screen example 601 also displays a text box for entering corrections and a button (software key) for instructing the system to re-infer.

[0077] The presentation control unit 104A can allow the applicant to confirm the inference results by presenting a UI screen, such as the example screen 601. The reception unit 105A can also accept corrections to the inference results via a UI screen, such as the example screen 601.

[0078] In the example screen 601, when the content of the correction is entered into the text box and the button to instruct re-inference is operated, the receiving unit 105A receives the entered content of the correction as a correction instruction. For example, the receiving unit 105A may accept a correction instruction that describes in natural language why the inference result is incorrect. Alternatively, for example, the receiving unit 105A may accept a correction instruction that describes in natural language whether or not the candidate item that was the subject of the inference corresponds to a related notification item.

[0079] Then, the determination unit 103A re-determines whether the candidate item corresponds to the relevant notification item based on the correction instruction received by the reception unit 105A. For example, the determination unit 103A may generate the prompt 602 shown in Figure 6 and input it to the language model M2, and then re-determine whether the candidate item corresponds to the relevant notification item based on the re-inference result 603 output from the language model M2.

[0080] Prompt 602 contains the text of the correction instructions received by the reception unit 105A and instructs the system to infer the relationship between candidate items and related disclosure items based on those correction instructions. Prompt 602 also includes the sentence, "However, do not output any words that are not included in the answer format." Thus, even in prompts that instruct re-inference, sentences specifying output conditions may be included. This prevents unnecessary content (for example, sentences such as "Apologies. I will re-output." or "Content has been updated") from being output from the language model M2, thus preventing any problems with inputting information into insurance input forms, etc. Prompt 602 can also be generated using a predetermined template, similar to prompt 501 shown in Figure 5. In addition, prompt 602 may also contain a list of past medical history (candidate items), related disclosure items, and an answer format, similar to prompt 501.

[0081] In the re-inference result 603 shown in Figure 6, the inference result regarding the relationship between the candidate item and the related disclosure item has changed from that shown on the UI screen 601, reflecting the content of the "comment," or correction instruction, shown in prompt 602. Specifically, in the re-inference result 603, the inference result regarding whether the candidate item "has been diagnosed with diabetes" is a related disclosure item has changed to "does not fall under the disclosure item." Furthermore, the basis for this inference has changed to the sentence "does not meet the condition of having been hospitalized for diabetes." In this way, by performing re-inference using prompt 602 which includes the natural language sentence received as a correction instruction, it is possible to output a re-inference result 603 that reflects the content of the correction instruction.

[0082] As described above, the determination unit 103A may generate a prompt (for example, prompt 501 in Figure 5) that instructs the system to output the reasoning behind the reasoning, along with the reasoning behind the reasoning, which is the reasoning behind

[0083] Furthermore, as described above, the information processing device 1A is equipped with a receiving unit 105A that receives instructions to correct the inference results. The determination unit 103A then re-determines whether the candidate items correspond to the relevant notification items based on the correction instructions received by the receiving unit 105A. This provides the additional benefit of correcting errors in the inference results so that applicants can make the notifications correctly, in addition to the effects achieved by the information processing device 1.

[0084] Furthermore, as described above, the reception unit 105A may accept correction instructions expressed in natural language. In this case, the determination unit 103A generates a prompt (for example, prompt 602 in Figure 6) that includes the correction instructions received by the reception unit 105A and instructs the determination unit 103A to infer the relationship between the candidate items and related notification items based on the correction instructions. Based on the output obtained by inputting the generated prompt into the language model M2, the determination unit 103A re-determines whether the candidate items correspond to related notification items. This provides the effect of enabling appropriate re-determination by taking into account the intent of the correction instructions, in addition to the effects performed by the information processing device 1.

[0085] One example of how the intent behind correction instructions can be understood is the absorption of variations in wording. For example, suppose the inference result obtained is that the candidate item "I have been diagnosed with diabetes" corresponds to the related disclosure item "I have been hospitalized for diabetes within the last 5 years," as shown in Figure 6. In this case, even if the applicant enters a correction instruction that is different from, for example, "I only receive outpatient treatment" or "My hospitalization was more than 10 years ago," as shown in Figure 6, the re-inference result can be determined to be "Does not apply to the disclosure item."

[0086] Furthermore, accepting correction instructions expressed in natural language has the advantage of allowing the content of the correction instructions to be reflected in other inferences. For example, suppose the inference result that the candidate item "Diagnosed with diabetes at Clinic X" corresponds to the related disclosure item "Have been hospitalized for diabetes within the last 5 years" is corrected by the input correction instruction "Clinic X does not have inpatient facilities." In this case, the judgment unit 103A can not only correct the inference result for the candidate item to "Does not correspond to the disclosure item," but can also update other inference results based on the fact that Clinic X does not have inpatient facilities.

[0087] (Reuse of correction instructions) The content of the correction instructions described above may be recorded in the memory unit 11A or an external database, etc., and used for subsequent inferences. For example, the comment of the correction instruction, "X Clinic does not have inpatient facilities," may be recorded, and this comment may be used for inferences regarding other candidate items. This will allow subsequent inferences (which may be limited to inferences regarding the same applicant, or may be reused for inferences regarding other applicants) to obtain inference results that take into account that X Clinic does not have inpatient facilities.

[0088] Furthermore, the presentation control unit 104A may present the recorded correction instructions to the applicant, and the reception unit 105A may accept corrections, deletions, and additions to the correction instructions. This allows the content of the correction instructions, which align with the applicant's intentions, to be reflected in subsequent inferences without requiring retraining of the language model M2. The correction instructions can be reused in the same way as when they are used for the first time, by including them in the prompt. In other words, the determination unit 103A generates a prompt that includes the recorded correction instructions and instructs the language model M2 to infer the relationship between the candidate items and related notification items based on the correction instructions, and then inputs the generated prompt to the language model M2 to output an inference result indicating whether or not the candidate items correspond to related notification items.

[0089] (Regarding iterative reasoning) Since the language model M2 is a probabilistic model, even if the exact same prompt is entered, the inference results in multiple inferences may differ. Furthermore, it has been found that inference results that differ from the facts tend not to be output repeatedly. For this reason, the determination unit 103A may perform the process of inputting a prompt to the language model M2 and outputting an inference result multiple times. In this case, the determination unit 103A may determine that candidate items with large variations in inference results do not fall under the category of related notification items. For example, the determination unit 103A may calculate a score for each candidate item indicating the magnitude of the variation in inference results (for example, the proportion of inference results that differ in content from other inference results out of all inference results), and determine that candidate items whose calculated score exceeds a predetermined threshold do not fall under the category of related notification items.

[0090] (Regarding checking for any missing entries) As mentioned above, if an insurance contract is concluded with omissions in the disclosure of required information, there is a possibility that the insurance payout will not be received. For this reason, the information processing device 1A may prompt the applicant to check whether there are any omissions in the disclosure of required information. This will be explained with reference to Figure 7. Figure 7 shows an example of a UI screen presented to prompt the applicant to check for any missing information.

[0091] The example screen 701 shown in Figure 7 displays each candidate item indicated in the notification information acquired by the acquisition unit 101A, and the result of the determination of whether or not those candidate items correspond to related notification items. The example screen 701 also displays text prompting the user to check for any information other than the above candidate items, such as past medical history, and to enter such information if available. In addition to a button (software key) for confirming the notification content, the example screen 701 also displays a text box for entering unentered items (candidate items) and a button (software key) for instructing the determination of whether or not those items correspond to notification items.

[0092] The display control unit 104A can show the applicant a UI screen like the example screen 701, allowing the applicant to check for any missing information. In particular, there are many cases where applicants mistakenly believe they have fulfilled their disclosure obligations by telling their insurance agent, etc., and thus fail to disclose information. Therefore, it is preferable to display a message prompting the applicant to also enter the information they have told their insurance agent, etc., as shown in the example screen 701.

[0093] In the example screen 701, if the button for confirming the disclosure content is pressed, the input support unit 106A confirms the candidate items that the determination unit 103A has determined to be related disclosure items at that time as disclosure items for the insurance that the applicant intends to contract.

[0094] On the other hand, if a new candidate item is entered into the text box and a button is pressed to instruct whether or not that candidate item corresponds to a notice item, the acquisition unit 101A acquires the entered new candidate item as new notice information. Subsequently, as in the case where notice information was acquired earlier, the extraction unit 102A extracts related notice items and the determination unit 103A makes a determination, and the result of the determination of whether or not the new candidate item corresponds to a notice item is presented.

[0095] (Regarding the use of historical information) The information to be disclosed acquired by the acquisition unit 101A may be information entered by the applicant that they believe may fall under the disclosure items, or it may be the applicant's medical history information, or it may be both. In this case, if the acquisition unit 101A acquires both the information entered by the applicant and the history information as disclosure information, there may be items that are not shown in the information entered by the applicant but are shown in the history information that may fall under the disclosure items.

[0096] Therefore, the acquisition unit 101A may perform a process of acquiring candidate items from the information entered by the applicant, as well as acquiring candidate items from the history information, and then comparing these candidate items. If the presentation control unit 104A detects candidate items that were not acquired from the information entered by the applicant but were acquired from the history information through the above process, it may present those candidate items to the applicant and prompt them to input whether or not they want to be notified. This makes it possible to supplement the information entered by the applicant and provide comprehensive notifications.

[0097] Furthermore, the historical information acquired by the acquisition unit 101A may not contain sufficient information to determine whether each item shown in the historical information corresponds to an insurance disclosure item. For this reason, the presentation control unit 104A may present candidate items acquired from the historical information to the applicant and prompt them to input information to determine whether or not they correspond to a disclosure item. This will be explained with reference to Figure 8.

[0098] Figure 8 shows an example of a UI screen that accepts input for a description of historical information. In the example screen 801 shown in Figure 8, a message is displayed indicating that historical information potentially related to the notification has been detected, along with the detected historical information itself. Specifically, this historical information indicates that the patient was hospitalized at Y Hospital between September 1st and 5th, 2021.

[0099] Furthermore, screen example 801 displays text prompting the user to enter the reason for hospitalization and press the "Determine" button. In addition, screen example 801 also displays a text box for entering the reason for hospitalization and a button (software key) to instruct the system to determine whether the information shown in the history information falls under the category of matters to be disclosed, taking the entered reason into consideration. The reception unit 105A can accept input of the reason for hospitalization through such a UI screen.

[0100] In example screen 801, when the reason for hospitalization is entered in the text box and the "Judge" button is operated, the extraction unit 102A extracts related disclosure items. In this case, the extraction unit 102A treats the pair of detected history information and the entered reason for hospitalization as one candidate item and extracts related disclosure items associated with that candidate item. For example, if the entered reason for hospitalization is "heart failure", the extraction unit 102A inputs the sentence "2021 / 9 / 1-5: Admitted to Y Hospital, Reason for hospitalization: Heart failure" as a candidate item into the extraction model M1 and extracts related disclosure items. After that, the judgment unit 103A makes a judgment and presents the judgment result of whether or not the item shown in the history information corresponds to a disclosure item.

[0101] In the example shown in Figure 8, the detected history information indicates a hospitalization history. Therefore, the reception unit 105A accepts input for the reason for hospitalization. Thus, the reception unit 105A only needs to accept input for additional information corresponding to the content of the history information, in other words, for explanatory text that explains the history information. For example, if history information indicating a drug prescription history is detected, the reception unit 105A only needs to accept input for explanatory text about the prescribed drug.

[0102] As described above, the information processing device 1A includes a presentation control unit 104A that presents the applicant's history information regarding the notification items, and a reception unit 105A that receives input of an explanatory text describing the presented history information. The extraction unit 102A then extracts related notification items associated with the candidate item, treating the set of history information and the explanatory text as one candidate item. This provides the effect of being able to extract appropriate related notification items that take into account the input explanatory text, in addition to the effects performed by the information processing device 1. Furthermore, the determination unit 103A may use the language model M2 to determine whether the candidate item, including the history information and the explanatory text, corresponds to a related notification item. This provides the effect of being able to obtain an appropriate determination result that takes into account the input explanatory text.

[0103] (Process flow) The processing flow performed by the information processing device 1A will be explained with reference to Figure 9. Figure 9 is a flowchart showing the processing flow performed by the information processing device 1A. The flowchart in Figure 9 includes each process of the support method according to this exemplary embodiment.

[0104] In S11 (acquisition process), the acquisition unit 101A acquires disclosure information indicating candidate items that may fall under the disclosure items that the applicant for the insurance contract must disclose at the time of the contract. For example, the acquisition unit 101A may acquire disclosure information that the applicant inputs to the information processing device 1A.

[0105] In S12 (extraction process), the extraction unit 102A uses an extraction model M1, which has been trained to take a document and data pair as input and output the portion of the document related to the data, to extract related disclosure items from the document containing the insurance disclosure items, which are disclosure items related to the candidate items shown in the disclosure information obtained in S11. Note that the document containing the insurance disclosure items may be input along with the disclosure information in S11, or it may be obtained from a predetermined database as shown in the example in Figure 4.

[0106] In S13, the determination unit 103A generates a prompt to be input to the language model M2. Specifically, the determination unit 103A generates a prompt that includes the relevant notification item extracted in S12 and the candidate item shown in the notification information acquired in S11, and instructs the M2 to infer the relationship between the relevant notification item and the candidate item, in other words, whether the candidate item corresponds to the relevant notification item.

[0107] In S14, the determination unit 103A inputs the prompt generated in S13 into the language model M2 to infer the relationship between the relevant notification items and the candidate items. In S15, the presentation control unit 104A presents the inference result from S14 to the applicant. In S16, the reception unit 105A determines whether there are any correction instructions for the inference result presented in S15. If the result in S16 is YES, the process proceeds to S17; if the result in S16 is NO, the process proceeds to S19.

[0108] In S15, the presentation control unit 104A may present the inference result by displaying a UI screen such as the example screen 503 in Figure 5 or the example screen 601 in Figure 6. In that case, the reception unit 105A may receive correction instructions via the UI screen. The process of presenting the inference result may be omitted. In that case, the process proceeds to S19 after S14.

[0109] In S17, the reception unit 105A records the contents of the received correction instruction in the storage unit 11A or an external database, etc. The timing of recording the contents of the correction instruction is arbitrary. For example, the reception unit 105A may record the contents of the correction instruction after the determination described later has been completed, or after the insurance application has been completed.

[0110] In S18, the determination unit 103A generates a prompt that reflects the content of the correction instruction received by the reception unit 105A, specifically a prompt that includes the correction instruction and instructs the system to infer the relationship between the candidate items and related notification items based on the correction instruction. After this, the process returns to S14, where the determination unit 103A inputs the newly generated prompt into the language model M2 to infer the relationship between the candidate items and related notification items.

[0111] In S19, the determination unit 103A determines, based on the inference result of the language model M2 (or the latest inference result among multiple inference results if multiple inferences were performed), which candidate items from the notification information obtained in S11 correspond to the related notification items extracted in S12.

[0112] In S20, the presentation control unit 104A presents the candidate items that were determined to be related to the notification items in S19 to the applicant as items that the applicant should notify. For example, the presentation control unit 104A may present the inference results by displaying a UI screen such as the example screen 701 in Figure 7.

[0113] In S21, the input support unit 106A determines whether or not the notification items have been finalized. For example, if the "Finalize" button in the example screen 701 of Figure 7 is pressed, the input support unit 106A may determine that all of the candidate items that were determined to be related notification items in S19 have been finalized as notification items.

[0114] If the result in S21 is YES, the process proceeds to S22. On the other hand, if the result in S21 is NO, the process returns to S11. In S11, which is the transitioned step from S21, the acquisition unit 101A acquires new notification information. For example, if a UI screen like the example screen 701 in Figure 7 is displayed in S20, the acquisition unit 101A only needs to acquire the items entered as unentered items via the UI screen as new notification information.

[0115] In S22, the input support unit 106A inputs each candidate item that was determined to be a disclosure item in S21 into the input form for the insurance that the applicant intends to contract. If there are no candidate items that are determined to be related disclosure items, the input support unit 106A inputs a message indicating that there are no disclosure items into the input form. This completes the process shown in Figure 10. Alternatively, in S22, the input support unit 106A may send the completed form to a designated recipient to proceed with the insurance contract.

[0116] Furthermore, as mentioned above, the inference results of the language model M2 are not always correct. For this reason, the input support unit 106A may also input the notification information obtained in S11 and the inference results in S14 into the input form as reference information. In addition, if the reasoning is output along with the inference results, the input support unit 106A may also input the reasoning as reference information. This allows the underwriter, who decides whether or not to underwrite the insurance contract, to take this reference information into consideration and make an appropriate decision on whether or not to underwrite the insurance contract.

[0117] Furthermore, if any of the disclosure items apply, the insurance premium may be higher than the applicant expected, or the insurance application may be refused. For this reason, if the presentation control unit 104A determines that there are candidate items that fall under the category of related disclosure items, it may present insurance products that do not require disclosure of those related disclosure items.

[0118] [Reference example 1] Figure 10 is a block diagram showing the configuration of the information processing device 1B according to this reference example. As shown in the figure, the information processing device 1B includes an acquisition unit 101B and a determination unit 103B.

[0119] The acquisition unit 101B acquires disclosure information that indicates candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, similar to the acquisition unit 101A in the exemplary embodiment 2.

[0120] Similar to the determination unit 103A in the exemplary embodiment 2, the determination unit 103B uses a language model trained on natural language to determine whether the candidate items shown in the notification information acquired by the acquisition unit 101B correspond to notification items.

[0121] Furthermore, the above-mentioned disclosure items may be extracted using an extraction model from a document containing disclosure items (i.e., related disclosure items), as in exemplary embodiment 1 or 2, or they may not be extracted using an extraction model. For example, the acquisition unit 101B may acquire, in addition to candidate items, disclosure items corresponding to those candidate items, in other words, disclosure items for which the determination of whether or not the candidate items apply, by having the applicant input them. Alternatively, for example, the determination unit 103B may extract a portion of a document containing insurance disclosure items (for example, a sentence or paragraph with a coherent content) by analyzing the document, and use that portion to perform the above determination.

[0122] As described above, the information processing device 1B includes an acquisition unit 101B that acquires disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and a determination unit 103B that uses a language model trained on natural language to determine whether or not the candidate items shown in the disclosure information acquired by the acquisition unit 101B fall under the disclosure items. With this information processing device 1B, an objective determination result can be obtained as to whether or not the candidate items fall under the disclosure items. Therefore, the effect is obtained that it is possible to support the disclosure of disclosure items in insurance contracts in an easy manner.

[0123] Furthermore, how the determination result of the determination unit 103B is used to facilitate notification is at the user's discretion. For example, the information processing device 1B may be equipped with a presentation control unit 104A similar to the information processing device 1A in the exemplary embodiment 2. In this case, the presentation control unit 104A can be made to present the determination result of the determination unit 103B, or candidate items that the determination unit 103B has determined to be subject to notification. This allows the applicant to make notification of the notification items by referring to the presented determination result or candidate items.

[0124] Furthermore, for example, the information processing device 1B may be provided with an input support unit 106A similar to that of the information processing device 1A in the exemplary embodiment 2. In this case, the input support unit 106A can be instructed to input candidate items that the determination unit 103B has determined to be subject to notification as notification items. As a result, the applicant can input notification items into a predetermined input form simply by inputting the candidate items into the information processing device 1B.

[0125] (Support Program) The functions of the information processing device 1B described above can also be implemented by a program. The support program in this reference example is an insurance contract application support program in which a computer functions as an acquisition means for acquiring disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and as a determination means for determining whether the candidate items shown in the disclosure information acquired by the acquisition means fall under the disclosure items using a language model that has been trained on natural language. This support program has the effect of making it easier to disclose disclosure items in an insurance contract.

[0126] (Support method) The support method described in this reference example is a method for supporting applications for insurance contracts, in which at least one processor performs an acquisition process to acquire disclosure information indicating candidate items that may fall under the disclosure items that the insurance contract applicant must disclose at the time of the contract, and a determination process to determine whether the candidate items shown in the disclosure information acquired in the acquisition process fall under the disclosure items using a language model that has been trained on natural language. This support method has the effect of making it easier to disclose disclosure items in insurance contracts.

[0127] [Reference example 2] In the exemplary embodiments described above, information processing devices 1 and 1A were described that obtain disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and extract related disclosure items related to the above candidate items from a document containing the insurance disclosure items.

[0128] Furthermore, the above-mentioned reference example describes an information processing device 1B that acquires disclosure information indicating potential items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and determines whether or not such potential items fall under the disclosure items.

[0129] These information processing devices 1, 1A, and 1B can be used not only to assist in the disclosure of disclosure matters in insurance contracts, but also to assist in the generation of arbitrary deliverables. For example, when preparing application documents for a subsidy, the applicant determines whether or not they meet the requirements for receiving the subsidy and prepares application documents that demonstrate that they meet those requirements.

[0130] In this process, the applicant can input potentially relevant information into Information Processing Device 1 or 1A instead of notification information, and have Information Processing Device 1 refer to the document containing the eligibility requirements instead of the document containing the notification information. This allows the device to extract the requirements from the document containing the eligibility requirements that are relevant to the information entered by the applicant. By having Information Processing Device 1 or 1A present the extracted requirements, the applicant can easily determine whether they meet the requirements and efficiently proceed with preparing the application.

[0131] Furthermore, applicants may input information that may qualify them for the subsidy into the information processing device 1B instead of the notification information, and may also have the information processing device 1B refer to the subsidy eligibility requirements. This allows the information processing device 1B to output a result indicating whether or not the information entered by the applicant meets the eligibility requirements. The applicant can then use the presented result as a reference to efficiently prepare their application.

[0132] In addition, for example, information processing devices 1, 1A, and 1B can also be used to create product manuals that conform to the specifications, or to create documents that comply with laws, regulations, and other provisions.

[0133] [Variation] The entities executing each process described in the above-described exemplary embodiments and reference examples are arbitrary and not limited to the examples given above. For example, a system having the same functions as information processing devices 1, 1A, and 1B can be constructed using multiple devices that can communicate with each other. Furthermore, the entities executing each process shown in the flowchart in Figure 9 may be a single device (which can also be called a processor) or multiple devices (which can also be called processors).

[0134] [Examples of implementation using software] Some or all of the functions of the information processing devices 1, 1A, and 1B (hereinafter also referred to as "the above devices") may be implemented by hardware such as integrated circuits (IC chips) or by software.

[0135] In the latter case, each of the above devices is implemented, for example, by a computer that executes instructions for a program, which is software that realizes each function. An example of such a computer (hereinafter referred to as computer C) is shown in Figure 11. Figure 11 is a block diagram showing the hardware configuration of computer C, which functions as each of the above devices.

[0136] Computer C comprises at least one processor C1 and at least one memory C2. Memory C2 stores a program (support program) P that causes computer C to operate as each of the above-mentioned devices. In computer C, processor C1 reads program P from memory C2 and executes it, thereby realizing each of the above-mentioned devices.

[0137] For processor C1, for example, a CPU (Central Processing Unit), GPU (Graphic Processing Unit), DSP (Digital Signal Processor), MPU (Micro Processing Unit), FPU (Floating Point Number Processing Unit), PPU (Physics Processing Unit), TPU (Tensor Processing Unit), quantum processor, microcontroller, or a combination thereof can be used. For memory C2, for example, flash memory, HDD (Hard Disk Drive), SSD (Solid State Drive), or a combination thereof can be used.

[0138] Computer C may also be equipped with RAM (Random Access Memory) for loading program P at runtime and for temporarily storing various data. Furthermore, computer C may be equipped with communication interfaces for sending and receiving data with other devices. Additionally, computer C may be equipped with input / output interfaces for connecting input / output devices such as keyboards, mice, displays, and printers.

[0139] Furthermore, program P can be recorded on a non-temporary, tangible recording medium M that is readable by computer C. Such a recording medium M could be, for example, tape, disk, card, semiconductor memory, or programmable logic circuitry. Computer C can acquire program P via such a recording medium M. Program P can also be transmitted via a transmission medium. Such a transmission medium could be, for example, a communication network or broadcast waves. Computer C can also acquire program P via such a transmission medium.

[0140] Furthermore, each of the above functions of each of the above devices may be implemented by a single processor in a single computer, by multiple processors in a single computer working together, or by multiple processors in each of multiple computers working together. In addition, the programs for implementing each of the above functions in each of the above devices may be stored in a single memory in a single computer, distributed and stored in multiple memories in a single computer, or distributed and stored in multiple memories in each of multiple computers.

[0141] [Additional Notes] This disclosure includes the technologies described in the following appendices. However, the present invention is not limited to the technologies described in the following appendices, and various modifications are possible within the scope of the claims.

[0142] (Note A1) An information processing device comprising: an acquisition means for acquiring disclosure information indicating candidate items that may fall under the disclosure items that an applicant for an insurance contract must disclose at the time of the contract; and an extraction means for extracting related disclosure items that are disclosure items related to the candidate items from a document containing the disclosure items for the insurance, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document that is related to the data.

[0143] (Appendix A2) The information processing apparatus according to Appendix A1, comprising: a determination means for determining whether the candidate items fall under the related notification items using a language model that has been trained on natural language; and a presentation control means for presenting the candidate items that the determination means has determined to fall under the related notification items as items that the applicant should notify.

[0144] (Note A3) The information processing device described in Appendix A2, wherein the determination means includes the related notification item and the candidate item, generates a prompt instructing the system to infer the relationship between the related notification item and the candidate item, and determines whether the candidate item corresponds to the related notification item based on the output obtained by inputting the generated prompt to the language model.

[0145] (Note A4) The information processing device according to Appendix A3, wherein the determination means generates a prompt instructing the determination means to output the basis for the inference along with the inference result of the relationship between the related notification item and the candidate item, and the presentation control means presents the inference result output by the language model or the determination result of the determination means along with the basis.

[0146] (Note A5) The information processing device described in Appendix A4, comprising a receiving means for receiving instructions for correction of the inference result, wherein the determination means re-determines whether the candidate item falls under the related notification item based on the correction instructions received by the receiving means.

[0147] (Note A6) The information processing device described in Appendix A5, wherein the receiving means receives a modification instruction expressing the modification content in natural language, the determination means generates a prompt that includes the modification instruction received by the receiving means and instructs the determination means to infer the relationship between the candidate item and the related notification item based on the modification instruction, and inputs the generated prompt into the language model and re-determines whether the candidate item corresponds to the related notification item based on the output obtained.

[0148] (Note A7) An information processing device as described in any of Appendix A1 to A6, comprising: a presentation control means for presenting the applicant's history information relating to the aforementioned notification item; and a reception means for receiving input of an explanatory text describing the presented history information, wherein the extraction means extracts related notification items related to the candidate item, with the set of the history information and the explanatory text being treated as one candidate item.

[0149] (Note A8) An information processing device comprising: an acquisition means for acquiring disclosure information indicating candidate items that may fall under the disclosure items that an applicant for an insurance contract must disclose at the time of the contract; and a determination means for determining whether or not the candidate items fall under the disclosure items using a language model that has been trained on natural language.

[0150] (Note B1) A support method comprising: an acquisition process in which at least one processor performs an acquisition process to acquire disclosure information indicating candidate items that may fall under disclosure items that an applicant for an insurance contract must disclose at the time of the contract; and an extraction process to extract related disclosure items that are disclosure items related to the candidate items from a document containing the insurance disclosure items, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document that is related to the data.

[0151] (Note B2) The support method according to Appendix B1, which includes a determination process in which at least one processor determines whether the candidate item falls under the related notification item using a language model that has been trained on natural language, and a presentation control process in which the at least one processor presents the candidate item that has been determined to fall under the related notification item in the determination process as an item that the applicant should notify.

[0152] (Note B3) The support method described in Appendix B2, wherein in the determination process, the at least one processor generates a prompt that includes the related notification item and the candidate item and instructs the processor to infer the relationship between the related notification item and the candidate item, and determines whether the candidate item corresponds to the related notification item based on the output obtained by inputting the generated prompt to the language model.

[0153] (Note B4) The support method according to Appendix B3, wherein in the determination process, the at least one processor generates a prompt instructing the output of the basis for the inference along with the inference result of the relationship between the related notification item and the candidate item, and in the presentation control process, the at least one processor presents the inference result output by the language model or the determination result of the determination process along with the basis.

[0154] (Note B5) The support method according to Appendix B4, wherein the at least one processor includes an acceptance process for receiving instructions for correction of the inference result, and a process for re-determining whether the candidate item falls under the related notification item based on the correction instructions received in the acceptance process.

[0155] (Note B6) The support method described in Appendix B5, wherein in the reception process, at least one processor receives a correction instruction expressing the content of the correction in natural language; in the re-determination process, at least one processor generates a prompt that includes the correction instruction received in the reception process and instructs the system to infer the relationship between the candidate item and the related notification item based on the correction instruction; and re-determines whether the candidate item corresponds to the related notification item based on the output obtained by inputting the generated prompt to the language model.

[0156] (Note B7) The support method described in any of Appendix B1 to B6, comprising: a presentation control process in which at least one processor presents the applicant's history information relating to the notification item; and a reception process in which at least one processor accepts input of an explanatory text describing the presented history information, wherein in the extraction process, the at least one processor extracts related notification items relating to the candidate item, with the pair of the history information and the explanatory text being one candidate item.

[0157] (Note B8) A support method in which at least one processor performs an acquisition process to acquire disclosure information indicating candidate items that may fall under disclosure items that an applicant for an insurance contract must disclose at the time of the contract, and a determination process to determine whether the candidate items indicated in the disclosure information acquired in the acquisition process fall under disclosure items using a language model that has been trained on natural language.

[0158] (Note C1) A support program that enables a computer to function as an acquisition means for acquiring disclosure information indicating candidate items that may fall under disclosure items that an insurance contract applicant must disclose at the time of the contract, and an extraction means for extracting related disclosure items, which are disclosure items related to the candidate items, from a document containing the insurance disclosure items, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document that is related to the data.

[0159] (Note C2) The support program described in Appendix C1 causes the computer to function as a determination means for determining whether the candidate items fall under the related notification items using a language model that has been trained on natural language, and a presentation control means for presenting the candidate items that the determination means has determined to fall under the related notification items as items that the applicant should notify.

[0160] (Note C3) The support program described in Appendix C2 includes the relevant notification item and the candidate item, generates a prompt instructing the system to infer the relationship between the relevant notification item and the candidate item, and determines whether the candidate item corresponds to the relevant notification item based on the output obtained by inputting the generated prompt to the language model.

[0161] (Note C4) The support program as described in Appendix C3, wherein the determination means generates a prompt instructing the output of the basis for the inference along with the inference result of the relationship between the related notification item and the candidate item, and the presentation control means presents the inference result output by the language model or the determination result of the determination means along with the basis.

[0162] (Note C5) The support program described in Appendix C4, wherein the computer functions as a receiving means for receiving instructions to correct the inference result, and the determination means re-determines whether the candidate item falls under the related notification item based on the correction instructions received by the receiving means.

[0163] (Appendix C6) The support program described in Appendix C5, wherein the receiving means receives a correction instruction expressing the content of the correction in natural language, the determination means generates a prompt that includes the correction instruction received by the receiving means and instructs the determination means to infer the relationship between the candidate item and the related notification item based on the correction instruction, and inputs the generated prompt into the language model and re-determines whether the candidate item corresponds to the related notification item based on the output obtained.

[0164] (Note C7) The computer functions as a presentation control means for presenting the applicant's history information relating to the notification items, and a reception means for receiving input of an explanatory text describing the presented history information, and the extraction means is a support program as described in any of Appendix C1 to C6 for extracting related notification items related to a candidate item, with the pair of the history information and the explanatory text being treated as one candidate item.

[0165] (Note C8) A support program that enables a computer to function as an acquisition means for acquiring disclosure information indicating potential items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and as a determination means for determining whether the candidate items shown in the disclosure information acquired by the acquisition means fall under the disclosure items using a language model that has been trained on natural language.

[0166] (Note D1) An information processing device comprising at least one processor, the at least one processor performing an acquisition process for acquiring disclosure information indicating candidate items that may fall under disclosure items that an applicant for an insurance contract must disclose at the time of the contract, and an extraction process for extracting related disclosure items that are disclosure items related to the candidate items from a document containing the insurance disclosure items, using an extraction model that has been trained to take a pair of documents and data as input and output the portion of the document related to the data.

[0167] The information processing device may also include memory. Furthermore, the memory may store a program that causes at least one processor to execute each of the aforementioned processes.

[0168] (Note D2) The information processing apparatus described in Appendix D1, wherein the at least one processor performs a determination process that determines whether the candidate item falls under the related notification item using a language model trained on natural language, and a presentation control process that presents the candidate item determined to fall under the related notification item in the determination process as an item that the applicant should notify.

[0169] (Note D3) The information processing apparatus according to Appendix D2, wherein in the determination process, the at least one processor generates a prompt that includes the related notification item and the candidate item and instructs the processor to infer the relationship between the related notification item and the candidate item, and determines whether the candidate item corresponds to the related notification item based on the output obtained by inputting the generated prompt to the language model.

[0170] (Note D4) The information processing apparatus according to Appendix D3, wherein in the determination process, the at least one processor generates a prompt instructing to output the basis for the inference along with the inference result of the relationship between the related notification item and the candidate item, and in the presentation control process, the at least one processor presents the inference result output by the language model or the determination result of the determination process along with the basis.

[0171] (Note D5) The information processing apparatus according to Appendix D4, wherein it performs an acceptance process to receive instructions for correction of the inference result, and in the determination process, at least one processor re-determines whether the candidate item falls under the related notification item based on the instructions for correction received in the acceptance process.

[0172] (Note D6) The information processing apparatus according to Appendix D5, wherein in the reception process, at least one processor receives a correction instruction expressing the content of the correction in natural language; in the determination process, at least one processor generates a prompt that includes the correction instruction received in the reception process and instructs the system to infer the relationship between the candidate item and the related notification item based on the correction instruction; and inputs the generated prompt to the language model and re-determines whether the candidate item corresponds to the related notification item based on the output obtained.

[0173] (Note D7) The information processing apparatus according to any one of the appendices D1 to D6, wherein the at least one processor performs a presentation control process for presenting the applicant's history information relating to the notification item, and a reception process for receiving input of an explanatory text describing the presented history information, and in the extraction process, the at least one processor extracts related notification items relating to the candidate item, with the pair of the history information and the explanatory text as one candidate item.

[0174] (Note D8) An information processing device comprising at least one processor, the at least one processor performing an acquisition process for acquiring disclosure information indicating candidate items that may fall under disclosure items that an applicant for an insurance contract must disclose at the time of the contract, and a determination process for determining whether the candidate items indicated in the disclosure information acquired in the acquisition process fall under disclosure items using a language model that has been trained on natural language.

[0175] (Note E1) A non-temporary recording medium that records a support program that causes a computer to perform an acquisition process to acquire disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and an extraction process that uses an extraction model trained on machine learning to take a pair of documents and data as input and output the portion of the document in which the data is relevant, to extract related disclosure items that are disclosure items related to the candidate items from a document containing the insurance disclosure items.

[0176] (Note E2) A non-temporary recording medium that records a support program that causes a computer to execute an acquisition process that obtains disclosure information indicating candidate items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and a determination process that determines whether or not the candidate items shown in the disclosure information obtained in the acquisition process fall under the disclosure items using a language model that has been trained on natural language. [Explanation of Symbols]

[0177] 1. Information Processing Device 101 Acquisition unit (acquisition means) 102 Extraction part (extraction means) 1A Information Processing Device 101A Acquisition part (acquisition means) 102A Extraction part (extraction means) 103A Judgment unit (judgment means) 104A Presentation Control Unit (Presentation Control Means) 105A Reception area (reception method) M1 Extraction Model M2 Language Model

Claims

1. A means of obtaining disclosure information that indicates potential items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, An information processing device comprising: an extraction means for extracting related disclosure items, which are disclosure items related to the candidate items, from a document containing the disclosure items for insurance, using an extraction model trained on machine learning to take a pair of documents and data as input and output the portion of the document related to the data in the document.

2. A determination means for determining whether the aforementioned candidate item falls under the aforementioned related notification item using a language model that has been trained on natural language, The information processing apparatus according to claim 1, further comprising: a presentation control means for presenting candidate items that the determination means has determined to fall under the related notification items as items that the applicant should notify.

3. The information processing apparatus according to claim 2, wherein the determination means includes the related notification item and the candidate item, generates a prompt instructing the system to infer the relationship between the related notification item and the candidate item, and determines whether the candidate item corresponds to the related notification item based on the output obtained by inputting the generated prompt to the language model.

4. The determination means generates a prompt that instructs the system to output the basis for the inference, along with the inference result regarding the relationship between the related notification item and the candidate item. The information processing apparatus according to claim 3, wherein the presentation control means presents the inference result output by the language model or the determination result of the determination means together with the basis.

5. The system includes a receiving means for receiving instructions to modify the aforementioned inference results, The information processing apparatus according to claim 4, wherein the determination means re-determines whether the candidate item falls under the related notification item based on the correction instruction received by the receiving means.

6. The aforementioned receiving means receives correction instructions that express the content of the correction in natural language, The information processing apparatus according to claim 5, wherein the determination means includes a correction instruction received by the receiving means, generates a prompt instructing the determination means to infer the relationship between the candidate item and the related notification item based on the correction instruction, and re-determines whether the candidate item corresponds to the related notification item based on the output obtained by inputting the generated prompt to the language model.

7. A presentation control means for presenting the applicant's history information regarding the aforementioned notification items, The system includes a receiving means for receiving input of an explanatory text describing the presented historical information, The information processing apparatus according to any one of claims 1 to 6, wherein the extraction means extracts related notification items associated with a candidate item, with the set of the historical information and the explanatory text being treated as one candidate item.

8. A means of obtaining disclosure information that indicates potential items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, An information processing device comprising: determination means for determining whether the aforementioned candidate items fall under the aforementioned notification items using a language model that has been trained on natural language.

9. At least one processor, An acquisition process to obtain disclosure information that indicates potential items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, A support method for performing an extraction process that extracts related disclosure items, which are disclosure items related to the candidate items, from a document containing the disclosure items for the insurance, using an extraction model trained on machine learning to take a pair of documents and data as input and output the portion of the document related to the data in the document.

10. Computers, A means for obtaining disclosure information that indicates potential items that may fall under the disclosure items that an insurance contract applicant must disclose at the time of the contract, and A support program that functions as an extraction means for extracting related disclosure items, which are disclosure items related to the candidate items, from a document containing the disclosure items for the insurance, using an extraction model trained on machine learning to take a pair of documents and data as input and output the portion of the document related to the data in question.

Images