Information processing device, information processing method, and program

Abstract

An information processing device 1 includes: an acquisition unit 131 that acquires document data, which is electronic data of a document to be processed; a summary data generation unit 132 that generates, from the document data acquired by the acquisition unit 131, summary data corresponding to a specific business-process viewpoint; an extraction unit 133 that extracts a feature amount from the summary data; and a determination unit 134 that compares the feature amount extracted by the extraction unit 133 with a prescribed feature amount corresponding to the specific business-process viewpoint, thereby determining whether the document data acquired by the acquisition unit 131 includes information necessary for performing the specific business-process.

Description

Information processing device, information processing method, and program

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

[0002] Techniques for simplifying the process of verifying evidence are known (for example, Patent Document 1).

[0003] JP 2023-29341 A

[0004] Companies perform a variety of administrative tasks and handle a wide range of data, so there was a problem that it was costly to prepare a system for each individual task to make it more efficient.

[0005] The present invention has been made in consideration of these points, and has as its object to make it possible to determine whether or not subsequent administrative processing can be carried out based on acquired data.

[0006] An information processing device of a first aspect of the present invention has an acquisition unit that acquires document data, which is electronic data of a document to be processed, an extraction unit that extracts features from summary data corresponding to a specific business process perspective generated based on the document data acquired by the acquisition unit, and a judgment unit that determines whether the document data acquired by the acquisition unit contains information necessary to process the specific business process by comparing the features extracted by the extraction unit with predetermined features corresponding to the specific business process perspective.

[0007] The apparatus may further include a summary data generation unit that generates summary data from the document data acquired by the acquisition unit according to the viewpoint of the specific administrative process, and the extraction unit may extract features from the summary data generated by the summary data generation unit.

[0008] The device may further have a memory unit that stores comparative feature information, which is features extracted from documents that have been acquired in the past and are the subject of specific administrative processing, and the determination unit may determine whether the document data acquired by the acquisition unit contains information necessary to process the specific administrative processing by comparing the features extracted from the summary data with the comparative feature information.

[0009] The device may further include an output unit that outputs the document data acquired by the acquisition unit so as to perform subsequent processing determined according to the viewpoint when the judgment unit determines that the document data acquired by the acquisition unit contains information necessary to process the specific business.

[0010] The device may further have a display control unit that displays candidate information to supplement the missing information when the determination unit determines that the document data acquired by the acquisition unit does not contain the information necessary to process the specific business.

[0011] The device may further have a display control unit that displays a message indicating that information is missing when the determination unit determines that the document data acquired by the acquisition unit does not contain the information necessary to process the specific business.

[0012] the determination unit determines whether or not the information included in the document data includes an abnormal value by comparing the feature extracted from the summary data with comparison feature information, and the information processing device further includes a display control unit that, when the determination unit determines that the information included in the document data includes an abnormal value, displays a message indicating that the information includes an abnormal value.

[0013] The summary data generation unit may generate the summary data including information necessary for the specific administrative processing that is different from the information contained in the document data and is generated based on the information contained in the document data.

[0014] An information processing method of a second aspect of the present invention includes the steps of: acquiring document data, which is electronic data of a specified document that is necessary for processing a specific task, executed by a computer; extracting features from summary data corresponding to the perspective of the specific task that is generated based on the document data acquired in the acquiring step; and determining whether the document data acquired in the acquiring step contains information necessary for processing the specific task by comparing the features extracted from the summary data with specified features corresponding to the perspective of the specific task.

[0015] In a third aspect of the program of the present invention, a computer is caused to execute the steps of acquiring document data, which is electronic data of a specified document that is a document necessary for processing a specific task; extracting features from summary data corresponding to the perspective of the specific task to be processed, which is generated based on the document data acquired in the acquiring step; and determining whether the document data acquired in the acquiring step contains information necessary for processing the specific task by comparing the features extracted from the summary data with specified features corresponding to the perspective of the specific task.

[0016] According to the present invention, it is possible to determine whether or not subsequent administrative processing can be performed based on the acquired data.

[0017] FIG. 1 is a diagram for explaining an overview of an information processing system S. FIG. 2 is a block diagram showing the configuration of an information processing device 1. FIG. 3 is a diagram showing an example of processing by a summary data generation unit 132. FIG. 4 is a diagram showing an example of the data structure of item information stored in a storage unit 12. FIG. 5 is a diagram showing an example of the data structure of an additional information table stored in a storage unit 12. FIG. 6 is a diagram showing an example of the data structure of complementary information stored in a storage unit 12. FIG. 7 is a flowchart showing the flow of processing in an information processing device 1.

[0018] 1 is a diagram illustrating an overview of an information processing system S. The information processing system S is a system for supporting business processing. The information processing system S includes an information processing device 1 and an information terminal 2.

[0019] The information processing device 1 determines whether the acquired document data includes information necessary for business processing, and performs processing according to the determination result. The information processing device 1 is, for example, a server.

[0020] The information terminal 2 is a terminal used by a user of the information processing system S. The information terminal 2 is, for example, a personal computer, a smartphone, or a tablet. The user of the information processing system S is a person in charge of accounting or other operations at a business.

[0021] The processing in the information processing system S will be described. The information processing device 1 acquires document data ((1) in FIG. 1). The document data is electronic data of documents created and received in business and targeted for processing in specific business processes. Examples of document data include invoices, contracts, delivery notes, receipts, etc. The document data includes information that is referenced or processed in specific business processes. The information processing device 1 may acquire the document data from the information terminal 2, or may acquire the document data from an external device (not shown).

[0022] The information processing device 1 generates summary data based on document data ((2) in FIG. 1). As will be described in detail later, the information processing device 1 generates the summary data by inputting the document data into large language models (LLMs). The summary data is data generated by extracting information necessary for a specific business process from the information contained in the document data.

[0023] The information processing device 1 extracts features from the generated summary data ((3) in FIG. 1). The features are information obtained by converting the information contained in the summary data into a format suitable for computer calculation. The features are, for example, feature vectors that indicate the characteristics of the information contained in the summary data. Based on the extracted features, the information processing device 1 determines whether the document data contains information necessary for specific administrative processing ((4) in FIG. 1).

[0024] The information processing device 1 outputs information based on the determination result ((5) in FIG. 1). If it is determined that the document data contains information necessary for specific administrative processing, the information processing device 1 outputs the document data so that subsequent processing can be performed. As an example, the information processing device 1 outputs the document data to an external device that executes subsequent processing. If it is determined that the document data does not contain information necessary for specific administrative processing, the information processing device 1 displays, for example, a message on the information terminal 2 indicating that the document data does not contain the necessary information.

[0025] By configuring the information processing system S in this way, it is possible to determine whether or not subsequent administrative processing can be executed based on the acquired data.

[0026] 2 is a block diagram showing the configuration of the information processing device 1. The information processing device 1 has a communication unit 11, a storage unit 12, and a control unit 13. The control unit 13 has an acquisition unit 131, a summary data generation unit 132, an extraction unit 133, a determination unit 134, an output unit 135, and a display control unit 136.

[0027] The communication unit 11 is a communication interface for transmitting and receiving data to and from other devices via a network. The storage unit 12 is a storage medium including a read-only memory (ROM), a random access memory (RAM), a solid-state drive (SSD), a hard disk drive, etc. The storage unit 12 stores in advance a program to be executed by the control unit 13.

[0028] The control unit 13 is a processor such as a CPU (Central Processing Unit), and functions as an acquisition unit 131, a summary data generation unit 132, an extraction unit 133, a determination unit 134, an output unit 135, and a display control unit 136 by executing a program stored in the storage unit 12.

[0029] The process of generating summary data by summary data generation unit 132 will be described with reference to Fig. 3. Acquisition unit 131 acquires document data D1, which is electronic data of a document to be processed. In Fig. 3, document data D1 is an invoice.

[0030] The extraction unit 133 extracts features from summary data corresponding to a specific business process perspective generated based on the document data acquired by the acquisition unit 131. More specifically, the summary data generation unit 132 generates summary data corresponding to the specific business process perspective from the document data acquired by the acquisition unit 131. As an example, the storage unit 12 stores item information associating a specific business process with items required for the specific business process. FIG. 4 is a diagram illustrating an example of the data structure of the item information. The item information includes "description items" and "extraction targets." The description items indicate information that may be included in the input document data. The "extraction targets" indicate whether the description items are targets to be extracted in the summary data. The "extraction targets" are associated with each business process perspective, indicating whether they are targets for extraction. In the example illustrated in FIG. 4, whether each description item is a target for extraction is associated with each business process perspective, such as "tax calculation," "validity confirmation," "journalization," "invoice," and "payment," as business processes processed by the information processing device 1. In the example shown in FIG. 4, if "Extraction target" is "◯", it indicates that the description item is an item to be extracted from the summary data.

[0031] 3, the summary data includes information contained in the document data that is necessary for the specific business process to be processed. The summary data generation unit 132 generates the summary data by inputting the document data acquired by the acquisition unit 131 and instructions (prompts) for extracting predetermined items into a known large-scale language model.

[0032] The transaction for which summary data is to be generated (hereinafter referred to as the "target transaction") may be predetermined, or the reception unit (not shown) may receive information indicating the target transaction from the information terminal 2. The summary data generation unit 132 identifies the target transaction based on the predetermined transaction or information received from the information terminal 2. The summary data generation unit 132 references the item information to identify items to be extracted in the identified target transaction. The summary data generation unit 132 inputs the document data acquired by the acquisition unit 131 and instructions for extracting the identified items to be extracted into a large-scale language model to generate summary data. In the example shown in FIG. 3, if the target transaction is "tax," the summary data generation unit 132 generates summary data D2 by extracting items associated with "tax" in the item information. If the target transaction is "payment," the summary data generation unit 132 generates summary data D3 by extracting items associated with "payment" in the item information.

[0033] The extraction unit 133 extracts feature quantities from the summary data. The extraction unit 133 may store the extracted feature quantities in the storage unit 12 in association with the target business process.

[0034] The determination unit 134 compares the feature quantities extracted by the extraction unit 133 with predetermined feature quantities corresponding to the viewpoint of the specific business process to determine whether the document data acquired by the acquisition unit 131 contains information necessary for processing the specific business process. For example, the determination unit 134 compares the extracted feature quantities with the predetermined feature quantities. The predetermined feature quantities may be feature quantities extracted from previously generated summary data (hereinafter referred to as "comparison feature quantities"), or feature quantities extracted from document data with or without anomalies. For example, the determination unit 134 calculates the distance or similarity between the extracted feature quantities and the predetermined feature quantities, and determines whether the document data contains information necessary for processing the specific business process based on the calculated distance or similarity. If the calculated distance or similarity is equal to or less than a predetermined threshold, the determination unit 134 determines that the document data contains information necessary for processing the specific business process. In addition, when comparing the features extracted by the extraction unit 133 with the features extracted from document data containing an abnormality, the judgment unit 134 judges that the document data does not contain the information necessary to process a specific task if the calculated distance or similarity is below a predetermined threshold.

[0035] As an example, the memory unit 12 stores comparative feature information. The comparative feature information is a feature extracted from a document that is the subject of a specific administrative process and that has been previously acquired. More specifically, the comparative feature information is a feature extracted from summary data generated based on previously acquired document data. As an example, the memory unit 12 stores the feature extracted by the extraction unit 133 in association with the administrative process that is the subject of the feature. The determination unit 134 compares the feature extracted from the summary data with the comparative feature information to determine whether the document data acquired by the acquisition unit 131 contains information necessary to process the specific administrative process.

[0036] By configuring the information processing device 1 in this way, it is possible to determine whether or not subsequent administrative processing can be executed based on the acquired data.

[0037] The information processing device 1 may be configured to generate summary data that further includes information derived from information included in the document data.

[0038] Summary data generation unit 132 generates summary data that further includes additional information by inputting the document data into the large-scale language model. The additional information is information different from the information included in the document data acquired by acquisition unit 131 and is information necessary for specific administrative processing that is generated based on the information included in the document data. In this case, summary data generation unit 132 further inputs an instruction to generate the additional information into the large-scale language model.

[0039] As an example, the storage unit 12 stores an additional information table. FIG. 5 is a diagram showing an example of the data structure of the additional information table stored in the storage unit 12. In the additional information table shown in FIG. 5, "additional information," "target business," and "generation method" are associated with each other. "Additional information" in FIG. 5 indicates the additional information to be generated. "Target business" indicates the target business for which the additional information is to be generated. "Generation method" indicates the information referenced to generate the additional information and the generation process. If the target business process matches the business process included in the additional information table, the summary data generation unit 132 generates additional information in the additional information table corresponding to the business process that matches the target business process, based on information included in the document data acquired by the acquisition unit 131. The summary data generation unit 132 generates summary data that further includes the generated additional information.

[0040] An example of the additional information is a monthly cost allocation for the invoice amount or subtotal amount, calculated based on the invoice amount or subtotal amount included in the invoice and the invoice period. For example, if the period included in the document data that is an invoice is from April 25 to May 24 and the invoice amount is 10,000 yen, the summary data generation unit 132 generates additional information indicating that the allocation amounts for April and May are 2,000 yen and 8,000 yen, respectively.

[0041] In another example, the additional information is the start and end dates of the contract, which are derived from the details of the contract. For example, if the document data that is a contract includes information indicating the contract date as May 1, 2023 and the contract period as one year, but does not include information indicating the start time of the contract, the summary data generation unit 132 generates additional information that sets the start and end dates of the contract as May 1, 2023 and April 30, 2024, respectively.

[0042] In another example, the additional information is the amount of withholding tax included in the invoice. When the details included in the document data include items subject to withholding tax and items not subject to tax, the summary data generation unit 132 calculates the product of the total amount of the invoice for the items subject to withholding tax and the withholding tax rate, and generates the calculated product as the additional information.

[0043] By configuring the information processing device 1 in this manner, for example, if the acquired document data is missing some of the information necessary for administrative processing, the missing information can be supplemented, allowing subsequent administrative processing to proceed smoothly.

[0044] The information processing device 1 may be configured to execute subsequent administrative processing based on the determination result of the determination unit 134.

[0045] If the determination unit 134 determines that the document data acquired by the acquisition unit 131 contains information necessary to process a specific transaction, the output unit 135 outputs the document data acquired by the acquisition unit 131 to execute subsequent processing determined according to the viewpoint. For example, the memory unit 12 stores information associating a target transaction with an output destination of the document data for the target transaction. If the determination unit 134 determines that the document data acquired by the acquisition unit 131 contains information necessary to process a specific transaction, the output unit 135 outputs the document data to an output destination corresponding to the target transaction. For example, if the target transaction is "journalization," the output unit 135 outputs the document data to an accounting processing device (not shown) that executes the journalization. The information processing device 1 may have a functional unit that executes subsequent processing such as journalization and payment processing, and the output unit 135 may output the document data to a functional unit corresponding to the target transaction. Furthermore, the output unit 135 may output summary data generated by the summary data generation unit 132 in addition to or instead of the document data.

[0046] By configuring the information processing device 1 in this way, it is possible to automate subsequent processing when there is no problem with the document data.

[0047] [Complementary Processing] When acquired document data contains insufficient information, the information processing device 1 may be configured to complement the insufficient information based on other information.

[0048] When the determination unit 134 determines that the document data acquired by the acquisition unit 131 does not contain information necessary to process a specific task, the display control unit 136 displays candidate information for supplementing the missing information. As an example, when the determination unit 134 determines that the document data acquired by the acquisition unit 131 does not contain information necessary to process a specific task, the determination unit 134 identifies the missing items. For example, the determination unit 134 compares the feature extracted by the extraction unit 133 with the comparison feature, and identifies, as the missing item, an item corresponding to a dimension whose value is missing only in the feature extracted by the extraction unit 133.

[0049] The display control unit 136 acquires data for completing the missing items identified by the determination unit 134. As an example, the storage unit 12 stores completion information. FIG. 6 is a diagram showing an example of the data structure of the completion information stored in the storage unit 12. In the completion information, a "complementary item," "reference data," and a "complementation method" are associated with each other. A "complementary item" indicates the item to be completed. "Reference data" indicates data to be referenced to complete the "complementary item." A completion method indicates a process for generating information on the "complementary item."

[0050] The display control unit 136 refers to the supplementary information stored in the memory unit 12 and identifies data to be referenced to supplement missing information in the document data. The display control unit 136 refers to the identified data and displays it on the information terminal 2 as candidate information to supplement the acquired data. When an operation to supplement the acquired data with the displayed information is performed on the information terminal 2, the output unit 135 outputs the summary information generated by the summary data generation unit 132 and the information acquired by the display control unit 136 as supplementary information to an output destination corresponding to the target business process.

[0051] The complementation process will be explained using an example in which the acquired invoice data does not include the payee's account number. As an example, the complementation information includes a requirement that if the "payee account number" is not included in the "payment transaction," the "payee master" stored in the memory unit 12 should be referenced. The "payee master" is information that associates a payee with the payee's account number. When the determination unit 134 determines that the payee account number is missing from the document data, the display control unit 136 acquires the payee account number in the payee master that corresponds to the payee's name and address included in the document data. The display control unit 136 displays this on the information terminal 2 as a candidate for information to complement the acquired payee account number.

[0052] In addition, the memory unit 12 may store a table that associates the target administrative process with candidate information for supplementing the process when the judgment unit 134 judges that the necessary information is not included based on the summary data for the target administrative process, and the display control unit 136 may refer to the table and display candidate information for supplementing the missing information.

[0053] By configuring the information processing device 1 in this way, it is possible to achieve the effect of enabling the execution of administrative processing even when the acquired document data contains insufficient information.

[0054] The information processing device 1 may be configured to display a message when the document data lacks information necessary for paperwork processing.

[0055] When the determination unit 134 determines that the document data acquired by the acquisition unit 131 does not contain information necessary to process a specific task, the display control unit 136 displays a message indicating that the information is missing. The display control unit 136 may display the message indicating that the information is missing on the information terminal 2, or may output the message to be displayed on an external device (not shown). The display control unit 136 may further display the content of the missing information together with the message indicating that the information is missing.

[0056] The determination unit 134 may compare the feature extracted from the summary data with the comparison feature information to determine whether the information included in the document data contains an abnormal value. For example, in document data that is an invoice, if the feature corresponding to the invoice amount differs from the feature corresponding to the invoice amount in the comparison feature by a predetermined threshold or more, the determination unit 134 determines that the document data contains an abnormal value. When the determination unit 134 determines that the document data contains an abnormal value, the display control unit 136 displays a predetermined message on an external device (not shown). For example, when the determination unit 134 determines that the document data contains an abnormal value, the display control unit 136 displays a message indicating that the acquired document data contains an abnormal value.

[0057] By configuring the information processing device 1 in this way, it is possible to provide information that allows a user to determine whether or not the subsequent process can be executed.

[0058] [Processing Flow in Information Processing Device 1] Fig. 7 is a flowchart showing the processing flow in the information processing device 1. The flowchart in Fig. 7 starts from the point at which document data is acquired.

[0059] The acquisition unit 131 acquires document data (S01). The summary data generation unit 132 identifies the target business process (S02). As an example, the summary data generation unit 132 identifies the target business process based on information received from the information terminal 2. The summary data generation unit 132 generates summary data based on the document data and the target business process (S03). The extraction unit 133 extracts features from the generated summary data (S04).

[0060] The determination unit 134 determines whether or not the summary data contains information necessary for the administrative processing (S05) based on the feature extracted by the extraction unit 133. If the determination unit 134 determines that the summary data contains information necessary for the administrative processing (YES in S05), the output unit 135 outputs the document data to an output destination corresponding to the target administrative processing (S06), and the information processing device 1 ends the processing.

[0061] If the determination unit 134 determines that the summary data does not contain information necessary for the administrative process (NO in S05), the display control unit 136 determines whether or not the missing items in the summary data can be supplemented (S07). For example, the display control unit 136 determines that the missing items in the document data can be supplemented if they match the items included in the supplemental information stored in the storage unit 12.

[0062] If the missing items in the summary data can be supplemented (YES in S07), the display control unit 136 displays candidates for supplementary information on the information terminal 2 (S08). If an operation indicating that the displayed information will be used for supplementation is performed on the information terminal 2, the information processing device 1 proceeds to S06.

[0063] If the missing items cannot be supplemented in the summary data (NO in S07), the display control unit 136 outputs a message indicating that information is missing (S09), and the information processing device 1 ends the process.

[0064] [Effects of the Present Embodiment] As described above, the present embodiment has the effect of being able to determine whether or not subsequent administrative processing can be executed based on the acquired data.

[0065] The present invention has been described above using embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist of the present invention. For example, all or part of the device can be configured by functionally or physically distributing or integrating in any unit. Furthermore, new embodiments resulting from any combination of multiple embodiments are also included in the embodiments of the present invention. The effects of the new embodiments resulting from the combination also have the effects of the original embodiments.

[0066] REFERENCE SIGNS LIST 1 Information processing device 2 Information terminal 11 Communication unit 12 Storage unit 13 Control unit 131 Acquisition unit 132 Summary data generation unit 133 Extraction unit 134 Determination unit 135 Output unit 136 Display control unit

Claims

1. An information processing device having: an acquisition unit that acquires document data, which is electronic data of a document to be processed; an extraction unit that extracts features from summary data corresponding to a specific business process perspective generated based on the document data acquired by the acquisition unit; and a judgment unit that judges whether the document data acquired by the acquisition unit contains information necessary to process the specific business process by comparing the features extracted by the extraction unit with predetermined features corresponding to the specific business process perspective.

2. The information processing device according to claim 1, further comprising a summary data generation unit that generates summary data from the document data acquired by the acquisition unit according to the viewpoint of the specific business process, and wherein the extraction unit extracts features from the summary data generated by the summary data generation unit.

3. An information processing device as described in claim 1, further comprising a memory unit that stores comparative feature information, which is features extracted from a document that was previously acquired and is the subject of a specific administrative process, and the determination unit determines whether the document data acquired by the acquisition unit contains information necessary to process the specific administrative process by comparing the features extracted from the summary data with the comparative feature information.

4. The information processing device of claim 1, further comprising an output unit which outputs the document data acquired by the acquisition unit so as to execute subsequent processing determined according to the viewpoint when the determination unit determines that the document data acquired by the acquisition unit contains information necessary to process the specific task.

5. The information processing device of claim 1, further comprising a display control unit that displays candidates for information that supplement the missing information when the determination unit determines that the document data acquired by the acquisition unit does not contain information necessary to process the specific task.

6. The information processing device of claim 1, further comprising a display control unit that displays a message indicating that information is missing when the determination unit determines that the document data acquired by the acquisition unit does not contain information necessary to process the specific transaction.

7. The information processing device of claim 1, wherein the determination unit determines whether the information contained in the document data includes an abnormal value by comparing the features extracted from the summary data with comparison feature information, and the information processing device further has a display control unit that, when the determination unit determines that the information contained in the document data includes an abnormal value, displays a message indicating that an abnormal value is included.

8. The information processing device according to claim 2, wherein the summary data generation unit generates the summary data including information necessary for the specific administrative process that is different from information contained in the document data and is generated based on information contained in the document data.

9. An information processing method executed by a computer, comprising: a step of acquiring document data, which is electronic data of a specified document that is a document necessary for processing a specific task; a step of extracting features from summary data corresponding to the viewpoint of the specific task that is generated based on the document data acquired in the acquiring step; and a step of determining whether or not the document data acquired in the acquiring step contains information necessary for processing the specific task by comparing the features extracted from the summary data with predetermined features corresponding to the viewpoint of the specific task.

10. A program that causes a computer to execute the following steps: acquiring document data, which is electronic data of a specified document that is a document necessary for processing a specific task; extracting features from summary data corresponding to the viewpoint of the specific task that is generated based on the document data acquired in the acquiring step; and determining whether or not the document data acquired in the acquiring step contains information necessary for processing the specific task by comparing the features extracted from the summary data with the specified features corresponding to the viewpoint of the specific task.

Images