Document processing system, document processing device, document processing method, and program

Abstract

The present disclosure provides a document processing system, a document processing device, a document processing method, and a program capable of determining an inclusion relationship of requirements to be compared among a plurality of documents. A requirement reading means reads a first requirement included in a first document and a second requirement included in a second document. An instruction means outputs instructions for determining an inclusion relationship between the first requirement and the second requirement. A determination means determines the inclusion relationship between the first requirement and the second requirement according to the instructions. An output means outputs the result of the determination by the determination means.

Description

Document processing system, document processing device, document processing method and program

[0001] The present disclosure relates to a document processing system, a document processing device, a document processing method, and a program.

[0002] When introducing an information system or IT (Information Technology) service, the client submits a Request for Proposal (RFP) to the vendor to whom the order will be placed. The vendor then submits a proposal based on the RFP to the client in order to make a specific proposal to the client.

[0003] For example, the client may own a package product for configuring an information system. In this case, the client must determine whether the package product's functions can satisfy the requirements set out in the RFP. Therefore, the client must compare the RFP with the package product's specifications and other documents to determine whether the package product's functions can satisfy the requirements set out in the RFP.

[0004] One method for comparing two documents, such as an RFP and a package product specification, is for a person to visually inspect the documents. However, if the two documents each contain a large amount of information, or if the information is complex and there are overlaps, this requires a lot of manpower. Furthermore, manual comparison makes it difficult to prevent mistakes, such as oversights.

[0005] In response to this, for example, Patent Document 1 proposes a document verification support device that supports verification of correspondence between contents between documents whose contents correspond to each other. This document verification support device includes a granularity determination unit, a division unit, and an extraction unit. The granularity determination unit determines the division granularity for dividing the reference document and the verification target document for each reference document and the verification target document based on the characteristics of the reference document and the verification target document. The division unit divides the reference document and the verification target document into one or more elements using the determined division granularity based on language specification information. The extraction unit extracts elements that correspond to each other between the divided reference document and the verification target document based on predetermined extraction criteria. The extraction unit then outputs the extracted results as verification support information.

[0006] JP 2017-111742 A

[0007] However, a method of simply comparing correspondences between a plurality of documents may not be sufficient to determine the consistency of corresponding written content between the plurality of documents.

[0008] A document processing system according to one embodiment includes a requirement reading means for reading a first requirement contained in a first document and a second requirement contained in a second document, an instruction means for outputting an instruction for determining the inclusion relationship between the first requirement and the second requirement, a determination means for determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction, and an output means for outputting the determination result by the determination means.

[0009] A document processing device according to one embodiment includes a requirement reading means for reading a first requirement contained in a first document and a second requirement contained in a second document, an instruction means for outputting an instruction for determining the inclusion relationship between the first requirement and the second requirement to a determination means for determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction, and an output means for outputting the determination result by the determination means.

[0010] A document processing method according to one embodiment reads a first requirement contained in a first document and a second requirement contained in a second document, outputs an instruction to determine the inclusion relationship between the first requirement and the second requirement, determines the inclusion relationship between the first requirement and the second requirement in accordance with the instruction, and outputs the result of the determination.

[0011] A program according to one embodiment causes a computer to perform the following processes: reading a first requirement contained in a first document and a second requirement contained in a second document; outputting an instruction to determine the inclusion relationship between the first requirement and the second requirement; determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction; and outputting the result of the determination.

[0012] According to the present disclosure, it is possible to provide a document processing system, a document processing device, a document processing method, and a program that can determine the inclusion relationship of requirements to be compared between multiple documents.

[0013] FIG. 1 is a block diagram schematically showing the configuration of a document processing device according to an embodiment; FIG. 2 is a block diagram schematically showing the configuration of a document processing system according to an embodiment; FIG. 3 is a flowchart of document processing operations of the document processing system according to an embodiment; FIG. 4 is a diagram showing an example of a requirement pair; FIG. 5 is a block diagram schematically showing a modified example of a document processing device according to an embodiment; FIG. 6 is a flowchart of operations of a preprocessing device according to an embodiment; FIG. 7 is a block diagram schematically showing the configuration of a document processing system according to an embodiment;

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same elements are designated by the same reference numerals, and redundant explanations will be omitted as necessary.

[0015] When referring to one embodiment below, it means that the present invention can be applied to any one of the embodiments described below or a combination of two or more embodiments, and is not limited to a specific embodiment.

[0016] First Embodiment Generally, when a government agency, a company, or the like introduces an information system or an IT service, the government agency or company procures the information system or the IT service from an IT vendor, etc. When considering the introduction of an information system or an IT service, the government agency or company that is the ordering party obtains information and specific proposals necessary for selecting a supplier from the IT vendor, etc.

[0017] A Request for Information (RFI) is a document that a client submits to an IT vendor to request information necessary for selecting a supplier. An RFI is a document that requests basic information, technical information, product information, etc. from potential IT vendors when the client is selecting products or services, planning outsourcing, bidding, and procurement.

[0018] In addition, a request for proposal (RFP) is a document in which a client presents requirements for an information system or IT service and requests specific proposals from IT vendors. The RFP describes specific requirements such as the scope of the proposal, required functions, and constraints, including, for example, an outline, objectives, required functions, service levels, and contract terms.

[0019] Generally, an RFI is created by a client to gather information about a product or service. Based on the information collected through the PFI, the client creates an RFP to request specific proposals from IT vendors. The client selects a supplier based on the proposal prepared by the IT vendor, which is a response to the RFP.

[0020] When an IT vendor prepares a proposal in response to an RFP, the IT vendor must consider whether the system or service it can provide satisfies the requirements described in the RFP. Therefore, the IT vendor compares the RFP with a reference document that describes the functional requirements that can be realized by the vendor's system or service, and determines whether the functional requirements in the reference document satisfy the requirements described in the RFP.

[0021] In this embodiment, a document processing system and device that analyzes the inclusion relationships of the description contents of multiple documents will be described. FIG. 1 is a block diagram schematically illustrating the configuration of a document processing device according to one embodiment. The document processing device 100 is a device that constitutes a document processing system together with an external determination unit. The document processing device 100 has a requirement reading unit 11, an instruction unit 12, and a determination result output unit 13. The determination result output unit is also referred to as output means or output unit.

[0022] 2 is a block diagram showing a configuration of a document processing system according to an embodiment. FIG. 3 is a flowchart showing a document processing operation of the document processing system according to an embodiment. The document processing system 1000 includes a document processing device 100 and a determination processing device 10.

[0023] As shown in step S1 of FIG. 3, the requirements reading unit 11 reads a requirements pair P consisting of a reference requirement R1 of a reference document D1 and a target requirement R2 of a verification target document D2, which is a comparison target for the reference requirement R1. FIG. 4 is a diagram showing an example of a requirements pair. The requirements pair P is created in advance by performing a predetermined process on the reference document D1 and the verification target document D2 to associate the reference requirement R1 with the target requirement R2. The requirements reading unit 11 outputs the read requirements pair P to the instruction unit 12. The reference document D1 is also referred to as the first document. The verification target document D2 is also referred to as the second document. The reference requirement R1 is also referred to as the first requirement. The target requirement R2 is also referred to as the second requirement.

[0024] In this embodiment, the reference document D1 is, for example, a document such as a design document or specification that specifies functional requirements that can be implemented by a package product for building an information system provided by a supplier or potential supplier such as a vendor. The verification target document D2 is, for example, a document such as an RFP or RFI that the supplier presents to the supplier or potential supplier. Hereinafter, the supplier that holds the reference document D1 will be referred to as the user of the document processing system 1000 and the document processing device 100, or simply as the user. The supplier that presents the verification target document D2 to the user, the supplier or potential supplier, will be referred to as the customer.

[0025] The requirement reading unit 11 may read the requirement pair P in response to a user operation. Alternatively, the requirement reading unit 11 may read the requirement pair P in accordance with a predetermined procedure and timing.

[0026] The requirement reading unit 11 may read each requirement pair P one by one. Alternatively, the requirement reading unit 11 may read a plurality of requirement pairs P. In this case, the requirement reading unit 11 may select one requirement pair P from the plurality of read requirement pairs P and sequentially output the selected one to the instruction unit 12.

[0027] The requirement pair P may be stored in advance in an external storage means. In this case, the requirement reading unit 11 may read the requirement pair P from the external storage means as appropriate. FIG. 5 is a block diagram schematically showing a modified example of the document processing device according to an embodiment. As shown in FIG. 5, the document processing device 100 may be provided with a storage unit 14 that stores one or more requirement pairs P. The requirement reading unit 11 may read the requirement pair P from the storage unit 14.

[0028] The requirement pair P may be created in advance by performing a predetermined process on the reference document D1 and the verification target document D2. As shown in Fig. 5, the document processing system 1000 may be provided with a preprocessing device 120 that creates the requirement pair P. The preprocessing device 120 includes a document reading unit 121, a document dividing unit 122, a correspondence relationship analysis unit 123, and a requirement output unit 124.

[0029] The operation of the preprocessing device 120 will now be described with reference to Fig. 6. Fig. 6 is a flowchart showing the operation of the preprocessing device according to one embodiment.

[0030] 6, the document reading unit 121 reads the reference document D1 and the verification target document D2. For example, the document reading unit 121 may read the reference document D1 and the verification target document D2 provided via an input operation by a user.

[0031] As shown in step S12 of Fig. 6, the document divider 122 divides the reference document D1 and the verification target document D2 into multiple elements based on a predetermined rule. Here, the data obtained by dividing the reference document D1 is referred to as reference document d1. The data obtained by dividing the verification target document D2 is referred to as verification target document d2. Here, the document divider 122 may use various methods, such as morphological analysis, to divide the documents.

[0032] 6, the correspondence analysis unit 123 analyzes the multiple elements of the divided reference document d1 and the multiple elements of the divided verification target document d2. As a result, the correspondence analysis unit 123 detects correspondences between one or more reference requirements R3 included in the multiple elements of the reference document d1 and one or more target requirements R4 included in the multiple elements of the verification target document d2. The correspondence analysis unit 123 then generates correspondence information INF indicating the correspondences between each of the reference requirements R3 and each of the target requirements R4.

[0033] The processes in the document reading unit 121, the document segmentation unit 122, and the correspondence analysis unit 123 may apply any method for analyzing the correspondence between two documents, such as various document processing methods such as morphological analysis, or a large-scale language model trained using a large amount of text. Furthermore, the natural language processing model may be a natural language processing model constructed by machine learning that is capable of estimating the similarity between two documents.

[0034] 6, the requirements output unit 124 generates a requirements pair P consisting of a reference requirement R1 and a target requirement R2 by combining one reference requirement R3 and one target requirement R4 that correspond to each other based on the correspondence information INF. The requirements output unit 124 may output the generated requirements pair P to the requirements reading unit 11 or the storage unit 14 of the document processing device 100 as appropriate. FIG. 5 illustrates an example of a configuration in which the requirements output unit 124 outputs the requirements pair P to the storage unit 14.

[0035] 3, the instruction unit 12 provides the determination processing device 10 with a prompt PRM instructing the determination processing device 10 to determine the relationship between the reference requirement R1 and the target requirement R2. Here, in this embodiment, the instruction unit 12 instructs the determination processing device 10 by the prompt PRM to determine the inclusion relationship between the reference requirement R1 and the target requirement R2.

[0036] As shown in step S3 of FIG. 3 , the determination processing device 10 executes processing instructed by the prompt PRM provided by the instruction unit 12. The determination processing device 10 determines the inclusion relationship between the reference requirement R1 and the target requirement R2, for example, by inputting the prompt PRM into a large-scale language model. The large-scale language model is not limited to a specific model, and may be any of various other large-scale language models. In this embodiment, the determination processing device 10 determines the inclusion relationship between the reference requirement R1 and the target requirement R2 by distinguishing between cases.

[0037] Here, the instruction unit 12 instructs the judgment processing device 10 by a prompt PRM to judge the inclusion relationship between the reference requirement R1 and the target requirement R2 by dividing them into cases A to C. Case A is a case where the reference requirement R1 includes the target requirement R2. Case B is a case where the target requirement R2 includes the reference requirement R1. Case C is a case where the reference requirement R1 and the target requirement R2 are not in an inclusion relationship. Furthermore, for each of cases A to C, more detailed case divisions are made depending on the analysis results. Cases A to C are also referred to as the first to third cases, respectively.

[0038] Below, examples of the prompt PRM that the instruction unit 12 gives to the determination processing device 10 are shown. [Example prompt] "Please determine whether or not the reference requirement R1 in your company's system function description satisfies the target requirement R2, which is a customer's request for system functions, and explain why. When making your decision, please consider which of the following patterns applies as a result of comparing the reference requirement R1 and the target requirement R2. Case A: Reference requirement R1 encompasses the target requirement R2 A1: Reference requirement R1 and target requirement R2 are consistent A2: Reference requirement R1 encompasses the target requirement R2 Case B: Target requirement R2 encompasses the reference requirement R1 B1: Target requirement R2 encompasses the reference requirement R1 B2: Reference requirement R1 and target requirement R2 are almost consistent, but there are differences in the quantitative descriptions (e.g., time, number of pages, etc.) Case C: Reference requirement R1 and target requirement R2 are not in an inclusive relationship C1: The sentences and words are completely different C2: The sentences are similar, but the words are different (documents to be presented are requested, but received documents, etc. are listed) C3: It describes related functions, but the requirements are different. Please respond in the following format. Judgment of target requirement R2: Corresponds to A2: Reference requirement R1 includes target requirement R2. Reason for the above judgment: Below are the sentences for reference requirement R1 and target requirement R2. Reference requirement R1: The system date is automatically entered for drafting, settlement, execution, draft document information, and drafting date. However, this can be corrected manually, and it is also possible to specify a date that is earlier than the drafting date, or a future date for advance drafting. Target requirement R2: The system date should be automatically entered for the drafting date.

[0039] In response to the above prompt PRM, the judgment processing device 10 determines whether the inclusion relationship between the reference requirement R1 and the target requirement R2 is one of cases A1, A2, B1, B2, or C1 to C3. The judgment processing device 10 also generates a reason for the judgment. Hereinafter, information including the judged case and the reason for the judgment will be referred to as the judgment result RES1. The judgment processing device 10 outputs the judgment result RES1 to the judgment result output unit 13.

[0040] 3, the determination result output unit 13 outputs the determination result RES1 generated by the determination processing device 10. The determination result output unit 13 may output the determination result RES1 to a display means such as a display device (not shown), for example. In this case, the display means visibly displays the determined cases and the reasons for the determination to the user, allowing the user to recognize to what extent the requirements of the reference document D1 match the requirements of the verification target document D2.

[0041] This allows the user to know the inclusion relationship between the reference requirement R1 and the target requirement R2. If the reference requirement R1 includes the target requirement R2, the user can determine that the requirements requested by the customer can be met without customizing the company's package product.

[0042] If target requirement R2 includes reference requirement R1, the user can recognize that the functionality of their package product only partially satisfies target requirement R2. In this case, the user can determine that customization of their package product is necessary to satisfy target requirement R2.

[0043] If the reference requirement R1 and the target requirement R2 are not in an inclusive relationship, the user can determine that the functions of the company's package product cannot satisfy the target requirement R2. In response, the user can consider adding new functions to the company's product. The user can also consult with the customer about how to handle the unsatisfied requirements or whether customization to add functions is appropriate.

[0044] Based on the information about the inclusion relationship between the reference requirement R1 and the target requirement R2, the user can create a proposal that describes an appropriate proposal for the customer's request.

[0045] As described above, the document processing system 1000 and the document processing device 100 can determine the inclusion relationship of requirements compared between multiple documents. The document processing system 1000 and the document processing device 100 can then effectively support the user in determining the similarity of requirements between multiple documents.

[0046] As a result, users of the document processing system 1000 and the document processing device 100 can use the determination results of the document processing system 1000 and the document processing device 100 to determine whether the requirements described in the RFP submitted by the customer can be met with functions that can be realized by their own package product. As a result, users can make specific proposals to customers, such as whether or not customization of their package product is necessary to meet the requirements described in the RFP. Furthermore, users can estimate or determine the amount of customization required for their own package product. This also enables users to propose to customers the cost and time required for customization.

[0047] The document processing system 1000 and the document processing device 100 can automatically determine inclusion relationships, which would take a long time to do manually. This makes it possible to efficiently reduce the manpower and time required to review RFPs in order to create proposals.

[0048] Furthermore, users of the document processing system 1000 and the document processing device 100 can also accumulate information on the customization requirements of their own package products estimated in accordance with requests written in RFPs from various customers, etc. This allows users to quickly propose to customers the costs and time required for system construction based on past performance.

[0049] Furthermore, users of the document processing system 1000 and the document processing device 100 can accumulate a history of customization items that have become necessary for their own packages based on the determination results of the inclusion relationships of requirements. This allows users to grasp trends in items that require customization for their own packages. Based on the grasped trends in customization requests, users can make proposals to their most recent customers that are in line with the trends. Users can also develop and improve their own package products so that they can keep up with the grasped trends.

[0050] The document processing system 1000 and the document processing device 100 can determine the inclusion relationships of requirements between multiple sentences without manual intervention, thereby eliminating the influence of variations in the interpretation of documents by the person making the determination. Therefore, the inclusion relationships of requirements can be determined based on more stable determination criteria than when the inclusion relationships of requirements are determined manually.

[0051] In the first embodiment, a document processing system and an apparatus for determining inclusion relationships between corresponding requirements among a plurality of documents using a determination processing device that implements a large-scale language model have been described. However, since large-scale language models generally require high computational resources and a large amount of learning processing, it may be difficult to build and maintain a large-scale language model on one's own.

[0052] Therefore, it is conceivable that multiple users will use a language processing service using a large-scale language model provided by a specific vendor. In this case, resources for the service using the large-scale language model are allocated to each user, which may result in a situation where the available resources per user are limited.

[0053] It is anticipated that language processing services using large-scale language models may be provided as paid services. Depending on the language processing service using a large-scale language model, a fee may be charged for each call to an API (Application Programming Interface) to provide a prompt to the large-scale language model. Therefore, the cost increases as the number of API calls increases. Therefore, when using a paid language processing service using a large-scale language model, it is desirable to limit the number of API calls.

[0054] Furthermore, when the document to be used as the criterion is a large document, it is difficult to input the document that will serve as the criterion into the large-scale language model itself or the API. Therefore, it is desirable to narrow down the sentences to be input into the large-scale language model itself or the API from the document that will serve as the criterion in advance.

[0055] Therefore, in the document processing system and device according to this embodiment, prior to the inclusion relationship determination process of embodiment 1, a similarity analysis of requirements is performed using another method to select requirements that require more accurate analysis.The document processing system and device then determine the inclusion relationship of the selected requirements using a large-scale language model, thereby reducing the frequency of use of the large-scale language model.

[0056] 7 is a block diagram showing a schematic configuration of a document processing system according to an embodiment. The document processing system 2000 has a configuration in which a similarity analyzer 20 and a result integration unit 30 are further provided in the document processing system 2000.

[0057] The similarity analysis device 20 includes a document reading unit 21, a document dividing unit 22, a correspondence relationship analysis unit 23, a similarity calculation unit 24, a similarity determination unit 25, and a requirement selection unit 26. The operation of the similarity analysis device 20 will be described below with reference to Fig. 8. Fig. 8 is a flowchart showing the operation of the similarity analysis device according to one embodiment.

[0058] The configurations of the document reading unit 21, document division unit 22, and correspondence relationship analysis unit 23 are respectively the same as those of the document reading unit 121, document division unit 122, and correspondence relationship analysis unit 123 according to the first embodiment, and therefore redundant explanations will be omitted. Also, steps S21 to S23 showing the operations of the document reading unit 21, document division unit 22, and correspondence relationship analysis unit 23 in Fig. 8 are respectively the same as steps S11 to S13 in Fig. 6, and therefore redundant explanations will be omitted.

[0059] As shown in step S24 of FIG. 8 , the similarity calculation unit 24 analyzes the similarity between the corresponding reference requirement R3 and target requirement R4 by natural language processing based on the detected reference requirement R3, target requirement R4, and correspondence relationship information INF. The reference requirement R3 is also referred to as the third requirement. The target requirement R4 is also referred to as the fourth requirement. Here, any natural language processing model capable of analyzing the similarity between two documents, such as a natural language processing model implementing multilingual-e5, may be applied. The natural language processing model may also be a natural language processing model constructed by machine learning and capable of estimating the similarity between two documents.

[0060] As shown in step S25 of FIG. 8 , the similarity determination unit 25 determines whether the relationship between the reference requirement R3 and the target requirement R4 is similar, dissimilar, or neither similar nor dissimilar, based on the similarity analyzed by the similarity calculation unit 24. The similarity determination unit 25 then outputs the determination result RES2 to the requirement selection unit 26 and the result integration unit 30. At this time, the similarity determination unit 25 may obtain the similarity as a numerical value, for example, by using various techniques for embedding sentences into a vector space using a natural language processing model. Furthermore, the similarity determination unit 25 may perform a threshold determination on the obtained similarity using a predetermined threshold to determine whether the relationship between the reference requirement R3 and the target requirement R4 is similar, dissimilar, or neither similar nor dissimilar.

[0061] 8, the requirement selection unit 26 outputs pairs of reference requirement R3 and target requirement R4 that are determined to be neither similar nor dissimilar based on the determination result RES2 to the document processing device 100. If there are two or more pairs that are determined to be neither similar nor dissimilar, the requirement selection unit 26 may output two or more pairs together to the document processing device 100, or may output the pairs one by one to the document processing device 100.

[0062] The document processing device 100 is the same as that in the first embodiment. The requirement reading unit 11 reads the requirement pair P output by the requirement selecting unit 26, and the judgment result output unit 13 outputs the judgment result RES1 to the result integrating unit 30.

[0063] The result integration unit 30 integrates the determination result RES1 of the inclusion relationship of requirements in the document processing device 100 and the determination result RES2 of the similarity analysis in the similarity analysis device 20, and outputs the result as a determination result RES3.

[0064] As described above, the user of document processing system 2000 can obtain the result of the similarity determination of requirements between reference document D1 and verification target document D2 performed by similarity analysis device 20 by referring to determination result RES2 included in determination result RES3. Furthermore, the user of document processing system 2000 can also obtain determination result RES1 of the inclusion relationship between requirements for which similarity determination was not possible by similarity analysis device 20. This allows the user to obtain information from determination result RES3 that is useful for determining to what extent reference document D1 covers the requirements described in verification target document D2.

[0065] Furthermore, in the document processing system 2000, the document processing device 100 determines the inclusion relationship only for requirements for which it is difficult to determine similarity using the similarity analysis device 20. This reduces the use of the document processing device 100, thereby enabling the limited resources of the determination processing device 10 to be used effectively.

[0066] Furthermore, in the case where a fee is charged each time the determination processing device 10 is used, by reducing the use of the document processing device 100, it is possible to reduce the overall cost required for document analysis.

[0067] Other Embodiments The present disclosure has been described above with reference to the embodiments, but the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0068] In the above-described embodiments, the document processing system and document processing device according to the present disclosure have been described primarily as hardware configurations, but this is not limited thereto. The document processing system and document processing device according to the present disclosure can also be realized by having a computer execute a computer program to perform any process. These processes may be realized by having a computer including at least one processor (e.g., a microprocessor, CPU, GPU, MPU, or DSP (Digital Signal Processor)) execute the program. Specifically, one or more programs including instructions for causing a computer to perform these algorithms related to transmission signal processing or reception signal processing may be created, and the programs may be supplied to the computer.

[0069] A computer program can be stored and provided to a computer using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may be provided to the computer by various types of transient computer-readable media. Examples of transient computer-readable media include electrical signals, optical signals, and electromagnetic waves. The transient computer-readable media can provide the program to the computer via a wired communication path such as an electrical wire or optical fiber, or via a wireless communication path.

[0070] An example of the configuration of a computer for realizing a document processing system and a document processing device is shown below. FIG. 9 is a diagram showing an example of the configuration of a computer for realizing a document processing system and a document processing device. The document processing system and document processing device can be realized by a computer 9000 such as a dedicated computer or a personal computer (PC). However, the computer does not need to be physically a single computer; multiple computers may be used when performing distributed processing. As shown in FIG. 9, the computer 9000 includes, for example, a processor 9001, a ROM (Read Only Memory) 9002, a RAM (Random Access Memory) 9003, a storage unit 9004, a communication interface 9005, and a user interface 9006.

[0071] The processor 9001, ROM 9002, RAM 9003, storage unit 9004, communication interface 9005, and user interface 9006 are connected to each other so as to be able to communicate with each other via a bus 9007. Note that although explanation of the OS software for operating the computer is omitted, it is also installed in the computer 9000 as appropriate.

[0072] The ROM 9002 is configured by, for example, a nonvolatile semiconductor memory device, etc. The ROM 9002 stores information such as various programs used by the computer 9000.

[0073] The storage unit 9004 is configured with various storage devices such as a hard disk, a solid state disk, etc. Furthermore, the storage unit 9004 is not limited to a storage device installed in the computer 9000, but may be a storage device external to the computer 9000. The external storage device may be a cloud storage device connected to the computer 9000 via various communication means, for example, a network. The storage unit 9004 stores information such as various programs and data used by the computer 9000.

[0074] The RAM 9003 is configured by a volatile semiconductor memory device, etc. Programs, data, and other information used by the processor 9001 are loaded into the RAM 9003 from one or both of the ROM 9002 and the storage unit 9004 as appropriate.

[0075] The processor 9001 may be configured with, for example, a CPU (Central Processing Unit). Furthermore, the processor 9001 may include not only a CPU but also a GPU (Graphics Processing Unit). A GPU is suitable for performing routine processing in parallel, and by applying it to neural network processing, for example, it is possible to improve processing speed compared to a CPU. The processor 9001 executes various processes based on various programs stored in the ROM 9002 or various programs and data held in the RAM 9003, as appropriate. Furthermore, the processor 9001 may store data generated by the processing in the RAM 9003 or the storage unit 9004, as appropriate.

[0076] The communication interface 9005 is an interface that connects the computer 9000 to a communication network such as the Internet or an intranet via various wired communication means or wireless communication means, etc. This allows the computer 9000 to communicate with other devices, systems, sensors, etc. that are connected to the communication network.

[0077] The user interface 9006 includes, for example, a display unit that provides information so that the user can recognize it using a display device or the like, and an audio output unit that outputs audio. The user interface 9006 also includes an input unit that allows the user to input information to the computer 9000 by operating it, such as a keyboard, a mouse, or a touch panel. The user interface 9006 may also include devices such as sensors that obtain information useful to the user.

[0078] Although the computer 9000 has been described as a single device here, this is merely an example. The computer 9000 may be composed of multiple physically separated devices. Some of the multiple devices may be portable devices, and the other devices may be stationary devices.

[0079] Each drawing is merely an example for describing one or more embodiments. Each drawing may not relate to only one particular embodiment, but may also relate to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessary to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0080] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes.

[0081] (Appendix 1) A document processing system comprising: a requirement reading means for reading a first requirement contained in a first document and a second requirement contained in a second document; an instruction means for outputting an instruction for determining an inclusion relationship between the first requirement and the second requirement; a determination means for determining an inclusion relationship between the first requirement and the second requirement in accordance with the instruction; and an output means for outputting the determination result by the determination means.

[0082] (Supplementary Note 2) The document processing system according to Supplementary Note 1, wherein the determining means determines an inclusion relationship between the first requirement and the second requirement using a large-scale language model.

[0083] (Supplementary Note 3) The document processing system according to Supplementary Note 2, wherein the large-scale language model determines an inclusion relationship between the first requirement and the second requirement based on a prompt given as the instruction.

[0084] (Appendix 4) The document processing system described in Appendix 3, wherein the determination means provides the instruction to the determination means through an API (Application Programming Interface), and the output means receives the determination result from the determination means through the API.

[0085] (Appendix 5) A document processing system described in any one of Appendices 1 to 4, wherein the instruction means instructs the determination means to determine whether the inclusion relationship between the first requirement and the second requirement is a first case in which the first requirement matches the second requirement, or the first requirement encompasses the second requirement, a second case in which the second requirement encompasses the first requirement, or a third case in which the first requirement and the second requirement are not in an inclusion relationship with each other.

[0086] (Appendix 6) The document processing system described in Appendix 5, wherein the instruction means instructs the judgment means to include in the judgment result the reason for determining that the document is one of the first to third cases.

[0087] (Appendix 7) A document processing system as described in any one of Appendices 1 to 6, further comprising a similarity analysis means for analyzing the similarity between a plurality of third requirements contained in the first document and a plurality of fourth requirements contained in the second document that correspond to the plurality of third requirements, and selecting one or more sets of the third requirements and the fourth requirements that are determined to require further similarity analysis; and a result integration means for integrating and outputting the analysis results of the similarity between the plurality of third requirements in the similarity analysis means and the fourth requirements that correspond to the plurality of third requirements, respectively, and the judgment results for the one or more sets of the first requirements and the second requirements, wherein the requirement reading unit reads the third requirements and the fourth requirements of each set of the one or more sets of the third requirements and the fourth requirements as the first requirements and the second requirements, respectively.

[0088] (Appendix 8) A document processing device comprising: a requirement reading means for reading a first requirement contained in a first document and a second requirement contained in a second document; an instruction means for outputting an instruction for determining an inclusion relationship between the first requirement and the second requirement to a determination means for determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction; and an output means for outputting the determination result by the determination means.

[0089] (Appendix 9) A document processing method that reads a first requirement contained in a first document and a second requirement contained in a second document, outputs an instruction to determine the inclusion relationship between the first requirement and the second requirement, determines the inclusion relationship between the first requirement and the second requirement in accordance with the instruction, and outputs the result of the determination.

[0090] (Appendix 10) A program that causes a computer to perform the following processes: reading a first requirement contained in a first document and a second requirement contained in a second document; outputting an instruction to determine the inclusion relationship between the first requirement and the second requirement; determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction; and outputting the result of the determination.

[0091] Some or all of the elements (e.g., configurations and functions) described in Supplementary Notes 2 to 7 that are dependent on Supplementary Notes 1 may also be dependent on Supplementary Notes 8 to 10 in the same dependency relationship as Supplementary Notes 2 to 7. Some or all of the elements described in any Supplementary Note may be applied to various hardware, software, recording means for recording software, systems, and methods.

[0092] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the invention.

[0093] This application claims priority based on Japanese Patent Application No. 2024-29827, filed February 29, 2024, the disclosure of which is incorporated herein in its entirety by reference.

[0094] REFERENCE SIGNS LIST 10 Determination processing device 11 Requirement reading unit 12 Instruction unit 13 Determination result output unit 14 Memory unit 20 Similarity analysis device 21, 121 Document reading unit 22, 122 Document division unit 23, 123 Correspondence analysis unit 24 Similarity calculation unit 25 Similarity determination unit 26 Requirement selection unit 30 Result integration unit 100 Document processing device 120 Preprocessing device 124 Requirement output unit 1000, 2000 Document processing system 9000 Computer 9001 Processor 9002 ROM 9003 RAM 9004 Memory unit 9005 Communication interface 9006 User interface 9007 Bus D1, d1 Reference document D2, d2 Verification target document INF Correspondence information P Requirement pair R1, R3 Reference requirement R2, R4 Target requirement RPM prompt RES1, RES2, RES3 Judgment result

Claims

1. A document processing system comprising: a requirement reading means for reading a first requirement contained in a first document and a second requirement contained in a second document; an instruction means for outputting an instruction to determine the inclusion relationship between the first requirement and the second requirement; a determination means for determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction; and an output means for outputting the determination result by the determination means.

2. The document processing system according to claim 1, wherein the determining means determines the inclusion relationship between the first requirement and the second requirement using a large-scale language model.

3. The document processing system according to claim 2, wherein the large-scale language model determines an inclusion relationship between the first requirement and the second requirement based on a prompt given as the instruction.

4. The document processing system according to claim 3, wherein the determination means provides the instruction to the determination means through an API (Application Programming Interface), and the output means receives the determination result from the determination means through the API.

5. A document processing system as described in claim 1 or 2, wherein the instruction means instructs the determination means to determine whether the inclusion relationship between the first requirement and the second requirement is one of the following three cases: a first case in which the first requirement matches the second requirement, or the first requirement encompasses the second requirement; a second case in which the second requirement encompasses the first requirement; or a third case in which the first requirement and the second requirement are not in an inclusion relationship with each other.

6. The document processing system according to claim 5, wherein the instruction means instructs the determination means to include in the determination result the reason for determining that the document is one of the first to third cases.

7. A document processing system as described in claim 1 or 2, further comprising: a similarity analysis means for analyzing the similarity between a plurality of third requirements included in the first document and a plurality of fourth requirements included in the second document that respectively correspond to the plurality of third requirements, and selecting one or more sets of the third requirements and the fourth requirements that are determined to require further similarity analysis; and a result integration means for integrating and outputting the analysis result of the similarity between the plurality of third requirements and the fourth requirements that respectively correspond to the plurality of third requirements obtained by the similarity analysis means, and the judgment result for the one or more sets of the first requirements and the second requirements, wherein the requirement reading unit reads the third requirements and the fourth requirements of each set of the one or more sets of the third requirements and the fourth requirements as the first requirements and the second requirements, respectively.

8. A document processing device comprising: a requirement reading means for reading a first requirement contained in a first document and a second requirement contained in a second document; an instruction means for outputting an instruction to a determination means for determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction; and an output means for outputting the determination result by the determination means.

9. A document processing method comprising: reading a first requirement contained in a first document and a second requirement contained in a second document; outputting an instruction to determine the inclusion relationship between the first requirement and the second requirement; determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction; and outputting the result of the determination.

10. A program that causes a computer to perform the following processes: reading a first requirement contained in a first document and a second requirement contained in a second document; outputting an instruction to determine the inclusion relationship between the first requirement and the second requirement; determining the inclusion relationship between the first requirement and the second requirement in accordance with the instruction; and outputting the results of the determination.

Images