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

Abstract

Previously, it was not possible to perform masking based on rules according to the attribute values ​​of a document. [Solution] A document processing device 1 is provided with a rule management unit 111 in which masking rules are stored in association with one or more document attribute value conditions, a rule acquisition means 1312 that acquires one or more document attribute values ​​associated with a first document and acquires masking rules satisfied by the one or more document attribute values ​​from the rule management unit 111, confidentiality acquisition means 1313 that acquires confidential information, which is partial information that meets the confidentiality conditions of the masking rules, from the first document, a masking unit 133 that performs masking processing to mask the confidential information, and a document output unit 141 that outputs a second document after the masking processing has been performed by the masking unit 133.This enables masking based on rules corresponding to the attribute values ​​of the document.

Description

[Technical Field]

[0001] The present invention relates to a document processing device that masks a portion of a document. [Background technology]

[0002] In the past, in order to protect confidential information in confidential documents, there was a system that overlaid a substitute object on a character block containing the confidential information so that the confidential information was not visible within the document (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-155104 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the prior art, it was not possible to perform masking based on rules according to the attribute values ​​of a document.

[0005] In the prior art, for example, it was sometimes possible to guess the confidential information from the character string length of the masked confidential information. In the prior art, for example, it was at least possible to know that the information was masked. [Means for solving the problem]

[0006] The document processing device of the first invention is a document processing device comprising: a rule management unit in which masking rules including confidentiality conditions are stored in correspondence with one or more document attribute value conditions, which are conditions related to one or more document attribute values ​​of a first document; a document reception unit that receives a first document including two or more pieces of partial information corresponding to one or more document attribute values; a rule acquisition means that acquires one or more document attribute values ​​corresponding to the first document received by the document reception unit and acquires masking rules satisfied by the one or more document attribute values ​​from the rule management unit; confidentiality acquisition means that acquires confidential information from the first document, which is partial information that matches the confidentiality conditions of the masking rules acquired by the rule acquisition means; a masking unit that performs a masking process to mask the confidential information in the first document; and a document output unit that outputs a second document, which is the document after the masking process has been performed by the masking unit.

[0007] With this configuration, masking can be performed based on rules according to the attribute values ​​of the document.

[0008] Furthermore, the document processing device of the second invention is a document processing device in which, compared to the first invention, the one or more document attribute values ​​corresponding to the first document include one or more user attribute values ​​that are attribute values ​​of the user of the first document, and the document attribute value conditions include user attribute value conditions that are conditions related to the one or more user attribute values ​​of the user of the first document.

[0009] With this configuration, masking can be performed based on rules according to user attribute values.

[0010] Furthermore, the document processing device of the third invention is a document processing device in which, compared to the second invention, the one or more document attribute values ​​corresponding to the first document include two or more user attribute values, and the two or more user attribute values ​​include two or more hierarchical organization identifiers to which the users belong, the rule management unit stores masking rules corresponding to the organization identifiers of each of the two or more hierarchical levels, and the rule acquisition means acquires the two or more organization identifiers corresponding to the first document accepted by the document acceptance unit, and acquires the masking rules corresponding to each of the two or more organization identifiers from the rule management unit.

[0011] This configuration enables masking based on rules managed in pairs with the organization identifiers of the hierarchical organizations to which the users belong.

[0012] Furthermore, the document processing device of the fourth invention is a document processing device in which, compared to the first invention, the one or more document attribute values ​​corresponding to the first document include a document placement attribute value related to the location where the first document is placed, and the one or more document attribute value conditions of the rule management unit include a document placement attribute value condition that is a condition related to the document placement attribute value of the first document.

[0013] This configuration allows masking based on rules according to where the document is placed.

[0014] Furthermore, the document processing device of the fifth invention is a document processing device in which, compared to the fourth invention, the one or more document attribute values ​​include a device identifier that identifies the device on which the first document is placed, or a folder identifier that identifies the folder on which the first document is placed, and the one or more document attribute value conditions of the rule management unit include a device condition that is a condition related to the device identifier of the first document, or a folder condition that is a condition related to the folder identifier of the first document.

[0015] Such a configuration allows for rule-based masking depending on the device or folder in which the document is located.

[0016] Furthermore, the document processing device of the sixth invention is a document processing device in which, compared to the first invention, the one or more document attribute values ​​corresponding to the first document include a type identifier that identifies the type of the document, and the one or more document attribute value conditions of the rule management unit include a document type condition that is a condition related to the type identifier of the first document.

[0017] This configuration enables masking based on rules according to the type of document.

[0018] Furthermore, in the document processing device of the seventh invention, in contrast to the sixth invention, the first document is a file, and the type identifier is information for identifying the type of the file.

[0019] With this configuration, masking can be performed based on rules according to the type of document file. [Effects of the Invention]

[0020] The document processing device according to the present invention can perform masking based on rules according to the attribute values ​​of the document. [Brief explanation of the drawings]

[0021] [Figure 1] Conceptual diagram of document processing system A according to the first embodiment [Figure 2] Block diagram of the document processing system A [Figure 3] Block diagram of the document processing device [Figure 4] A flowchart illustrating an example of the operation of the document processing device 1. [Figure 5] A flowchart illustrating an example of the masking determination process [Figure 6] A flowchart illustrating an example of the rule acquisition process [Figure 7]A flowchart illustrating an example of the masking process [Figure 8] A flowchart illustrating an example of the confidential information acquisition process [Figure 9] A flowchart illustrating an example of the character string inspection process [Figure 10] A flowchart illustrating an example of the alternative acquisition process [Figure 11] A flowchart illustrating an example of the document selection process [Figure 12] A flowchart illustrating an example of the operation of the terminal device 2 [Figure 13] A diagram showing the rule management table [Figure 14] The figure shows the document attribute value condition management table. [Figure 15] A diagram showing the user management table [Figure 16] A diagram showing the document management table [Figure 17] Block diagram of the computer system DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, embodiments of a document processing apparatus and the like will be described with reference to the drawings. Note that components with the same reference numerals in the embodiments perform similar operations, and therefore repeated description may be omitted.

[0023] (Embodiment 1) In this embodiment, a document processing device will be described that changes confidential information in a received document to, for example, ambiguous information or fixed-length information, thereby creating a document that does not leak confidential information. In particular, in this embodiment, a document processing device will be described that changes confidential information in a received document to, for example, ambiguous information, such that the masking is not apparent.

[0024] In this embodiment, a document processing apparatus that manages rules for two or more document attribute value conditions and applies a rule according to the document attribute value to mask confidential information in a document will be described.

[0025] In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is not important. Information X and information Y may be linked, may exist in the same buffer, information X may be included in information Y, or information Y may be included in information X, etc.

[0026] Furthermore, in this specification, selecting or determining information Z means obtaining information Z, obtaining a pointer to information Z, obtaining the ID of information Z, setting a flag on information Z, etc., and it is sufficient if information Z can be accessed.

[0027] 1 is a conceptual diagram of a document processing system A according to the present embodiment. The document processing system A includes a document processing device 1 and one or more terminal devices 2.

[0028] The document processing device 1 is a device that masks confidential information in accepted documents. The document processing device 1 is typically a server, but may also be a terminal. When the document processing device 1 is a server, the document processing device 1 may be, for example, a cloud server or an ASP server, but the type does not matter. When the document processing device 1 is a terminal, the document processing device 1 may be, for example, a personal computer, a smartphone, or a tablet terminal, but the type does not matter. When the document processing device 1 is a terminal, the terminal device 2 is not required.

[0029] The terminal device 2 is a terminal used by a user. The user may be, for example, a person who masks confidential information in a document, a document creator, or a person who acquires a document with masked confidential information. The terminal device 2 may be, for example, a personal computer, a smartphone, or a tablet terminal, but the type of the terminal device is not important.

[0030] The document processing apparatus 1 and one or more terminal devices 2 can communicate with each other via a network such as the Internet.

[0031] Fig. 2 is a block diagram of a document processing system A according to the present embodiment, and Fig. 3 is a block diagram of a document processing apparatus.

[0032] The document processing device 1 includes a storage unit 11, a reception unit 12, a processing unit 13, and an output unit 14. The storage unit 11 includes a rule management unit 111, a user management unit 112, and a document management unit 113. The reception unit 12 includes a document reception unit 121. The processing unit 13 includes a confidentiality acquisition unit 131, a substitute acquisition unit 132, and a masking unit 133. The confidentiality acquisition unit 131 includes a disclosure acquisition means 1311, a rule acquisition means 1312, and a confidentiality acquisition means 1313. The output unit 14 includes a document output unit 141.

[0033] The terminal device 2 includes a terminal storage unit 21, a terminal reception unit 22, a terminal processing unit 23, a terminal transmission unit 24, a terminal reception unit 25, and a terminal output unit 26.

[0034] Various types of information are stored in the storage unit 11 that constitutes the document processing device 1. The various types of information include, for example, masking rules (described later), confidentiality conditions (described later), user information (described later), various prompts, a concept dictionary, and masking conditions. It goes without saying that the confidentiality conditions, masking conditions, and other conditions stored in the storage unit 11 may be embedded in a program.

[0035] Here, a prompt is a prompt given to the generation AI. A prompt may also be called an order, an instruction, or the like. Various types of prompts include, for example, a confidential information acquisition prompt and an alternative information acquisition prompt. Alternative information acquisition prompts include, for example, a higher-level information acquisition prompt, an ambiguous numerical information acquisition prompt, an image description information acquisition prompt, or a summary information acquisition prompt.

[0036] The generation AI here has the function of a text generation AI. The generation AI may be, for example, ChatGPT or Gemini, but this does not matter. The generation AI is a function possessed by a generation AI device (not shown). The generation AI device may also be called the generation AI. The generation AI device may be, for example, a cloud server or an ASP server, but the type does not matter. The document processing device 1 may also have the function of a generation AI. In such a case, a generation AI device is not required.

[0037] A confidential information acquisition prompt is a prompt for acquiring confidential information. For example, a confidential information acquisition prompt might be, "Please perform the task of analyzing the text contained in the given document below, detecting any corporate, personal, or confidential information contained therein, and returning it in JSON format. [Document] <Document>." In the prompt, strings enclosed in "<" and ">" are variables, and information is assigned to the variables. In other words, the accepted document is assigned to the variable <Document>.

[0038] The substitute information acquisition prompt is a prompt for acquiring substitute information.

[0039] A prompt for obtaining higher-level information is a prompt for obtaining higher-level information for confidential information. Higher-level information for confidential information is information that is a higher-level concept of confidential information. Higher-level information for confidential information is information that abstracts confidential information. An example of a prompt for obtaining higher-level information is, "Output the higher-level concept word for the following <confidential information>! [confidential information] <confidential information>."

[0040] The ambiguous numeric information acquisition prompt is a prompt for acquiring ambiguous numeric information, which is information that makes a numeric value ambiguous, when the confidential information is a number. Examples of ambiguous numeric information include numeric range information and approximate number information. Numeric range information is information that indicates a range of numbers. Approximate number information is information that indicates an approximate number of a number. An example of an ambiguous numeric information acquisition prompt is, "Output an appropriate numeric range that includes the following number! [Number] <Confidential information>". An example of an ambiguous numeric information acquisition prompt is, "Output an approximate number of the following number! [Number] <Confidential information>".

[0041] The image description information acquisition prompt is a prompt for acquiring information that describes an image when the confidential information is an image. For example, the image description information acquisition prompt is, "Output the hypernym of the company name of the image (company logo) below! The output should be "company logo of 'hypernym of company name'. [Image] <Confidential Information>."

[0042] A summary information acquisition prompt is a prompt for acquiring information that summarizes confidential information. For example, a summary information acquisition prompt might be, "Summarize the following text in 400 characters or less! When summarizing, please replace the company name with a broader term. [Text] <Confidential information>."

[0043] A concept dictionary is information that has two or more corresponding pieces of information, each having lower-level information and higher-level information. Lower-level information is information that is a lower concept than the higher-level information. Higher-level information is information that is a higher concept than the lower-level information. The concept dictionary may reside on an external server (not shown).

[0044] A masking condition is a condition for determining whether or not to perform masking processing on a document. A masking condition is usually a condition related to one or more document attribute values. Examples of masking conditions include "a masking flag is associated with the accepted document," "a non-masking flag is not associated with the accepted document," "the document type identifier is a predetermined type identifier," and "the document type identifier is not a predetermined type identifier."

[0045] The rule management unit 111 stores one or more masking rules. A masking rule is a rule for masking confidential information in a document. A masking rule typically includes a confidentiality condition. A masking rule may include, for example, an alternative definition. A masking rule may not include an alternative definition. If a masking rule does not include an alternative definition, the masking rule is the same as the confidentiality condition. Each of the one or more masking rules in the rule management unit 111 typically corresponds to a document attribute value condition.

[0046] Confidentiality conditions are conditions for determining confidential information. Confidential information is information that is to be masked. Confidential information is information that is not normally shown to users. Confidential information is, for example, information that may cause disadvantages if leaked to the outside. There are various possible contents of confidential information, and it is not restricted to these.

[0047] The confidentiality condition is, for example, that the information is a character string that belongs to a specific class. The confidentiality condition is, for example, that the information is an image that belongs to a specific class that is a predetermined class. The confidentiality condition is, for example, that the information is non-public information.

[0048] A specific class is a predetermined specific class. A class may be called, for example, a classification or an attribute. A specific class is, for example, a company name, a name, an address, a job title, an amount, a sales figure, or a number of customers. An amount is, for example, a sales amount, a profit amount, a salary amount, a raw material amount, a purchase price, a cost price, or an import price.

[0049] When the confidential information is an image, the confidentiality condition is, for example, that the image is an image of a specific class. An example of the specific class is a company logo. When the confidential information is audio information, the confidentiality condition is, for example, that the audio information contains information of a specific class. An example of the specific class is the company name, name, address, job title, amount, sales volume, and number of customers mentioned above. An example of the amount is sales amount, profit amount, salary amount, raw material amount, purchase price, cost price, and import price.

[0050] An alternative definition is information used to obtain alternative information. An alternative definition may be the alternative information itself. Alternative information is information placed in a document in place of confidential information. An alternative definition is, for example, information that defines a method for obtaining alternative information. An alternative definition is, for example, information about an API for obtaining alternative information, or information that includes a prompt for obtaining alternative information. An alternative definition is, for example, "higher-level information," "*****," "a five-character string consisting of "*," "ambiguous numeric information," "explanatory information," or "summary information." An alternative definition is, for example, information that defines a method for obtaining alternative information.

[0051] The alternative definition of "higher-level information" refers to obtaining information that is higher in level than the confidential information. The alternative definitions "*****" or "five characters of "*"" refer to a string of five asterisks. The alternative definition of "ambiguous numerical information" refers to obtaining information that obscures the numerical value when the confidential information is a number. "Ambiguous numerical information" refers to obtaining, for example, numerical range information or approximate number information. Numeric range information is information (e.g., "1000-2000") that indicates a range of numbers that includes the numerical value indicated by the confidential information (e.g., 1528). Approximate number information is an approximation (e.g., "1500") of the numerical value indicated by the confidential information (e.g., 1528). The alternative definition of "explanatory information" refers to obtaining information that explains the image when the confidential information is an image. The alternative definition of "summary information" refers to obtaining information that summarizes the confidential information. The alternative definition may also be "blacked out," which indicates blacking out.

[0052] A document attribute value condition is a condition related to one or more document attribute values ​​of a document to be masked. A document corresponds to one or more document attribute values. A document attribute value is an attribute value of a document. The document attribute value may be, for example, a user attribute value, a document location attribute value, or a type identifier.

[0053] A user attribute value is a user's attribute value. For example, a user attribute value is one or more organization identifiers, job titles, years of employment, age, or gender. An organization identifier is information that identifies an organization to which a user belongs. For example, an organization identifier is an organization name or organization ID. It is preferable that a user attribute value includes two or more hierarchical organization identifiers. For example, two or more hierarchical organization identifiers are a company identifier and a business division identifier. For example, two or more hierarchical organization identifiers are a company identifier, a business division identifier, and a department identifier. For example, two or more hierarchical organization identifiers are a company identifier and a department identifier. For example, two or more hierarchical organization identifiers are a company identifier, a department identifier, and a section identifier. For example, a company identifier is a company name or a company ID. For example, a business division identifier is a business division name or a business division ID. For example, a department identifier is a department name or a department ID. For example, a department identifier is a department name or a department ID. For example, a department identifier is a department name or a department ID. For example, a department identifier is a department name or a department ID. For example, a section identifier is a section name or a section ID. The position is the position of the user. Examples of the position are "department manager," "section manager," "supervisor," and "employee." The number of years of employment is the number of years the user has been employed in an organization (usually a company).

[0054] A document location attribute value is information about the location where a document is located. The location where a document is located is a location in electromagnetic space. A document location attribute value is the device identifier of the device where the document is located, or the folder identifier of the folder where the document is located. A device identifier is information that identifies a device. For example, a device name, a device ID, or an IP address. A folder identifier is information that identifies a folder. For example, a folder identifier is a folder name or a folder ID.

[0055] A type identifier is information that identifies the type of document. For example, a type identifier is a format type identifier or a content type identifier. A format type identifier is information that indicates the format of a file. For example, a format type identifier is "PPTX" or "DOCX." A content type identifier is information that indicates the type of document based on the content of the document. For example, a content type identifier is "proposal," "sales material," "internal material," or "brochure."

[0056] It is preferable that the rule management unit 111 stores masking rules associated with organization identifiers at two or more levels.

[0057] The user management unit 112 stores one or more pieces of user information. User information is information related to a user. User information is usually associated with a user identifier. A user identifier is information that identifies a user. The user identifier is, for example, a user ID, a name, a telephone number, or an email address. User information usually has one or more user attribute values.

[0058] The document management unit 113 stores one or more documents. Preferably, the documents here are documents for which masking, which will be described later, has been completed. However, the document management unit 113 may also store documents that have not been masked. When the document management unit 113 contains a mixture of masked and unmasked documents, it is preferable that information be present that enables determination of whether or not a document has been masked. When a mixture of masked and unmasked documents exists, it is preferable that, for example, each document be associated with a flag for determining whether or not it has been masked. Note that, where appropriate, an unmasked document will be referred to as a first document, and a masked document will be referred to as a second document.

[0059] The reception unit 12 receives various instructions and information. The reception unit 12 usually receives various instructions and information from the terminal device 2. The various instructions and information are, for example, documents and output instructions. An output instruction is an instruction to output a document. The output instruction has, for example, a document identifier of the document to be output. The output instruction has, for example, a user identifier.

[0060] Here, acceptance typically refers to the reception of information transmitted via a wired or wireless communication line, but may also be a concept that includes the reception of information input from an input device such as a keyboard, mouse, or touch panel, or the reception of information read from a recording medium such as an optical disk, magnetic disk, or semiconductor memory.

[0061] The document receiving unit 121 receives one or more documents. The document receiving unit 121 may also receive a document to which a judgment flag has been assigned. The document receiving unit 121 receives one or more documents from, for example, the terminal device 2. Note that such a document corresponds to one or more document attribute values ​​and includes two or more pieces of partial information.

[0062] The judgment flag is information indicating whether or not a document is subject to masking. The judgment flag may be information assigned to a document that is subject to masking, or information assigned to a document that is not subject to masking, or information indicating whether or not a document is subject to masking (for example, a judgment flag of "1" indicates that a document is subject to masking, and a judgment flag of "0" indicates that a document is not subject to masking), and may be information assigned to all files.

[0063] Partial information is information that constitutes a document. Partial information is usually a character string, but may also be image or audio information. Images are usually still images, but may also be moving images.

[0064] The processing unit 13 performs various types of processing, such as processing performed by the confidentiality acquisition unit 131, the alternative acquisition unit 132, or the masking unit 133.

[0065] The confidential information acquisition unit 131 acquires one or more pieces of confidential information from the document accepted by the document acceptance unit 121. The confidential information is partial information that meets the confidentiality conditions.

[0066] The confidential information acquisition unit 131 acquires confidential information, which is a character string belonging to a specific class, from a document accepted by the document acceptance unit 121, for example.

[0067] The confidential information acquisition unit 131 acquires confidential information, which is non-public information, from the document accepted by the document acceptance unit 121, for example.

[0068] More specifically, the confidential information acquisition unit 131 acquires confidential information from a document by, for example, one of the following methods (1) to (3). (1) Using generative AI

[0069] The confidential information acquisition unit 131 acquires a confidential information acquisition prompt from the storage unit 11. The confidential information acquisition unit 131 acquires a document to be masked. Next, the confidential information acquisition unit 131 provides the confidential information acquisition prompt and the document to the generation AI, and acquires one or more pieces of confidential information from the generation AI. (2) Language processing method

[0070] The confidential information acquisition unit 131 acquires two or more pieces of partial information from the document to be masked. For example, the confidential information acquisition unit 131 acquires one or two or more sentences from the document to be masked. The confidential information acquisition unit 131 performs morphological analysis on each of the one or two or more sentences to acquire two or more morphemes. Next, for each of the two or more morphemes, the confidential information acquisition unit 131 determines whether the class of the morpheme corresponds to a specific class such as "company name," "name," or "amount." The confidential information acquisition unit 131 acquires one or more morphemes (terms) that correspond to the specific class as confidential information.

[0071] The method for acquiring the class of a character string is not limited. The method for acquiring the class of a character string is a highly advanced technology. The confidential information acquisition unit 131 acquires the class of a character string, for example, by machine learning prediction processing. The confidential information acquisition unit 131 provides a natural language model such as BERT or GPT and the character string to a module that performs machine learning prediction processing, executes the module, and acquires the class of the character string. The confidential information acquisition unit 131 searches for a class that pairs with the character string from, for example, a dictionary that has two or more pieces of correspondence information having character strings and classes. The confidential information acquisition unit 131 acquires the class of a character string, for example, by a rule base. Note that the rules in the rule base are, for example, "terms including 'Kabushiki Kaisha' and 'Limited Company' belong to the class 'company name'" and "kanji strings of two to five characters belong to the class 'personal name'." (3) How to combine generative AI and language processing

[0072] The confidential information acquisition unit 131 acquires two or more pieces of partial information from the document to be masked. For example, the confidential information acquisition unit 131 acquires one or more sentences from the document to be masked. The confidential information acquisition unit 131 performs morphological analysis on each of the one or more sentences to acquire two or more morphemes. The confidential information acquisition unit 131 also acquires one or more images from the document to be masked.

[0073] Furthermore, the confidential information acquisition unit 131 acquires from the storage unit 11 a confidential information acquisition prompt, which is a prompt asking whether or not a morpheme (term) is a term of a specific class.

[0074] Next, the confidential information acquisition unit 131 provides the generation AI with two or more morphemes and the confidential information acquisition prompt, acquires a response from the generation AI, and acquires one or more pieces of confidential information based on the response.

[0075] The confidential information acquisition unit 131 also acquires from the storage unit 11 a confidential information acquisition prompt, which is a prompt asking whether one or more images in the document are images of a specific class (for example, "company logo").

[0076] Next, the confidential information acquisition unit 131 provides one or more images and the confidential information acquisition prompt to the generation AI, acquires a response from the generation AI, and acquires one or more pieces of confidential information based on the response.

[0077] The public information acquisition unit 1311 acquires public information. Public information is information that is publicly available. Public information does not include confidential information.

[0078] The public information acquisition unit 1311 acquires, for example, public information input by a user. The public information corresponds to, for example, the first document to be masked. The public information is, for example, a file. The public information is, for example, an HTML file of a company's homepage, a company introduction file, a product brochure, or a product introduction file.

[0079] The public information acquisition unit 1311 acquires public information, for example, by a web search. The web search conditions are provided, for example, by a user, but the method for acquiring the conditions does not matter. The web search conditions correspond, for example, to the first document to be masked. The information acquired by the web search is public information. In addition, the web search conditions are, for example, the "company name" and the URL of the company's homepage.

[0080] The rule acquisition means 1312 acquires one or more masking rules from the rule management unit 111 .

[0081] The rule acquisition unit 1312 acquires, for example, one or more document attribute values ​​associated with a document accepted by the document acceptance unit 121. Next, the rule acquisition unit 1312 acquires, from the rule management unit 111, one or more masking rules that are satisfied by the one or more document attribute values.

[0082] The rule acquisition unit 1312 acquires, for example, two or more organization identifiers associated with the document accepted by the document acceptance unit 121. Next, the rule acquisition unit 1312 acquires one or more masking rules associated with each of the two or more organization identifiers from the rule management unit 111.

[0083] The one or more document attribute values ​​associated with the document may include, for example, one or more user attribute values ​​that are attribute values ​​of a user of the document. The one or more document attribute values ​​associated with the document may include, for example, a document location attribute value. The one or more document attribute values ​​associated with the document may include, for example, a type identifier that identifies the type of the document.

[0084] The confidential information acquisition unit 1313 acquires, for example, confidential information, which is non-public information not included in the public information acquired by the public information acquisition unit 1311, from the document accepted by the document acceptance unit 121.

[0085] The confidential information acquisition unit 1313 acquires, for example, confidential information, which is partial information that matches the confidentiality conditions of the masking rules acquired by the rule acquisition unit 1312, from the document accepted by the document acceptance unit 121.

[0086] The confidential information acquisition means 1313 acquires, for example, one or more pairs of confidential information and location identification information. The location identification information is information that identifies a location within a document. The location identification information is, for example, a pointer, an inspection unit ID and an offset within the inspection unit, and an offset within the document.

[0087] The substitute acquisition unit 132 acquires substitute information that is information for replacing confidential information.

[0088] The alternative acquisition unit 132 acquires alternative information that is, for example, higher-level information, which is information of a higher concept than the confidential information, or fixed-length information that does not depend on the data length of the confidential information, or ambiguous numerical information that obscures the numerical value when the confidential information is a number, or explanatory information that explains the image when the confidential information is an image, or summary information that summarizes the confidential information.

[0089] The alternative acquisition unit 132 performs a first acquisition process of acquiring alternative information that is higher-level information of confidential information from a concept dictionary that has two or more pieces of correspondence information that have lower-level information that is information on lower concepts and higher-level information.

[0090] The alternative acquisition unit 132 performs a second acquisition process, for example, by providing the generated AI with confidential information and an alternative information acquisition prompt for acquiring higher-level information, and acquiring alternative information, which is higher-level information corresponding to the confidential information, from the generated AI.

[0091] The alternative acquisition unit 132 preferably acquires alternative information using information other than confidential information, including information in the document. The document is a document to be masked. The alternative acquisition unit 132 acquires alternative information that includes, for example, the difference in the position of the document subject relative to the company. The document subject is, for example, the user who sent the document to the document processing device 1 or the document creator. The position is, for example, a "rival company," "competitor company," "group company," or "business partner company" relative to the document subject.

[0092] The substitute acquisition unit 132 acquires substitute information that is, for example, numerical range information indicating a range including a numerical value that is confidential information. The substitute acquisition unit 132 acquires substitute information that is, for example, approximate number information that is an approximation of a numerical value that is confidential information. The substitute acquisition unit 132 performs a fifth acquisition process, for example, by providing confidential information and an alternative information acquisition prompt for acquiring the numerical range information or approximate number information of the confidential information to the generation AI and acquiring substitute information that is the numerical range information or approximate number information of the confidential information from the generation AI. The substitute acquisition unit 132 acquires, for example, numerical range information that includes an upper limit value rounded up for the numerical value that is confidential information and a lower limit value rounded up for the numerical value. The substitute acquisition unit 132 acquires, for example, approximate number information that adds a "approximate" to a numerical value obtained by rounding the numerical value that is confidential information.

[0093] The alternative acquisition unit 132 performs a third acquisition process, for example, by providing the generation AI with an image that is confidential information and an alternative information acquisition prompt for acquiring explanatory information about the image, and acquiring alternative information that is explanatory information that explains the image from the generation AI.

[0094] The substitute acquisition unit 132 performs a fourth acquisition process to acquire substitute information, which is, for example, a predetermined image or character string in place of an image that is confidential information. In this case, the image or character string is stored in the storage unit 11, for example.

[0095] The alternative acquisition unit 132 performs a fifth acquisition process in which the generation AI is provided with the confidential information and an alternative information acquisition prompt for acquiring summary information, and alternative information, which is summary information that summarizes the confidential information, is acquired from the generation AI. The alternative acquisition unit 132 may acquire the alternative information, which is summary information that summarizes the confidential information, by natural language processing, for example. Summarizing information by natural language processing is a well-known technique.

[0096] The alternative acquisition unit 132 may acquire information for blacking out confidential information, for example.

[0097] The masking unit 133 performs a masking process to mask confidential information in the first document.

[0098] The masking unit 133 performs a masking process to replace confidential information in the first document with the alternative information acquired by the alternative acquisition unit 132. The process of replacing confidential information with alternative information may be any process that hides the confidential information and displays the alternative information in the document to be masked. The process of replacing confidential information with alternative information may, for example, be a process of deleting the confidential information and inserting the alternative information in place of the confidential information. The process of replacing confidential information with alternative information may, for example, be a process of overwriting the confidential information with the alternative information. The process of replacing confidential information with alternative information may, for example, be a process of attaching a deletion flag to the confidential information and inserting the alternative information in place of the confidential information.

[0099] The masking unit 133 may, for example, black out confidential information in the first document.

[0100] The output unit 14 outputs various types of information. The various types of information are, for example, documents. The documents are preferably second documents after masking processing, but may also be first documents before masking processing.

[0101] Here, output typically means transmission to terminal device 2, but it may also be a concept that includes display on a display, projection using a projector, printing on a printer, sound output, transmission to an external device other than terminal device 2, storage on a recording medium, and handing over processing results to other processing devices or other programs.

[0102] The document output unit 141 outputs the second document, which is the document after the masking unit 133 has performed the masking process.

[0103] Various types of information are stored in the terminal storage unit 21 included in the terminal device 2. The various types of information are, for example, a user identifier or a document.

[0104] The terminal reception unit 22 receives various instructions and information, such as documents or output instructions.

[0105] The means for inputting various instructions and information may be any means, such as a touch panel, keyboard, mouse, or menu screen.

[0106] The device processing unit 23 performs various types of processing, such as processing to change instructions, information, etc. received by the device receiving unit 22 into instructions, information, etc. with a structure to be transmitted, and processing to change information received by the device receiving unit 25 into a structure to be output.

[0107] The terminal transmitting unit 24 transmits various instructions and information, such as documents and output instructions.

[0108] The terminal receiving unit 25 receives various instructions and information from the document processing apparatus 1. The various instructions and information are, for example, masked documents.

[0109] The terminal output unit 26 outputs various types of information, such as a masked document.

[0110] The storage unit 11, the rule management unit 111, the user management unit 112, and the terminal storage unit 21 are preferably non-volatile recording media, but can also be realized by volatile recording media.

[0111] There is no restriction on the process by which information is stored in the storage unit 11 etc. For example, information may be stored in the storage unit 11 etc. via a recording medium, information transmitted via a communication line etc. may be stored in the storage unit 11 etc., or information input via an input device may be stored in the storage unit 11 etc.

[0112] The reception unit 12 and the document reception unit 121 are preferably realized by wireless or wired communication means, but may also be realized by a means for receiving broadcasts, a device driver for an input means such as a touch panel or keyboard, or control software for a menu screen.

[0113] The processing unit 13, confidential information acquisition unit 131, alternative acquisition unit 132, masking unit 133, disclosure acquisition means 1311, rule acquisition means 1312, and confidential information acquisition means 1313 can usually be realized by a processor, memory, etc. The processing procedures of the processing unit 13, etc. are usually realized by software, and the software is recorded on a recording medium such as a ROM. However, they may also be realized by hardware (dedicated circuit). The processor may be a CPU, MPU, GPU, etc., and the type does not matter.

[0114] The output unit 14 and the document output unit 141 are usually realized by wireless or wired communication means, but may also be realized by driver software for an output device such as a display or speaker, or by a combination of driver software for an output device and the output device.

[0115] The terminal reception unit 22 can be realized by a device driver for an input means such as a touch panel or a keyboard, or control software for a menu screen.

[0116] The terminal transmitting unit 24 is usually realized by a wireless or wired communication means, but may also be realized by a broadcasting means.

[0117] The terminal receiving unit 25 is usually realized by a wireless or wired communication means, but may also be realized by a means for receiving broadcasts.

[0118] The terminal output unit 26 may or may not include an output device such as a display, a speaker, etc. The terminal output unit 26 may be realized by driver software for an output device, or by a combination of driver software for an output device and the output device, etc.

[0119] Next, an example of the operation of the document processing apparatus 1 will be described with reference to the flowchart of Fig. 4. Note that the document processing apparatus 1 will be described here when it is a server.

[0120] (Step S401) The document receiving unit 121 determines whether or not a document has been received from the terminal device 2. If a document has been received, the process proceeds to step S402; if not, the process proceeds to step S413. The received document is associated with, for example, a user identifier and one or more document attribute values. Such document attribute values ​​include, for example, a document location attribute value that specifies where the document is stored (for example, a folder, or a device, or a device and a folder). The received document may also be associated with, for example, public information or information for obtaining public information (for example, a URL, search conditions).

[0121] (Step S402) The processing unit 13 stores the document in a location specified by the document location attribute value associated with the document. Note that this document is the first document before the masking process.

[0122] (Step S403) The processing unit 13 performs a masking determination process. An example of the masking determination process will be described with reference to the flowchart in Fig. 5. The masking determination process is a process for determining whether or not to perform masking.

[0123] (Step S404) If the determination result in step S403 is "mask," the process proceeds to step S405; if the determination result is "do not mask," the process returns to step S401.

[0124] (Step S405) The confidential information acquisition unit 131 acquires one or more document attribute values ​​associated with the first document received in step S401. The confidential information acquisition unit 131 also acquires a user identifier associated with the document received in step S401, and acquires one or more user attribute values ​​paired with the user identifier from the user management unit 112. Note that these user attribute values ​​are also document attribute values.

[0125] (Step S406) The confidential information acquisition unit 131 acquires one or more masking rules associated with the one or more document attribute values ​​acquired in step S405 from the rule management unit 111. An example of such rule acquisition processing will be described with reference to the flowchart in FIG.

[0126] (Step S407) The processing unit 13 assigns 1 to the counter i.

[0127] (Step S408) The processing unit 13 determines whether the i-th unit of inspection exists in the received first document. If the i-th unit of inspection exists, the process proceeds to step S409; if not, the process proceeds to step S412.

[0128] The inspection unit is a unit of information for which confidential information is acquired and masked. The inspection unit is, for example, a page, a block of text or an image, or an object in a document. The inspection unit may also be the entire document.

[0129] (Step S409) The processing unit 13 acquires the i-th unit of inspection from the received first document.

[0130] (Step S410) The processing unit 13 performs masking processing on the i-th unit of inspection. An example of the masking processing will be described with reference to the flowchart of FIG.

[0131] The masking process is a process of detecting confidential information in the first document and masking the confidential information. For example, the masking process is a process of detecting confidential information in the first document and replacing the confidential information with alternative information.

[0132] (Step S411) The processing unit 13 increments the counter i by 1. The process returns to step S408.

[0133] (Step S412) The processing unit 13 stores the second document, which is a document obtained by masking the first document. The process returns to step S401.

[0134] It is preferable that the processing unit 13 stores the second document in association with the first document.

[0135] (Step S413) The reception unit 12 determines whether or not an output instruction has been received from the terminal device 2. If an output instruction has been received, the process proceeds to step S414, and if not, the process proceeds to step S416.

[0136] In addition, the output

[0137] (Step S414) The processing unit 13 acquires the document specified by the output instruction from the document management unit 113. An example of such document selection processing will be described with reference to the flowchart of FIG.

[0138] (Step S415) The document output unit 141 transmits the document acquired in step S414 to the terminal device 2. The process returns to step S401.

[0139] (Step S416) The reception unit 12 determines whether or not information has been received from the terminal device 2. If information has been received, the process proceeds to step S417, and if information has not been received, the process returns to step S401.

[0140] The information received here includes, for example, masking rules and user information.

[0141] (Step S417) The processing unit 13 stores the received information in the storage unit 11. The process returns to step S401.

[0142] Although the masking determination process is performed in step S403 of the flowchart in FIG. 4, the masking process may be performed on all received documents without performing the masking determination process.

[0143] In the flowchart of FIG. 4, the process ends when the power is turned off or an interrupt occurs to end the process.

[0144] Next, an example of the masking determination process in step S403 will be described with reference to the flowchart in FIG.

[0145] (Step S501) The confidential information obtaining unit 131 obtains the masking conditions from the storage unit 11.

[0146] (Step S502) The confidential information acquisition unit 131 acquires one or more document attribute values ​​of the received document. These document attribute values ​​are used to determine whether the masking conditions are satisfied.

[0147] (Step S503) The confidential information acquisition unit 131 determines whether or not one or more document attribute values ​​acquired in step S502 satisfy the masking condition. If the masking condition is satisfied, the process proceeds to step S504; if not, the process proceeds to step S505.

[0148] (Step S504) The confidential information obtaining unit 131 assigns "masking (for example, "1")" to the determination result, and returns to the upper level processing.

[0149] (Step S505) The confidential information obtaining unit 131 assigns "no masking (for example, "0")" to the determination result, and returns to the upper level processing.

[0150] Next, an example of the rule acquisition process in step S406 will be described with reference to the flowchart in FIG.

[0151] (Step S601) The rule acquisition unit 1312 acquires one or more document attribute values ​​of the received first document.

[0152] (Step S602) The rule acquiring means 1312 assigns 1 to a counter i.

[0153] (Step S603) Rule acquisition means 1312 determines whether or not the i-th masking rule exists in rule management unit 111. If the i-th masking rule exists, the process proceeds to step S604; if not, the process returns to the upper level process.

[0154] (Step S604) The rule acquisition unit 1312 acquires the document attribute condition associated with the i-th masking rule from the rule management unit 111.

[0155] (Step S605) The rule acquisition means 1312 determines whether or not the one or more document attribute values ​​acquired in step S601 satisfy the document attribute condition acquired in step S604. If the one or more document attribute values ​​satisfy the document attribute condition, the process proceeds to step S606; if not, the process proceeds to step S607.

[0156] (Step S606) The rule acquisition means 1312 acquires the i-th masking rule and temporarily stores it in a buffer (not shown).

[0157] (Step S607) The rule acquiring means 1312 increments the counter i by 1. The process returns to step S603.

[0158] Next, an example of the masking process in step S410 will be described with reference to the flowchart in FIG.

[0159] (Step S701) The confidential information obtaining unit 131 assigns 1 to a counter i.

[0160] (Step S702) The rule acquisition means 1312 determines whether or not the i-th masking rule exists among the masking rules temporarily stored in a buffer (not shown). If the i-th masking rule exists, the process proceeds to step S703; if not, the process returns to the upper process.

[0161] (Step S703) The confidentiality obtaining means 1313 obtains the confidentiality condition of the i-th blocking rule.

[0162] (Step S704) The confidential information acquisition means 1313 acquires one or more pieces of confidential information that are partial information that satisfy the confidentiality conditions from the inspection unit to be processed. An example of such confidential information acquisition processing will be described using the flowchart in Fig. 8. Note that there may be cases where confidential information cannot be acquired here.

[0163] (Step S705) The alternative acquisition unit 132 acquires an alternative definition that the i-th matching rule has.

[0164] (Step S706) The alternative acquisition unit 132 assigns 1 to a counter j.

[0165] (Step S707) The alternative acquisition unit 132 determines whether or not the j-th confidential information exists among the confidential information acquired in step S704. If the j-th confidential information exists, the process proceeds to step S708, and if not, the process proceeds to step S711.

[0166] (Step S708) The substitution acquisition unit 132 acquires substitution information using the substitution definition acquired in step S705. An example of such substitution acquisition processing will be described with reference to the flowchart in Fig. 10. Note that the substitution acquisition processing is processing for acquiring substitution information. In addition, the substitution information may be the substitution definition itself.

[0167] (Step S709) The replacement acquisition unit 132 places the replacement information acquired in step S708 at the j-th confidential information location in the first document, that is, replaces the confidential information with the replacement information.

[0168] (Step S710) The alternative acquisition unit 132 increments the counter j by 1. The process returns to step S707.

[0169] (Step S711) The confidential information obtaining unit 131 increments the counter i by 1. The process returns to step S702.

[0170] Next, an example of the confidential information acquisition process in step S704 will be described with reference to the flowchart in FIG.

[0171] (Step S801) The secret acquisition means 1313 determines whether the acquired secret conditions indicate the use of the generated AI. If the generated AI is to be used, the process proceeds to step S802; if not, the process proceeds to step S810.

[0172] In addition, the confidentiality acquisition means 1313 determines to use the generated AI if, for example, the rule management unit 111 has a confidential information acquisition prompt that is paired with the acquired confidentiality condition, and determines not to use the generated AI if the confidential information acquisition prompt does not exist.

[0173] (Step S802) The confidential information acquisition means 1313 determines whether or not the public information use conditions are satisfied. If the public information use conditions are satisfied, the process proceeds to step S803, and if not, the process proceeds to step S805.

[0174] The public information usage conditions are conditions for using public information when acquiring confidential information. For example, the public information usage conditions are that the public information is associated with the received first document. For example, the public information usage conditions are that the acquired masking rule has a public information usage flag or public information.

[0175] (Step S803) The public information acquisition means 1311 acquires public information.

[0176] (Step S804) The secret information acquisition means 1313 provides the public information acquired in step S803 to the generation AI. The secret information acquisition means 1313 usually provides the public information to the generation AI together with information indicating that the public information is public information.

[0177] (Step S805) The secret information obtaining means 1313 obtains the target inspection unit, and provides the inspection unit to the generating AI.

[0178] (Step S806) The confidential information acquisition means 1313 uses the confidentiality conditions to construct a confidential information acquisition prompt.

[0179] The confidential information acquisition means 1313 acquires, for example, a confidential information acquisition prompt template from the storage unit 11. Next, the confidential information acquisition means 1313, for example, replaces variables of the confidential information acquisition prompt template with confidentiality conditions to construct a prompt. The confidential information acquisition prompt template is, for example, "Please acquire confidential information that satisfies the following confidentiality conditions from the information below. At that time, please also associate location specifying information (page number and offset) that specifies the location where the confidential information exists with the confidential information to acquire the confidential information. Also, if public information is given, do not include information in the public information in the confidential information. [Information] <Inspection unit> [Confidentiality conditions] <Confidentiality conditions>." The inspection unit to be processed is substituted for the variable <Inspection unit> of the confidential information acquisition prompt template, and the acquired confidentiality conditions are substituted for the variable <Confidentiality conditions>.

[0180] (Step S807) The confidential information acquisition means 1313 gives a confidential information acquisition prompt to the generation AI.

[0181] (Step S808) The secret acquisition means 1313 determines whether or not an answer has been acquired from the generated AI. If an answer has been acquired, the process proceeds to step S809, and if an answer has not been acquired, the process returns to step S808.

[0182] (Step S809) The secret information acquisition means 1313 acquires one or more pieces of secret information from the response acquired in step S808. The secret information acquisition means 1313 also acquires location specifying information for one or more pieces of secret information. The secret information acquisition means 1313 then acquires one or more pairs of secret information and location specifying information, and temporarily stores them in a buffer (not shown). The process returns to the upper level process.

[0183] Here, the confidential information acquisition means 1313 may not be able to acquire the set of confidential information and location specifying information.

[0184] (Step S810) The confidential information acquisition means 1313 determines whether or not the inspection unit includes an image, etc. If the inspection unit includes an image, etc., the process proceeds to step S811, and if not, the process proceeds to step S816. Note that the image, etc. refers to image or audio information.

[0185] (Step S811) The confidential information acquisition means 1313 acquires one or more images etc. in the inspection unit.

[0186] (Step S812) The confidentiality acquisition means 1313 determines whether one or more images etc. satisfy the confidentiality conditions. If the images etc. satisfy the confidentiality conditions, the process proceeds to step S813, and if the images etc. do not satisfy the confidentiality conditions, the process proceeds to step S814.

[0187] (Step S813) The confidential information acquisition means 1313 acquires one or more confidential information items, such as images that satisfy confidentiality conditions. The confidential information acquisition means 1313 also acquires location identification information for each of the one or more confidential information items. The confidential information acquisition means 1313 then acquires one or more pairs of confidential information and location identification information, and temporarily stores them in a buffer (not shown).

[0188] (Step S814) The confidential information acquisition means 1313 judges whether or not the inspection unit contains a character string. If the inspection unit contains a character string, the process proceeds to step S815, and if not, the process returns to the upper processing.

[0189] (Step S815) The confidential information obtaining means 1313 obtains a character string from the inspection unit.

[0190] (Step S816) The confidential information acquisition means 1313 performs a character string inspection process and returns to the upper level process. An example of the character string inspection process will be described with reference to the flowchart in FIG.

[0191] In the flowchart of FIG. 8, if two or more images are acquired in step S811, it is preferable that the processes of steps S812 and S813 are performed for each of the two or more images.

[0192] Furthermore, in the flowchart of FIG. 8, if two or more character strings are acquired in step S815, it is preferable that the processing of step S816 be performed for each of the two or more character strings.

[0193] Next, an example of the character string inspection process in step S816 will be described with reference to the flowchart in FIG.

[0194] (Step S901) The confidential information acquisition means 1313 performs morphological analysis on the character string acquired in step S815 to acquire one or more morphemes.

[0195] (Step S902) The secret information acquiring means 1313 assigns 1 to the counter i.

[0196] (Step S903) The confidential information acquisition means 1313 determines whether the i-th independent word exists among the morphemes acquired in step S901. If the i-th independent word exists, the process proceeds to step S904; if not, the process returns to the upper process.

[0197] (Step S904) The security acquisition means 1313 judges whether the i-th independent word satisfies the security condition. If it does, the process proceeds to step S905, and if it does not, the process proceeds to step S906.

[0198] The confidentiality acquisition means 1313 determines, for example, whether the i-th independent word is a character string of a specific class that is a confidentiality condition.

[0199] (Step S905) The confidential information acquisition means 1313 acquires confidential information that is an independent word that satisfies the confidentiality conditions. The confidential information acquisition means 1313 also acquires location specifying information of the confidential information. The confidential information acquisition means 1313 then acquires a pair of confidential information and location specifying information, and temporarily stores the pair in a buffer (not shown).

[0200] (Step S906) The secret information acquiring means 1313 increments the counter i by 1. The process returns to step S903.

[0201] Next, an example of the alternative acquisition process in step S708 will be described with reference to the flowchart in FIG.

[0202] (Step S1001) The alternative acquisition unit 132 acquires the j-th confidential information of step S707.

[0203] (Step S1002) The substitution acquisition unit 132 determines whether the substitution definition acquired in step S705 is "higher-level information." If the substitution definition is "higher-level information," the process proceeds to step S1003; if not, the process proceeds to step S1004.

[0204] (Step S1003) The alternative acquisition unit 132 acquires higher-level information of the confidential information acquired in step S1001, temporarily stores it in a buffer (not shown), and returns to the higher-level processing.

[0205] (Step S1004) The alternative acquisition unit 132 determines whether the alternative definition acquired in step S705 is "fixed-length information." If the alternative definition is "fixed-length information," the process proceeds to step S1005; if not, the process proceeds to step S1006.

[0206] (Step S1005) The alternative acquisition unit 132 acquires fixed-length information (for example, "*****") from the rule management unit 111 and temporarily stores it in a buffer (not shown). The process returns to the upper level process. It is preferable that the fixed-length information is determined in advance. However, the alternative acquisition unit 132 may acquire different fixed-length information depending on, for example, the class of confidential information. For example, the alternative acquisition unit 132 acquires fixed-length information "*****" for confidential information of a company name, and fixed-length information "XXXX" for confidential information of a person's name.

[0207] (Step S1006) The alternative acquisition unit 132 determines whether the alternative definition acquired in step S705 is "ambiguous numerical information." If the alternative definition is "ambiguous numerical information," the process proceeds to step S1007; if not, the process proceeds to step S1008.

[0208] (Step S1007) The alternative acquisition unit 132 acquires ambiguous numerical information using the confidential information acquired in step S1001, and temporarily stores the ambiguous numerical information in a buffer (not shown). The process returns to the upper level process.

[0209] (Step S1008) The substitution acquisition unit 132 determines whether the substitution definition acquired in step S705 is "explanatory information." If the substitution definition is "explanatory information," the process proceeds to step S1009; if not, the process proceeds to step S1010.

[0210] (Step S1009) The alternative acquisition unit 132 acquires explanatory information for the confidential information acquired in step S1001, temporarily stores the information in a buffer (not shown), and returns to the upper level process.

[0211] (Step S1010) The substitution acquisition unit 132 determines whether the substitution definition acquired in step S705 is "summary information." If the substitution definition is "summary information," the process proceeds to step S1003; if not, the process proceeds to step S1004.

[0212] (Step S1011) The alternative acquisition unit 132 acquires summary information of the confidential information acquired in step S1001, temporarily stores the summary information in a buffer (not shown), and returns to the upper level processing.

[0213] (Step S1012) The substitution acquisition unit 132 acquires substitution information according to another substitution definition, temporarily stores the information in a buffer (not shown), and returns to the upper level processing.

[0214] Next, an example of the document selection process in step S414 will be described with reference to the flowchart in FIG.

[0215] (Step S1101) The processing unit 13 acquires the document identifier included in the output instruction.

[0216] (Step S1102) The processing unit 13 acquires, from the document management unit 113, one or more document attribute values ​​of the document identified by the document identifier.

[0217] (Step S1103) The processing unit 13 acquires the user identifier included in the output instruction.

[0218] (Step S1104) The processing unit 13 acquires from the user management unit 112 one or more user attribute values ​​paired with the user identifier.

[0219] (Step S1105) The processing unit 13 determines whether one or more document attribute values ​​and / or one or more user attribute values ​​satisfy the non-masking condition. If the non-masking condition is satisfied, the process proceeds to step S1106, and if not, the process proceeds to step S1107.

[0220] The unmasking condition is a condition for obtaining a first document that has not been masked. The unmasking condition is, for example, a condition regarding one or more document attribute values, or a condition regarding one or more user attribute values, or a condition regarding one or more document attribute values ​​and one or more user attribute values. The unmasking condition is, for example, that the user identifier corresponding to the output instruction matches the user identifier of the user who created the document. The unmasking condition is, for example, that the type identifier of the document corresponding to the output instruction is a specific type identifier (e.g., "internal material").

[0221] (Step S1106) The processing unit 13 acquires the non-masking document (first document) that is paired with the document identifier, and returns to the upper processing.

[0222] (Step S1107) The processing unit 13 acquires the masking document (second document) that is paired with the document identifier, and returns to the upper processing.

[0223] In the flowchart of FIG. 11, the processing unit 13 may always acquire the second document without determining whether the non-masking condition is met.

[0224] Next, an example of the operation of the terminal device 2 will be described with reference to the flowchart of FIG.

[0225] (Step S1201) The terminal reception unit 22 determines whether or not a document has been received. If the document has been received, the process proceeds to step S1202, and if not, the process proceeds to step S1204. Note that the document here is the first document.

[0226] (Step S1202) The device processing unit 23 acquires a user identifier from the device storage unit 21.

[0227] (Step S1203) The terminal transmitting unit 24 transmits the document accepted in step S1201 in association with the user identifier acquired in step S1202 to the document processing apparatus 1. The process returns to step S1201.

[0228] (Step S1204) The terminal reception unit 22 determines whether or not an output instruction has been received. If an output instruction has been received, the process proceeds to step S1205, and if not, the process proceeds to step S1207.

[0229] (Step S1205) The device processing unit 23 acquires the user identifier from the device storage unit 21.

[0230] (Step S1206) The terminal transmitting unit 24 transmits the output instruction accepted in step S1204 in association with the user identifier acquired in step S1205 to the document processing apparatus 1. The process returns to step S1201.

[0231] (Step S1207) The terminal reception unit 22 determines whether or not the information has been received. If the information has been received, the process proceeds to step S1208, and if the information has not been received, the process returns to step S1201. The information here may be, for example, a masking rule or user information.

[0232] (Step S1208) The terminal transmitting unit 24 transmits the information received in step S1207 to the document processing apparatus 1. The process returns to step S1201.

[0233] (Step S1209) The terminal receiving unit 25 determines whether or not the information has been received. If the information has been received, the process proceeds to step S1210, and if the information has not been received, the process returns to step S1201.

[0234] (Step S1210) The device processing unit 23 uses the information received in step S1209 to compose information to be output. The terminal output unit 26 outputs the information. Return to step S1201. The information is, for example, the first document or the second document.

[0235] In the flowchart of FIG. 12, the process ends when the power is turned off or an interrupt occurs to end the process.

[0236] A specific example of the operation of the document processing system A according to this embodiment will now be described.

[0237] The rule management unit 111 of the document processing device 1 stores a rule management table shown in FIG. 13. The rule management table is a table that manages one or more masking rules. The rule management table is a table that manages one or more records that have an "ID," "confidentiality condition," and "alternative definition." The "confidentiality condition" has a "data type condition," a "class condition," and a "confidential information acquisition prompt." The "alternative definition" has an "alternative type" and a "alternative information acquisition prompt." The "ID" is information that identifies a record.

[0238] "Data type condition" is the condition for the data type to be classified as confidential information. "Class condition" is the condition for the class to be classified as confidential information. "Confidential information acquisition prompt" is the prompt used when acquiring confidential information. If "Confidential information acquisition prompt" is "-(NULL)", for example, confidential information is acquired by natural language processing without using generation AI. "Alternative type" is the type of alternative information. "Alternative type" can be the alternative information itself. "Alternative information acquisition prompt" is the prompt used when acquiring alternative information.

[0239] The rule management unit 111 also stores a document attribute value condition management table shown in Fig. 14. The document attribute value condition management table is a table that manages masking rules for one or more document attribute value conditions. The document attribute value condition management table is a table that manages document attribute value conditions of documents to which masking rules are applied. In other words, a masking rule that pairs with the document attribute value condition is applied to a first document that corresponds to a document attribute value that satisfies the document attribute value condition, masking processing is performed, and a second document is obtained.

[0240] The document attribute value condition management table has an "ID," "document attribute value condition," and "rule identifier." A record with "ID=1" indicates that a masking rule with rule identifier "1" is applied to a document uploaded to the document processing device 1 by a user belonging to the organization identifier "Company A" or a document created by a user belonging to the organization identifier "Company A," thereby masking the document. A record with "ID=2" indicates that a masking rule with rule identifier "2" is applied to a document uploaded to the document processing device 1 by a user belonging to the organization identifier "Company A.A1 Department" (the "A1 Department" of Company A) or a document created by a user belonging to the organization identifier "Company A.A1 Department," thereby masking the document. A record with "ID=3" indicates that a masking rule with rule identifier "3" is applied to a document uploaded to the document processing device 1 by a user identified by user identifier "AU01," or a document created by a user identified by user identifier "AU01," thereby masking the document. In other words, masking rules are managed in association with the organizational identifiers of each hierarchical layer, and the masking rules of each layer in the hierarchical organization to which a user belongs are applied to documents uploaded by the user or created by the user. The record with "ID=4" indicates that if the type identifier indicating the document type is "Proposal," then the masking rules with rule identifiers "1" and "4" are applied to mask the document. The record with "ID=6" indicates that if the folder in which the document is located is "\root\...\Folder X," then the masking rule with rule identifier "6" is applied to mask the document. The record with "ID=7" indicates that the document attribute value condition is NULL, so the masking rule with rule identifier "7" is applied to all documents to be masked that meet the masking condition, and thus the document is masked.

[0241] It is assumed that the user management unit 112 stores a user management table shown in FIG. 15. The user management table is a table for managing user information. The user management table has "ID," "user identifier," "organization identifier," "job title," "years of employment," "age," and "gender." The "organization identifier" has a "company identifier," "division identifier," and "department identifier." If the "division identifier" is "-," it indicates that no division exists.

[0242] The document management unit 113 stores a document management table shown in FIG. 16. The document management table is a table for managing documents. The document management table is a table for managing one or more records having an "ID," "user identifier," "first document," "second document," "document placement attribute value," and "type identifier." The "first document" is the file name of the first document, which is the document before the masking process. The "second document" is the file name of the second document, which is the document after the masking process. The entity (file) of the first document and the entity (file) of the second document can be accessed by file name, and are stored in the document management unit 113, for example.

[0243] Furthermore, it is assumed that the masking condition "type identifier != internal material" is stored in the storage unit 11. Note that "!=" indicates no match.

[0244] In the above situation, assume that user U, identified by user identifier "AU01", inputs document "Proposal P1.pptx" and document location attribute value "\root\···\Folder X" into terminal device 2. Note that document "Proposal P1.pptx" is a document that corresponds to the type identified by type identifier "Proposal". Furthermore, the document "Proposal P1.pptx" includes the name of the proposed company (Company C), which is a company name other than the user's own company "Company A", the personal names of Company C's employees, Company C's telephone numbers, and Company C's employees' email addresses. Furthermore, the document "Proposal P1.pptx" includes the price of the proposed product. Furthermore, the document "Proposal P1.pptx" includes Company C's company logo.

[0245] Then, the terminal receiving unit 22 of the terminal device 2 receives the document "Proposal P1.pptx" and the document location attribute value "\root\···\Folder X". Also, it is assumed that the terminal processing unit 23 acquires a user identifier "UA01" from the terminal storage unit 21. Next, the terminal transmitting unit 24 pairs the user identifier "UA01" with the document location attribute value "\root\···\Folder X" and transmits the document to the document processing device 1.

[0246] Next, the document receiving unit 121 of the document processing device 1 receives the document "Proposal P1.pptx" that is paired with the user identifier "UA01" and the document placement attribute value "\root\···\Folder X" from the terminal device 2. Next, the processing unit 13 stores the document in the folder "\root\···\Folder X" in the document processing system A.

[0247] Next, the processing unit 13 passes the document to the generation AI. The processing unit 13 also obtains a prompt, "Tell me the type of this document! The document type is, for example, 'proposal' or 'internal document'," from the storage unit 11, and passes the prompt to the generation AI. The processing unit 13 then obtains the type identifier "proposal," which is the answer, from the generation AI.

[0248] Next, the processing unit 13 determines that the type identifier “proposal” satisfies the masking condition “type identifier !=internal document.” In other words, the processing unit 13 determines that the document is to be masked.

[0249] Next, the confidential information acquisition unit 131 acquires the document attribute values ​​of the document. Here, it is assumed that the confidential information acquisition unit 131 acquires a set of document attribute values ​​"<Type Identifier> Proposal <User Identifier> AU01 <Organization Identifier.Company Identifier> Company A <Organization Identifier.Department Identifier> A1 Division <Document Location Attribute Value> \root\···\Folder X···".

[0250] Next, the confidentiality acquisition unit 131 uses the acquired document attribute values ​​to acquire six masking rules of "ID=1, 2, 3, 4, 6, 7" from the rule management table (FIG. 14).

[0251] Next, according to the flowchart of FIG. 7, the confidential information acquisition unit 131 provides the document "Proposal P1.pptx" and a confidential information acquisition prompt for "ID=1" to the generation AI in accordance with the confidentiality conditions of the masking rule "ID=1," and acquires company names, personal names, telephone numbers, and email addresses excluding the company's own company "Company A," paired with their location-specific information. The confidential information acquisition unit 131 also performs morphological analysis on the document "Proposal P1.pptx" in accordance with the confidentiality conditions of the masking rule "ID=2," and acquires a string indicating an amount, which is an independent word that belongs to the "amount" class among the acquired independent words. The confidential information acquisition unit 131 also provides the document "Proposal P1.pptx" and a confidential information acquisition prompt for "ID=3" to the generation AI in accordance with the confidentiality conditions of the masking rule "ID=3," and acquires an image of the company logo of a company other than the company's own company. The confidential information acquisition unit 131 then provides the generation AI with a confidential information acquisition prompt for the document "Proposal P1.pptx" and "ID=4" in accordance with the confidentiality conditions of the masking rule for "ID=4," and acquires the proposal content from the document "Proposal P1.pptx." The confidential information acquisition unit 131 then acquires confidential information in accordance with the confidentiality conditions of the masking rules for "ID=6" and "ID=7."

[0252] Next, the substitution acquisition unit 132 uses the substitution definition (FIG. 13) of the masking rule "ID=1" to acquire higher-level information (e.g., "electricity manufacturer" or "rival company") that is substitute information for the confidential information that is one or more company names that have been acquired. The substitution acquisition unit 132 provides the generation AI with a substitution information acquisition prompt that pairs with the "higher-level information" of "ID=1" in the rule management table (FIG. 13) and the confidential information, and acquires the higher-level information of the confidential information that is one or more company names.

[0253] Then, the masking unit 133 replaces each of the one or more pieces of acquired substitute information with confidential information corresponding to the substitute information.

[0254] Furthermore, the substitution acquisition unit 132 uses the substitution definition (FIG. 13) of the masking rule "ID=1" to acquire fixed-length information (see the substitution type of "ID=1" in FIG. 13) that is substitution information for one or more acquired personal names, telephone numbers, and email addresses. Then, the masking unit 133 replaces each of the acquired one or more pieces of substitution information with confidential information corresponding to the substitution information.

[0255] Furthermore, the substitution acquisition unit 132 uses the substitution definition (FIG. 13) of the masking rule "ID=2" to provide each of the acquired one or more amounts to the generation AI, and provides a substitution information acquisition prompt for "ID=2" to the generation AI to acquire substitution information that is numerical range information corresponding to each of the one or more amounts. Then, the masking unit 133 replaces each of the acquired one or more pieces of substitution information with the amount corresponding to the substitution information.

[0256] Furthermore, the substitution acquisition unit 132 provides each of the acquired one or more company logos (images) to the generation AI using the substitution definition (FIG. 13) of the masking rule "ID=3," and provides a substitution information acquisition prompt for "ID=3" to the generation AI to acquire explanatory information for each of the one or more company logos. Then, the masking unit 133 replaces each of the acquired explanatory information with the company logo corresponding to the substitution information.

[0257] The alternative acquisition unit 132 then uses the alternative definition (FIG. 13) of the masking rule "ID=4" to provide the acquired proposal content to the generation AI, and provides an alternative information acquisition prompt for "ID=4" to the generation AI to acquire summary information of the proposal content. The masking unit 133 then replaces the acquired summary information with the proposal content corresponding to the alternative information.

[0258] Furthermore, the alternative acquisition unit 132 and the masking unit 133 perform masking processing in accordance with the masking rules for "ID=6" and "ID=7."

[0259] Through the above process, a second document is constructed by masking the received first document. Then, the processing unit 13 stores the second document in the folder "\root\...\Folder X" of the document management unit 113, naming it "Proposal P1M.pptx." Note that the processing unit 13 acquires, for example, the file name of the second document, which is the file name of the first document with "M" added to the end of the file name itself.

[0260] As described above, according to this embodiment, confidential information in a document can be masked so that it cannot be guessed by not masking the confidential information but maintaining its length. For example, confidential information in a document can be masked so that it cannot be guessed by replacing the confidential information with fixed-length information.

[0261] Furthermore, according to this embodiment, confidential information can be hidden by masking it with higher-level information, and a document that is easy to read can be provided.

[0262] Furthermore, according to this embodiment, confidential information that is a numerical value is masked with the fuzzy numerical information, thereby making it possible to conceal the confidential information and provide a document that is naturally readable.

[0263] Furthermore, according to this embodiment, confidential information in the form of an image can be masked with its explanatory information.

[0264] Furthermore, according to this embodiment, confidential information can be hidden by masking it with its summary information, and a document that is easy to read can be provided.

[0265] Furthermore, according to this embodiment, confidential information can be appropriately detected, and as a result, unnecessary masking can be reduced.

[0266] Furthermore, according to this embodiment, masking can be performed based on rules according to the attribute values ​​of the document.

[0267] Furthermore, according to this embodiment, masking can be performed based on rules according to user attribute values.

[0268] Furthermore, according to this embodiment, masking can be performed based on rules managed in pairs with the organization identifiers of the hierarchical organizations to which the users belong.

[0269] Furthermore, according to this embodiment, masking can be performed based on rules according to the location where the document is placed. In particular, according to this embodiment, masking can be performed based on rules according to the device or folder where the document is placed.

[0270] Furthermore, according to this embodiment, masking can be performed based on rules according to the type of document.

[0271] The document processing device 1 in this embodiment may also function as a terminal. In such a case, the reception unit 12 and the document reception unit 121 typically receive information from a user. The output unit 14 and the document output unit 141 typically display information or output information as audio. A block diagram of the document processing device 1 in such a case is also shown in FIG. 3.

[0272] The processes in this embodiment may be implemented by software. This software may be distributed by software download or the like. This software may also be recorded on a recording medium such as a CD-ROM and distributed. This also applies to the other embodiments in this specification. The software that implements the document processing device 1 in this embodiment is the following program. In other words, this program causes a computer to function as a document receiving unit that receives a first document containing two or more pieces of partial information; a confidentiality acquisition unit that acquires confidential information, which is partial information that meets confidentiality conditions, which are conditions for determining confidential information, from the first document received by the document receiving unit; a substitution acquisition unit that acquires alternative information, which is higher-level information that is a higher-level concept of the confidential information, or fixed-length information that is independent of the data length of the confidential information, or ambiguous numerical information that obscures the numerical value when the confidential information is a number, or explanatory information that explains the image when the confidential information is an image, or summary information that summarizes the confidential information; a masking unit that performs a masking process to replace the confidential information in the first document with the alternative information acquired by the substitution acquisition unit; and a document output unit that outputs the second document after the masking process has been performed by the masking unit.

[0273] Furthermore, software for realizing the document processing apparatus 1 according to the present embodiment is the following program. That is, this program causes a computer that can access a rule management unit that stores masking rules, each including a confidentiality condition that is a condition for determining confidential information in association with one or more document attribute value conditions, which are conditions related to one or more document attribute values ​​of a first document, to function as: a document reception unit that receives a first document including two or more pieces of partial information that are associated with one or more document attribute values; a rule acquisition unit that acquires one or more document attribute values ​​associated with the first document received by the document reception unit and acquires the masking rules that the one or more document attribute values ​​satisfy from the rule management unit; a confidentiality acquisition unit that acquires, from the first document, confidential information that is partial information that matches the confidentiality condition of the masking rule acquired by the rule acquisition unit; a masking unit that performs a masking process to mask the confidential information in the first document; and a document output unit that outputs the second document after the masking process has been performed by the masking unit.

[0274] FIG. 17 is a block diagram of a computer system 300 that executes the programs described in this specification to realize the document processing device 1 and the like according to the various embodiments described above.

[0275] In FIG. 17, a computer system 300 includes a computer 301 including a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.

[0276] 17, computer 301 includes, in addition to CD-ROM drive 3012, MPU 3013, bus 3014 connected to CD-ROM drive 3012 etc., ROM 3015 for storing programs such as a boot-up program, RAM 3016 connected to MPU 3013 for temporarily storing instructions of application programs and providing temporary storage space, and hard disk 3017 for storing application programs, system programs, and data. Although not shown here, computer 301 may further include a network card for providing connection to a LAN.

[0277] A program that causes the computer system 300 to execute the functions of the document processing device 1 of the above-described embodiment may be stored on a CD-ROM 3101, inserted into the CD-ROM drive 3012, and then transferred to the hard disk 3017. Alternatively, the program may be transmitted to the computer 301 via a network (not shown) and stored on the hard disk 3017. The program is loaded into the RAM 3016 when executed. The program may also be loaded directly from the CD-ROM 3101 or the network.

[0278] The program does not necessarily include an operating system (OS) or third-party programs that cause the computer 301 to execute the functions of the document processing device 1 of the above-described embodiment. The program need only include instructions that call appropriate functions (modules) in a controlled manner to achieve the desired results. How the computer system 300 operates is well known, and a detailed description thereof will be omitted.

[0279] In addition, in the above program, the steps of transmitting information and receiving information do not include processing performed by hardware, such as processing performed by a modem or interface card in the transmission step (processing that can only be performed by hardware).

[0280] The computer that executes the program may be a single computer or a plurality of computers, that is, it may perform centralized processing or distributed processing.

[0281] Furthermore, in each of the above embodiments, it goes without saying that two or more communication means present in one device may be physically realized by one medium.

[0282] Furthermore, in each of the above embodiments, each process may be realized by centralized processing in a single device, or may be realized by distributed processing in a plurality of devices.

[0283] The present invention is not limited to the above-described embodiment, and various modifications are possible, and it goes without saying that these modifications are also included within the scope of the present invention. [Industrial Applicability]

[0284] As described above, the document processing device according to the present invention has the effect of being able to mask confidential information in a document so that the confidential information cannot be guessed, and is useful as a document processing device or the like. [Explanation of symbols]

[0285] A Document Processing System 1 Document Processing Device 2. Terminal Device 11 Storage area 12 Reception 13 Processing section 14 Output section 21 Terminal storage section 22 Terminal Reception 23 Terminal processing section 24 Terminal transmitter 25 Terminal receiving section 26 Terminal Output Unit 111 Rules Management Department 112 User Management Department 113 Document Management Department 121 Document Reception Department 131 Confidential Acquisition Department 132 Alternative Acquisition Department 133 Masking Section 141 Document Output Unit 1311 Public Acquisition Means 1312 Rule Acquisition Method 1313 Confidential acquisition means

Claims

1. a rule management unit in which masking rules including confidentiality conditions for determining confidential information to be masked are stored in association with one or more document attribute value conditions, which are conditions related to one or more document attribute values ​​that are attribute values ​​of the first document; a document receiving unit that receives a first document including two or more pieces of partial information that correspond to one or more document attribute values ​​and that can be subject to masking; a rule acquisition unit that acquires document attribute values ​​that are attribute values ​​of the first document accepted by the document acceptance unit, and acquires the masking rules that the one or more document attribute values ​​satisfy from the rule management unit; confidentiality acquisition means for applying the masking rule acquired by the rule acquisition means to the first document and acquiring confidential information from the first document, the confidentiality information being partial information that matches the confidentiality condition included in the masking rule; a masking unit that performs a masking process to mask the confidential information in the first document; a document output unit that outputs the second document after the masking unit has performed the masking process.

2. the one or more document attribute values ​​that are attribute values ​​of the first document include one or more user attribute values ​​that are attribute values ​​of a user of the first document; The document processing device according to claim 1 , wherein the document attribute value conditions include a user attribute value condition that is a condition related to one or more user attribute values ​​of a user of the first document.

3. the one or more document attribute values ​​that are attribute values ​​of the first document include two or more user attribute values; the two or more user attribute values ​​include two or more hierarchical organization identifiers to which the user belongs; The rule management unit includes: Masking rules corresponding to organization identifiers at two or more levels are stored, The rule acquisition means The document processing apparatus according to claim 2 , wherein the document receiving unit obtains two or more organization identifiers corresponding to the first document received, and obtains the masking rules corresponding to the two or more organization identifiers from the rule management unit.

4. the one or more document attribute values ​​that are attribute values ​​of the first document include a document location attribute value related to a location where the first document is located; The one or more document attribute value conditions of the rule management unit include: The document processing device according to claim 1 , further comprising a document layout attribute value condition that is a condition related to a document layout attribute value of the first document.

5. the one or more document attribute values ​​include a device identifier that identifies a device on which the first document is located or a folder identifier that identifies a folder on which the first document is located; The one or more document attribute value conditions of the rule management unit include: The document processing device according to claim 4 , further comprising a device condition that is a condition related to a device identifier of the first document, or a folder condition that is a condition related to a folder identifier of the first document.

6. the one or more document attribute values ​​that are attribute values ​​of the first document include a type identifier that identifies a type of the document; The one or more document attribute value conditions of the rule management unit include: The document processing device according to claim 1 , further comprising a document type condition that is a condition related to a type identifier of the first document.

7. the first document is a file, 7. The document processing device according to claim 6, wherein the type identifier is information for identifying the type of file.

8. A document processing system comprising the document processing device according to any one of claims 1 to 7 and one or more terminal devices, the document receiving unit of the document processing device receives the first document from a terminal device; the document output unit of the document processing device transmits the second document to the terminal device; One or more of the terminal devices a terminal reception unit that receives a first document; a terminal transmitter that transmits the first document to the document processing device; a terminal receiving unit for receiving the second document from the document processing device; a terminal output unit that outputs the second document.

9. A document processing method including all processes performed by the document processing device according to any one of claims 1 to 7.

10. Computer, A program for causing the document processing device to function as the document processing device according to any one of claims 1 to 7.

Images