Generation device, generation method, and generation program

Abstract

To propose a generation device, a generation method, and a generation program that can easily generate proper sample data.SOLUTION: A generation device includes: an acquisition part for acquiring image data on a document containing input items; an analysis part for analyzing a structure of the document by inputting the image data acquired by the acquisition part to a first model; and a generation part for specifying input items of the document based on the structure analyzed by the analysis part, and generating sample data corresponding to the input items by inputting the contents of the input items to a second model.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a generating device, a generating method, and a generating program for generating sample data of a document or the like. [Background technology]

[0002] When a user prepares an application form or the like, it is useful to have sample data with the necessary information filled in, since the user can fill in or input the necessary information by referring to the sample.

[0003] As a technique for creating such sample data, a technique is known in which sample information and help information for a customer to fill out an application form as support information (for example, Patent Document 1). Also, a technique is known in which sample information can be flexibly presented according to the entry items by imaging a document and reading information in input fields (for example, Patent Document 2). Also, a technique is known in which the structure of a document can be recognized by analyzing the structure of an application form or the like using a learning model (for example, Patent Document 3). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 5482522 [Patent Document 2] JP 2018-112802 A [Patent Document 3] Patent Publication No. 2021-56796 Summary of the Invention [Problem to be solved by the invention]

[0005] Currently, many application forms and other documents are electronic, and users are often required to enter information into these electronic documents. However, there are a wide variety of application forms, such as official documents, and it is very time-consuming for the application receiving party to create sample data for all of them. On the other hand, if sample data does not exist, inexperienced users will not be able to enter information into the documents appropriately, which may cause delays in business operations or reduce usability.

[0006] Therefore, the present disclosure proposes a generating device, a generating method, and a generating program that can easily generate appropriate sample data. [Means for solving the problem]

[0007] In order to solve the above problems, one embodiment of a generation device according to the present disclosure is characterized in that it includes an acquisition unit that acquires image data of a document including input fields, an analysis unit that analyzes a structure of the document by inputting the image data acquired by the acquisition unit into a first model, and a generation unit that identifies input fields of the document based on the structure analyzed by the analysis unit and generates sample data corresponding to the input fields by inputting the contents of the input fields into a second model. Effect of the Invention

[0008] According to one aspect of the embodiment, appropriate sample data can be easily generated. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing an overview of a generation process according to an embodiment. [Diagram 2] FIG. 11 is a diagram for explaining internal processing of the generation process according to the embodiment. [Diagram 3] FIG. 1 is a diagram for explaining an example of a generation process according to an embodiment; [Figure 4] FIG. 13 is a diagram for explaining an example of the generation process according to the embodiment. [Diagram 5]FIG. 3 is a diagram for explaining an example of the generation process according to the embodiment. [Figure 6] FIG. 1 is a diagram showing an example of a user interface according to an embodiment. [Figure 7] FIG. 2 is a diagram showing an example of a user interface according to the embodiment. [Figure 8] FIG. 3 illustrates an example of a user interface according to the embodiment. [Figure 9] FIG. 2 is a diagram illustrating an example of the configuration of a generating device according to an embodiment. [Figure 10] 1 is a flowchart (1) showing a procedure of information processing according to the embodiment. [Figure 11] 11 is a flowchart (2) showing the procedure of information processing according to the embodiment. [Figure 12] FIG. 11 is a diagram illustrating an example of a generation process according to a modified example. [Figure 13] FIG. 2 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the generating device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the following embodiments, the same components are designated by the same reference numerals, and duplicated descriptions will be omitted.

[0011] (1. EMBODIMENT) (1-1. Overview of Generation Processing According to the Embodiment) The generation process according to the present disclosure is executed by a generation device 100 shown in Fig. 1. The generation device 100 is an information processing device that generates sample data related to various forms, such as application forms that require input by a user. For example, the generation device 100 is a PC (Personal Computer), a tablet terminal, a cloud server, or the like. Note that the generation device 100 may have a function for implementing OCR processing for converting forms into image data, or may be a device that cooperates with a dedicated OCR device.

[0012] 1, the generating device 100 generates sample data of the application form 10. First, the generating device 100 converts the application form 10 into image data using a scanning function. The generating device 100 also recognizes characters and the like included in the application form 10 using an OCR function.

[0013] At this time, the generating device 100 acquires structured data consisting of input item names in the application form and writing area information which is an area to enter information in. For example, the generating device 100 acquires structured data indicating the structure of the application form 10, such as rectangles corresponding to the input items of the application form 10 and the position where characters or the like are to be entered for each input item.

[0014] Specifically, the generating device 100 inputs image data of the application form 10 into a first learning model (hereinafter referred to as a "structure model") for generating structured data, and obtains the output structured data. The structure model is a learning model that uses samples of various forms and application forms as learning data, and is pre-trained to output layout information of the input items (coordinate data indicating the position of the input item in the document, etc.) and position information such as the position at which characters are input in the input item as structured data when image data of the form or the like is input. As the configuration and learning method of the structure model, for example, the technology disclosed by the same applicant of the present disclosure in the above-mentioned Patent Document 3 or various publicly-announced technologies may be used.

[0015] Furthermore, the generating device 100 acquires structured data of the application form 10 and also acquires character strings corresponding to each input item. In the example of Fig. 1, the generating device 100 acquires character strings corresponding to input items such as "application date", "department name", "name", and "telephone number". The generating device 100 also acquires character strings such as "management number", "software", and "reason for use". The generating device 100 also acquires character strings such as "head of department", "administrator", and "application form".

[0016] Then, the generating device 100 inputs each character string into a second learning model (hereinafter referred to as a "language model") that has previously learned the relationship between the input items of the form and the character strings input into the input items. The language model is generated in advance, for example, by re-learning a known large-scale language model for sample data of the form of the present disclosure. For example, when "application date" is input, the language model outputs sample characters and numbers (in this example, a specific date, etc.). Similarly, when "department name" is input, the language model outputs sample characters (in this example, a specific name such as "Research and Development Department, etc.)). This allows the generating device 100 to obtain each sample data to be entered into each input item.

[0017] The generating device 100 does not necessarily output only letters and numbers, but can also output image data, check marks, etc., depending on the learning result. For example, the generating device 100 can learn in advance that for an input item "head", image data corresponding to a seal is input to the input item. Therefore, using a trained language model, when the input item is input, the generating device 100 can output image data corresponding to a seal based on the input item "head" and a structure in which some data is written below the input item.

[0018] The language model may also learn attribute information regarding each input item. The attribute information is, for example, information indicating what type of characters (Chinese characters, Hiragana, Katakana, English letters, numbers, symbols, etc.) are to be input for each input item, how many digits the characters to be input have, and whether characters or image data such as a seal should be input. By learning such information, the generating device 100 can operate a function to check whether the input information is appropriate when the user inputs information into a form.

[0019] When the information of the application form 10 is analyzed by the two learning models as described above, the generating device 100 combines the structured data output from the structural model and the sample data output from the language model to generate a sample input application form 20. As shown in Fig. 1, the input application form 20 is sample data in which an appropriate character string or the like matching the input item is input for each input item.

[0020] In this way, the generating device 100 first analyzes the structure of the application form 10 using a structure model. Furthermore, the generating device 100 identifies input items from the structure, and outputs appropriate character strings, etc. to be input into the identified input items using a language model. The generating device 100 then reproduces the layout of the application form 10 based on the structured data and identifies the position where the sample data should be input, thereby generating an input application form 20 in which character strings, etc. are input with appropriate content and at appropriate positions. In other words, the generating device 100 can easily generate appropriate sample data by simply reading the application form as image data, without the trouble of manually creating sample data for each of a wide variety of application forms.

[0021] The flow of the generation process will be described in more detail with reference to Fig. 2. Fig. 2 is a diagram for explaining the internal processing of the generation process according to the embodiment.

[0022] First, the generating device 100 images the application form 10 (step S10). Then, the generating device 100 obtains structured data of the application form 10 using the structure model (step S11). As shown in data 30, the structured data includes, for example, input items and position information such as coordinates of the input items.

[0023] Next, the generating device 100 generates sample data (sample values) corresponding to the input items using the language model, and inserts the sample values ​​into the structured data (step S12). Data 40 shows an example of a data structure when the sample values ​​are inserted into the structured data. As shown in Fig. 2, the data 40 includes position information such as coordinates of the input items, and position information such as coordinates where the sample values ​​are input.

[0024] Then, the generating device 100 generates the inputted application form 20 based on the data 40. That is, the generating device 100 can generate sample image data in which sample data such as character strings are inserted into appropriate positions on the application form 10.

[0025] In addition, when generating sample image data from an unfilled form or application, the generating device 100 may perform the process shown in Fig. 3. An example of the generating process according to the embodiment will be described with reference to Fig. 3. Fig. 3 is a diagram (1) for explaining an example of the generating process according to the embodiment.

[0026] The generating device 100 images the application form 10, whose input item fields are blank (step S21), and acquires image data 11. The generating device 100 also separately prepares an application form 21, whose input item fields are filled in, images the application form 21 (step S22), and acquires image data 22.

[0027] Then, the generating device 100 calculates the difference between the image data 11 and the image data 22 to obtain the difference data 23. That is, the generating device 100 extracts only the input data entered in the input items. Next, the generating device 100 performs OCR processing on the difference data 23 to extract characters and the like included in the difference data 23 (step S23).

[0028] An example of the difference data 23 is shown in Fig. 4. Fig. 4 is a diagram (2) for explaining an example of the generation process according to the embodiment. Specifically, Fig. 4 shows character strings extracted as the difference data 23. As shown in Fig. 4, the difference data 23 does not extract input item fields or character strings indicating input items, and only includes input character strings that are the difference between the image data 11 and the image data 22.

[0029] Continuing the explanation by returning to Fig. 3, the generating device 100 inputs the extracted character strings into a language model, and generates sample data 24 corresponding to each input item (step S24).

[0030] Here, FIG. 5 shows an example of sample data 24. FIG. 5 is a diagram (3) for explaining an example of the generation process according to the embodiment. Specifically, FIG. 5 shows a character string generated by the language model through a dialogue with the user as sample data 24. As shown in FIG. 5, the user requests the language model to convert the actual input data described in the image data 22 into sample data. Then, the language model returns sample data in which the values ​​are converted without changing the attributes. In this case, the language model can generate sample data in which the character type, number of digits, and information attributes are not changed from the input data acquired as a sample by using, for example, the above-mentioned attribute information.

[0031] Continuing the explanation by returning to Fig. 3, the generating device 100 can generate input sample data 50 by combining the generated sample data with image data 11 of an application form 10 that has not yet been input.

[0032] Next, an example of a case where a user utilizes sample data generated by the generating device 100 will be shown with reference to Fig. 6 to Fig. 8. Fig. 6 is a diagram (1) showing an example of a user interface according to an embodiment. In the example shown in Fig. 6, the generating device 100 reads the application form 10 shown in Fig. 1 etc., analyzes the structure and input items, and generates the input application form 20.

[0033] For example, the generating device 100 provides a user interface 60 shown in Fig. 6 etc. to a user via a display device such as a display. The user interface 60 includes a display field for displaying image data of an unfilled application form 10 and an information field 62 in which various information is provided.

[0034] A user can input data into the input items on the user interface 60. At this time, when the user selects an input item 61 (for example, when the user moves a cursor near the input item 61 using a device such as a mouse), the generating device 100 displays input candidates 63 (sample data) corresponding to the input item in the information field 62. This allows the user to easily understand what information to input into the input item.

[0035] The generating device 100 may automatically input information displayed in the input candidates 63 into the corresponding input item. As an example, the generating device 100 may perform a support process such as automatically inputting the date on which the user is going to make an application into the corresponding input item.

[0036] 6 is an example, and the generating device 100 may display input candidates in a different manner. For example, when a user places a cursor near an input item 61, the generating device 100 may display input candidates in a pop-up in an adjacent blank area. Alternatively, the generating device 100 may display input candidates as placeholders in the adjacent blank area. When a user starts an operation, the generating device 100 may display input candidates in a pop-up in the vicinity of all input items, not limited to the input item 61.

[0037] Next, a case will be shown in which the user views sample data on the user interface 60. Fig. 7 is a diagram (2) showing an example of the user interface according to the embodiment.

[0038] For example, before entering the application form 10, the user presses a button 65 that reads "View Sample" displayed in the information field 62. When the button 65 is pressed, the generating device 100 switches the display of the application form 10 to a display of the inputted application form 20 in which sample data has been entered. In this way, the user can check the sample data at any time.

[0039] It should be noted that the generating device 100 may provide the user interface 60 in different modes. For example, the generating device 100 may always display the inputted application form 20 adjacent to the application form 10. Alternatively, the generating device 100 may automatically display the inputted application form 20 when the user is about to start inputting data into the application form 10. The generating device 100 may also display the inputted application form 20 by superimposing (overlaying) it on the application form 10.

[0040] Next, a case will be shown in which the application form input by the user is checked on the user interface 60. Fig. 8 is a diagram (3) showing an example of the user interface according to the embodiment.

[0041] The user enters information into the application form 10 on the user interface 60. After that, the user presses a "check" button 67 displayed in the information field 62. The generating device 100 then compares the content entered by the user with sample data generated in advance, and checks whether the input content complies with the format of the sample data.

[0042] For example, since the generating device 100 acquires attribute information for each input item, it acquires the correct value for the number of digits of a character string to be input into a certain input item. Therefore, when the content input by the user does not satisfy the number of digits that should be input into the input item, the generating device 100 displays a warning such as "The number of digits in the management number is incorrect," as shown in the check result 68. Alternatively, when an item to be input is not filled in, the generating device 100 displays a warning such as "The name is blank." In addition, when a space where the user should input or attach image data such as a seal is blank, the generating device 100 can give a warning such as "Be careful not to forget to affix a seal."

[0043] The generating device 100 may provide the user interface 60 in a different manner. For example, the generating device 100 may constantly maintain a check state for the information input by the user and display the check result 68 in real time without pressing the button 67. The generating device 100 may present the result to the user in a manner that displays an error in the relevant portion, rather than displaying the check result 68 in the information column 62. The generating device 100 may provide the user with the input application form 20 that may be the correct data, documents related to the input contents, and the like, together with the check result.

[0044] As described above and shown in FIG. 1 to FIG. 8, the generating device 100 generates sample data for various documents such as forms and applications, and provides a user interface to assist the user in input. In this manner, the generating device 100 can dynamically or on-demand generate and provide sample data when a user creates an application or the like for which sample data has not been generated in advance. This allows the generating device 100 to meet the user's demand for a sample immediately when the user creates a document for which sample data has not been prepared. Furthermore, the generating device 100 can improve usability by providing a user interface equipped with checks, sample display, and the like.

[0045] (1-2. Configuration of the Generation Device According to the Embodiment) Next, a configuration of the generating device 100 that executes the information processing according to the embodiment will be described. Fig. 9 is a diagram showing an example of the configuration of the generating device 100 according to the embodiment.

[0046] 9, the generating device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The generating device 100 may also include an input unit (e.g., a keyboard, a mouse, etc.) that receives various operations from an administrator who manages the generating device 100, and a display unit (e.g., a liquid crystal display, etc.) that displays various information.

[0047] The communication unit 110 is realized by, for example, a network interface card (NIC) or a network interface controller. The communication unit 110 is connected to a network N (for example, the Internet) by wire or wirelessly, and transmits and receives information to and from an OCR device, a scanner, a display, and the like via the network N. For example, the communication unit 110 may transmit and receive information using any communication standard or communication technology, such as Wi-Fi (registered trademark), Bluetooth, a subscriber identity module (SIM), or low power wide area (LPWA).

[0048] The storage unit 120 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk.

[0049] In the embodiment, the storage unit 120 has a structural model storage unit 121 and a language model storage unit 122. The structural model storage unit 121 stores a structural model capable of outputting structured data indicating the layout, input items, and input positions of an application form or the like based on image data of a form, application, or the like. The language model storage unit 122 stores a language model capable of outputting appropriate character strings, etc. as information to be input into an input item based on character strings and attribute information of the input item. Note that these learning models do not necessarily need to be stored in the storage unit 120 inside the generation device 100, and may be stored in, for example, an external storage device on a cloud.

[0050] The control unit 130 is realized by, for example, a central processing unit (CPU), a micro processing unit (MPU), a graphics processing unit (GPU), etc., executing a program stored inside the generating device 100 using a RAM or the like as a working area. The control unit 130 is also a controller, and is realized by, for example, an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

[0051] As shown in FIG. 9, the control unit 130 includes an acquisition unit 131, an analysis unit 132, a generation unit 133, and a provision unit 134.

[0052] The acquisition unit 131 acquires various information used in the generation process according to the embodiment. For example, the acquisition unit 131 acquires image data of a document including input items by cooperating with a scanner or the like. The acquisition unit 131 also acquires information such as character strings included in the image data of the document by cooperating with an OCR device or the like.

[0053] The analysis unit 132 analyzes the structure of the document by inputting the image data acquired by the acquisition unit 131 into a structural model. Specifically, the analysis unit 132 inputs the image data into the structural model and acquires structured data including the layout and input items included in the image data, and writing area information for entering information in the input items.

[0054] The generation unit 133 identifies input items of the document based on the structure analyzed by the analysis unit 132, and generates sample data corresponding to the input items by inputting the contents of the input items into a language model.

[0055] The generating unit 133 can acquire attribute information for each input item by learning what information is input for each input item for the document to be processed, using the input document as correct answer data. In other words, the generating unit 133 learns the format of information input into a plurality of input documents, learns input rules, and acquires the learned result as attribute information.

[0056] The generating unit 133 may generate sample data using a language model that has learned attribute information including the character type or number of digits that may be input into the input items. The attribute information may be automatically acquired by machine learning or the like, or may be created by an administrator or the like manually providing the generating device 100 with information on the number of digits or the character type. The attribute information may also include rules regarding the relationships between the input items. For example, in the case of an application form that allows only alternative descriptions, such as when there is an input into one input item and an input is not made into the other input items, the generating unit 133 may acquire the rules as attribute information.

[0057] In addition, the language model can utilize a so-called publicly known large-scale language model, but by using various information sources, it is possible to further improve accuracy and personalize the generation of input candidates (sample data), the generation of sample data, the check results, etc. For example, the generation unit 133 can perform learning to generate more realistic data or personalized data as input candidates or sample data by using application forms of other people having attributes similar to those of the input user as learning data, or other application forms, etc., input by the user himself in the past as learning data.

[0058] More specifically, the generation unit 133 can generate the name, etc. of the user as sample data for the corresponding input items by pre-reading information such as a card containing personal information, or authentication information of the user stored in a terminal, server, etc.

[0059] Furthermore, since the generation unit 133 generates sample data using a large-scale language model, it is also possible to generate sample data using not only character strings of input items but also information such as the context of the entire application form, label information attached to the application form, and the format of the application form data, etc. This allows the generation unit 133 to generate more natural and appropriate sample data compared to sample data output simply based on information of input items only.

[0060] Furthermore, the generating unit 133 generates sample image data showing a sample of the document by incorporating the sample data corresponding to the input items into structured data showing the structure of the document. For example, the generating unit 133 can generate an input application form 20 containing sample data based on the structure of an uninputted application form 10 and sample data for each input item, as shown in FIG.

[0061] The providing unit 134 displays the sample image data generated by the generating unit 133 on a user interface used for inputting documents, and provides the same to the user.

[0062] As shown in Fig. 6 to Fig. 8, the user interface may have a sample data display function, a check function, etc. Specifically, the providing unit 134 may determine whether the input content received from the user in the user interface is suitable for the attribute information, and provide the determination result to the user. This allows the user to immediately determine whether the information he or she has input is appropriate, and thus allows the user to smoothly input appropriate information.

[0063] (1-3. Information Processing Procedure According to the Embodiment) The procedure of the information processing according to the embodiment will be specifically described with reference to Figures 10 and 11. Figure 10 is a flowchart (1) showing the procedure of the information processing according to the embodiment.

[0064] 10, the generating device 100 acquires image data of a document to be processed, such as an application form (step S101). The generating device 100 analyzes the structure of the acquired image data to acquire structured data (step S102).

[0065] Next, the generating device 100 generates sample data for each input item identified by the structural analysis, using a language model (step S103).

[0066] Then, the generating device 100 generates sample image data corresponding to the application form or the like to be processed, in which information has been input into the input fields, based on the structured data and the sample data (step S104). The generating device 100 provides the generated sample image data to the user (step S105).

[0067] Next, a procedure for supporting input of an application form or the like using a user interface will be described with reference to Fig. 11. Fig. 11 is a flowchart (2) showing the procedure of information processing according to the embodiment.

[0068] The generating device 100 provides the user with an application input screen as a user interface (step S201). The generating device 100 determines whether or not an item has been selected or input on the screen (step S202). If no item has been selected or input (step S202; No), the generating device 100 waits until an item is selected or input.

[0069] On the other hand, if an item has been selected or input (step S202; Yes), the generating device 100 displays input candidates (sample data) for the item in an information display field such as a sidebar (step S203).

[0070] After that, the generating device 100 determines whether or not there is a request to end input of the application form or a request to check the application form (step S204). If there is no request to end input of the application form or a request to check the application form (step S204; No), the generating device 100 waits again for an input operation by the user.

[0071] On the other hand, if there is a request to finish inputting the application form or a request to check the application form (step S204; Yes), the generating device 100 displays the check result on the user interface (step S205).

[0072] Thereafter, the generating device 100 determines whether the user's input is complete (step S206). If the user's input is not complete (step S206; No), the generating device 100 waits again for the user to perform an input operation, etc. On the other hand, if the user's input is complete (step S206; Yes), the generating device 100 ends the input support process. Note that if the user attempts to input information related to another document, the generating device 100 may repeat the process shown in FIG. 11 for the number of application forms.

[0073] (2. Modifications of the embodiment) (2-1. Entering personal information) The generating device 100 may acquire personal information of the user and then automatically input the personal information into appropriate fields based on the structured data. This will be described with reference to Fig. 12. Fig. 12 is a diagram showing an example of a generating process according to a modified example.

[0074] The generating device 100 acquires analysis data (structured data) related to an application form to be processed. The generating device 100 also acquires read data of a personal card or the like that contains personal information via an OCR device, a dedicated card reader, or the like.

[0075] Thereafter, the generating device 100 provides a user interface such as that shown in FIG. 6 and accepts input of the application form from the user. At this time, the generating device 100 may provide a guide for creating the application form as appropriate. For example, the generating device 100 may display options for accepting input from the user. As an example, the generating device 100 displays an option for automatically inputting personal information, an option for manually inputting information, an option for displaying a guidance for inputting information, an option for checking after inputting information, an option for requesting printing of the inputted application form, and the like.

[0076] At this time, if the user selects automatic input, the generating device 100 analyzes the personal information acquired from the card or the like, and then uses the structured data to automatically input the personal information into the appropriate input fields that should be input.

[0077] Furthermore, when obtaining personal information from a card or the like, the generating device 100 may perform personal authentication via a card reader or the like. That is, the generating device 100 may authenticate the identity of the person who has executed the card reading, and then automatically input personal information only for users who have been successfully authenticated. This allows the generating device 100 to prevent unauthorized input or application by anyone other than the person in question.

[0078] After that, the generating device 100 can automatically submit the application form that the user has completed inputting to the application destination, or print it if the user's seal is required. In this case, the generating device 100 may be configured to allow only users who have been personally authenticated to submit or print.

[0079] In this way, the generating device 100 can provide an environment that supports the user in all tasks, including automatic input, generation of sample data, execution of checks, submission and printing of input application forms, etc., even for applications that use the user's personal information. In addition, the generating device 100 can execute highly secure electronic applications by performing personal authentication.

[0080] (2-2. Other Modifications) The document in the present disclosure may be a paper medium that has been scanned and OCR processed, or may be an electronic document. The medium handled by the generating device 100 is not necessarily limited to a document, and may be any medium that accepts applications and inputs. The user does not need to use a dedicated scanner or the like when reading a document into the generating device 100, and may convert the document into image data using a personally owned camera or the like.

[0081] In the above embodiment, the configuration of the generating device 100 is shown in Fig. 9, but the generating device 100 is not limited to such a configuration. For example, the generating device 100 may have a scanner or OCR function, or may also have a function of creating an input document by printing, etc. In addition, the generating device 100 is not necessarily an individual generating device, and may be configured by combining a cloud server, a virtual server, a program, etc.

[0082] Furthermore, the information stored in the storage unit 120 described above does not necessarily have to be saved in the storage unit 120 inside the generating device 100. For example, each piece of information may be stored in a storage area of ​​a predetermined external storage device (such as cloud storage) or in an external device owned by an advertiser or the like.

[0083] Furthermore, in the above embodiment, the description has been given on the assumption that the user who uses the generating device 100 is an individual. However, the user may be in any form, such as a company, an organization, or an association.

[0084] (3. Other embodiments) The processing according to the above-described embodiment may be implemented in various different forms other than the above embodiment.

[0085] For example, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by a known method. In addition, the information including the processing procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

[0086] In addition, each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads, usage conditions, etc.

[0087] Furthermore, the above-described embodiments and modifications can be appropriately combined as long as the processing contents are not contradictory.

[0088] Furthermore, the effects described in this specification are merely examples and are not limiting, and other effects may also be obtained.

[0089] (4. Effects of the Generation Device According to the Present Disclosure) As described above, the generating device according to the present disclosure (generating device 100 in the embodiment) includes an acquiring unit (acquiring unit 131 in the embodiment), an analyzing unit (analyzing unit 132 in the embodiment), and a generating unit (generating unit 133 in the embodiment). The acquiring unit acquires image data of a document including input items. The analyzing unit inputs the image data acquired by the acquiring unit into a first model (a structural model in the embodiment) to analyze the structure of the document. The generating unit identifies input items of the document based on the structure analyzed by the analyzing unit, and generates sample data corresponding to the input items by inputting the contents of the input items into a second model (a language model in the embodiment).

[0090] In this way, the generating device according to the present disclosure analyzes the structure of documents such as applications that require some input by the user, and then generates sample data corresponding to the input items using a language model. This allows the generating device to generate sample data that is written in the correct input position without human intervention, making it possible to generate sample data quickly and easily.

[0091] The generation unit also generates sample data using a second model that has learned attribute information including the character type or number of digits that can be input into the input item.

[0092] In this way, the generation device generates sample data composed of appropriate character types and number of digits by learning attribute information of the information to be entered into the document in advance, thereby improving the accuracy of the sample data.

[0093] The generating unit also generates sample image data showing a sample of the document by incorporating sample data corresponding to the input items into structured data showing the structure of the document.

[0094] In this way, the generating device generates a sample image with sample data inserted in the appropriate input position after analyzing the layout itself, so it can easily generate a useful sample that allows the user to understand at a glance what items to enter. This allows the generating device to meet the needs of users who want to immediately view a sample of an application form or the like for which a sample is not prepared.

[0095] The generating device may further include a providing unit (in the embodiment, the providing unit 134) that displays the sample image data generated by the generating unit on a user interface used for inputting documents and provides it to the user.

[0096] In this way, the generation device provides a user interface that can display sample image data when a user creates a document, thereby providing a highly usable creation environment in which the user can create a document while looking at a sample.

[0097] The providing unit may also determine whether or not the input content received from the user in the user interface is compatible with the attribute information, and provide the user with the determination result.

[0098] In this way, the generating device according to the present disclosure can assist the user in inputting appropriate content by providing a user interface with a function for checking the content entered by the user.

[0099] (5. Hardware Configuration) An information device such as the generating device 100 according to the embodiment described above is realized by a computer 1000 having a configuration as shown in Fig. 13, for example. The generating device 100 according to the embodiment will be described below as an example. Fig. 13 is a hardware configuration diagram showing an example of a computer 1000 that realizes the functions of the generating device 100. The computer 1000 has a CPU 1100, a RAM 1200, a ROM (Read Only Memory) 1300, a HDD (Hard Disk Drive) 1400, a communication interface 1500, and an input / output interface 1600. Each unit of the computer 1000 is connected by a bus 1050.

[0100] The CPU 1100 operates and controls each unit based on a program stored in the ROM 1300 or the HDD 1400. For example, the CPU 1100 loads the program stored in the ROM 1300 or the HDD 1400 into the RAM 1200 and executes processing corresponding to the various programs.

[0101] The ROM 1300 stores boot programs such as a basic input output system (BIOS) executed by the CPU 1100 when the computer 1000 is started, and programs that depend on the hardware of the computer 1000.

[0102] HDD 1400 is a computer-readable recording medium that non-temporarily records programs executed by CPU 1100 and data used by such programs. Specifically, HDD 1400 is a recording medium that records a program that executes information processing according to the present disclosure, which is an example of program data 1450.

[0103] The communication interface 1500 is an interface for connecting the computer 1000 to an external network 1550 (e.g., the Internet). For example, the CPU 1100 receives data from other devices and transmits data generated by the CPU 1100 to other devices via the communication interface 1500.

[0104] The input / output interface 1600 is an interface for connecting the input / output device 1650 and the computer 1000. For example, the CPU 1100 receives data from an input device such as a keyboard or a mouse via the input / output interface 1600. The CPU 1100 also transmits data to an output device such as a display, a speaker, or a printer via the input / output interface 1600. The input / output interface 1600 may also function as a media interface for reading a program or the like recorded on a predetermined recording medium. The medium may be, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0105] For example, when the computer 1000 functions as the generating device 100 according to the embodiment, the CPU 1100 of the computer 1000 executes a generating program loaded onto the RAM 1200 to realize functions of the control unit 130, etc. Also, the HDD 1400 stores a program for executing information processing according to the present disclosure and data in the storage unit 120. Note that the CPU 1100 reads and executes the program data 1450 from the HDD 1400, but as another example, the CPU 1100 may obtain these programs from other devices via an external network 1550.

[0106] The above describes the embodiments of the present application in detail based on the drawings. However, these are merely examples, and the present invention can be embodied in other forms that include the aspects described in the Disclosure of the Invention section, as well as in other forms that include various modifications and improvements based on the knowledge of those skilled in the art. [Explanation of symbols]

[0107] 10 Application form 20 Completed application form 60 User Interface 100 generator 110 Communications Department 120 Storage section 121 Structural model storage unit 122 Language model storage unit 130 Control section 131 Acquisition Department 132 Analysis Department 133 Generation part 134 Provision Department

Claims

1. An acquisition unit that acquires image data of a document including input items; an analysis unit that analyzes a structure of the document by inputting the image data acquired by the acquisition unit into a first model; a generation unit that identifies input items of a document based on the structure analyzed by the analysis unit, and generates sample data corresponding to the input items by inputting contents of the input items into a second model; A generating device comprising:

2. The generation unit is generating the sample data using the second model that has learned attribute information including a character type or a number of digits that can be input into the input item; The generating device according to claim 1 .

3. The generation unit is generating sample image data showing a sample of the document by incorporating sample data corresponding to the input items into structured data showing the structure of the document; The generating device according to claim 1 .

4. a providing unit that displays the sample image data generated by the generating unit on a user interface used for inputting the document and provides the sample image data to a user; The generating device of claim 3 further comprising:

5. The generation unit is generating the sample data using the second model that has learned attribute information including a character type or a number of digits that can be input into the input item; The providing unit is determining whether or not an input content received from a user via the user interface is compatible with the attribute information, and providing a result of the determination to the user; 5. The generating device according to claim 4.

6. The computer Obtain image data of a document including input fields; Analyzing a structure of the document by inputting the acquired image data into a first model; Identifying input items of the document based on the analyzed structure, and inputting the contents of the input items into a second model to generate sample data corresponding to the input items. A method of generating a signal, comprising:

7. Computer, An acquisition unit that acquires image data of a document including input items; an analysis unit that analyzes a structure of the document by inputting the image data acquired by the acquisition unit into a first model; a generation unit that identifies input items of a document based on the structure analyzed by the analysis unit, and generates sample data corresponding to the input items by inputting contents of the input items into a second model; A generating program that causes the device to function as a generating device comprising:

Images