Processing device, processing program, and processing method

The processing system accurately determines regulatory compliance of content data by using a multi-check approach with trained models, addressing the challenge of complex content compliance in pharmaceuticals, quasi-drugs, and medical devices.

JP2024160817APending Publication Date: 2024-11-15REGAL CORE CO LTD

Patent Information

Application Number
JP2023076217
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-02
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

Existing systems face challenges in accurately determining whether data in various content types complies with regulatory laws and guidelines, particularly for pharmaceuticals, quasi-drugs, cosmetics, and medical devices, due to the complexity and variability of the content.

Method used

A processing system that includes a client terminal, server device, and checker terminal devices to analyze content data, utilizing trained models and multiple checks by authorized personnel to determine compliance with regulations, generating comprehensive results.

Benefits of technology

Enhances the accuracy of determining regulatory compliance by integrating multiple checks and trained models, ensuring precise adherence to laws and guidelines for content data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024160817000001_ABST
    Figure 2024160817000001_ABST
Patent Text Reader

Abstract

To provide a method to high-accurately perform the determination whether or not the data content included in contents is regulated by prescribed laws and regulations and by a guideline.SOLUTION: In a processing system 1, a server device of a processing device includes at least one processor. At least one processor accepts the input of the contents including at least one data of a determination target from a client terminal and acquires the data from the contents, determines the data availability in the contents based on the regulation by the prescribed laws and regulations or the guideline for each of the acquired data, transmits the request of the determination result of the data availability and of the propriety of the determination result to a checker terminal, and transmits the total result generated based on the determination result and on the check result received from the checker terminal to the client terminal device.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a processing device, a processing program, and a method for determining whether data included in content is subject to regulation by a predetermined law or guideline. [Background technology]

[0002] In the past, because the way in which drugs, quasi-drugs, cosmetics, medical devices, or regenerative medicine products were advertised could have a significant impact on health and hygiene, there was, for example, the "Law for Ensuring the Quality, Efficacy and Safety of Drugs, Medical Devices, etc." This regulated inappropriate advertising of drugs, quasi-drugs, cosmetics, medical devices, or regenerative medicine products.

[0003] In relation to this situation, for example, Patent Document 1 describes a system that has an NG comment database in which NG sentences and NG words are registered in advance, and determines whether or not utterance data entered as text is NG by comparing it with the NG comment database, and reports the result.

[0004] However, although there is a demand for judging data containing various contents, there is a risk that, depending on the contents of the data, judgment may not be possible or the accuracy of judgment may decrease. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2011-204178 A Summary of the Invention [Problem to be solved by the invention]

[0006] The present disclosure has been made against the above-mentioned background, and provides a processing device, processing program, or processing method for determining with greater accuracy whether the content of data included in content is subject to regulation by specified laws, regulations, or guidelines. [Means for solving the problem]

[0007] According to one aspect of the present disclosure, there is provided a processing device including at least one processor, the at least one processor being configured to perform processing for accepting input of content including at least one data to be judged from a requester terminal device, acquiring the data from the content, judging for each of the acquired data whether or not the data can be used in the content based on regulations under predetermined laws or guidelines, transmitting a judgment result on whether or not the data can be used and a request for the appropriateness of the judgment result to a checker terminal device, and transmitting an overall result generated based on the judgment result and the check result received from the checker terminal device to the requester terminal device."

[0008] According to one aspect of the present disclosure, there is provided a processing program that, when executed by at least one processor, causes the at least one processor to function as follows: accept input of content including at least one data to be judged from a requester terminal device, acquire the data from the content, judge for each of the acquired data whether or not the data can be used in the content based on regulations imposed by predetermined laws or guidelines, transmit a judgment result of whether or not the data can be used and a request for the appropriateness of the judgment result to a checker terminal device, and transmit to the requester terminal device a comprehensive result generated based on the judgment result and the check result received from the checker terminal device."

[0009] According to one aspect of the present disclosure, there is provided a "processing method executed by at least one processor, comprising the steps of: accepting an input of content including at least one data to be judged from a requester terminal device, and acquiring the data from the content; judging, for each of the acquired data, whether or not the data can be used in the content based on regulations under predetermined laws or guidelines; transmitting a judgment result on whether or not the data can be used and a request for the appropriateness of the judgment result to a checker terminal device; and transmitting a comprehensive result generated based on the judgment result and the check result received from the checker terminal device to the requester terminal device." Effect of the Invention

[0010] According to the present disclosure, it is possible to provide a processing device, a processing program, or a processing method for determining with higher accuracy whether data included in content is subject to regulation by a specified law or guideline.

[0011] It should be noted that the above effects are merely illustrative for the convenience of explanation and are not limiting. In addition to or instead of the above effects, any effect described in this disclosure or any effect obvious to a person skilled in the art may be achieved. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating an overview of a screening process performed by a processing system 1 according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is a conceptual diagram illustrating a schematic configuration of a processing system 1 according to an embodiment of the present disclosure. [Diagram 3] FIG. 3 is a block diagram illustrating an example of a configuration of the client terminal device 100 according to an embodiment of the present disclosure. [Figure 4] FIG. 4 is a block diagram illustrating an example of a configuration of the server device 200 according to an embodiment of the present disclosure. [Diagram 5]FIG. 5 is a block diagram illustrating an example of a configuration of the first checker terminal device 300 according to an embodiment of the present disclosure. [Figure 6] FIG. 6 is a block diagram illustrating an example of a configuration of the second checker terminal device 400 according to an embodiment of the present disclosure. [Figure 7A] FIG. 7A is a diagram conceptually illustrating a database for screening stored in the server device 200 according to an embodiment of the present disclosure. [Figure 7B] FIG. 7B is a diagram conceptually illustrating a database for screening stored in the server device 200 according to the embodiment of the present disclosure. [Figure 7C] FIG. 7C is a diagram conceptually illustrating a content table stored in the server device 200 according to an embodiment of the present disclosure. [Figure 8A] FIG. 8A is a diagram showing a processing sequence executed among the client terminal device 100, the server device 200, the first checker terminal device 300, and the second checker terminal device 400 according to the embodiment of the present disclosure. [Figure 8B] FIG. 8B is a diagram showing an example of a screen output to the client terminal device 100 according to an embodiment of the present disclosure. [Figure 8C] FIG. 8C is a diagram showing an example of a screen output to the first checker terminal device 300 and the second checker terminal device 400 according to an embodiment of the present disclosure. [Figure 8D] FIG. 8D is a diagram showing an example of a screen output to the client terminal device 100 according to an embodiment of the present disclosure. [Figure 9] FIG. 9 is a diagram showing a processing sequence executed among the server device 200, the first checker terminal device 300, and the second checker terminal device 400 according to the embodiment of the present disclosure. [Figure 10A] FIG. 10A is a diagram showing a process flow executed in the server device 200 according to an embodiment of the present disclosure. [Figure 10B] FIG. 10B is a diagram showing a process flow executed in the server device 200 according to an embodiment of the present disclosure. [Figure 11A]FIG. 11A is a diagram conceptually illustrating generation management of a trained model stored in a server device 200 according to an embodiment of the present disclosure. [Figure 11B] FIG. 11B is a diagram showing a process flow executed in the server device 200 according to an embodiment of the present disclosure. [Figure 12] FIG. 12 is a diagram showing a process flow executed in the server device 200 according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] The embodiments of the present disclosure will be described with reference to the accompanying drawings, in which the same reference numerals are used to designate the same components.

[0014] <Embodiment> 1. Overview of processing system 1 according to the present disclosure The processing system 1 according to the present disclosure is used, for example, to screen whether or not content such as advertisements uses wording or expressions regulated by laws and regulations, such as the Act on Ensuring the Quality, Effectiveness, and Safety of Pharmaceuticals, Medical Devices, etc., or guidelines. When a user inputs the content to be screened and the processing system 1 accepts the content, it acquires predetermined data (e.g., text or images) from the content. The processing system 1 then screens the acquired predetermined data for wording or expressions regulated by the laws and regulations or guidelines, and outputs an overall result that is the result of the screening.

[0015] FIG. 1 is a diagram showing an outline of the screening process of the processing system 1 according to the embodiment of the present disclosure. Specifically, FIG. 1 shows a general flow from a screening request made by a user who is a requester desiring screening to a result of screening being returned to the requester. As shown in FIG. 1, content information to be screened and a screening request are transmitted from a requester terminal device 100 used by the user to a server device 200. In the server device 200, a first screening process is performed on the received content information. In addition, the server device 200 requests the first checker terminal device 300 and the second checker terminal device 400 to check the result of the first screening process.

[0016] In the first checker terminal device 300 and the second checker terminal device 400, the result of the primary judgment process is checked, and the usability of not only the comments on the primary judgment result but also the predetermined data in the content is judged. By carrying out this process, the first checker terminal device 300 and the second checker terminal device 400 generate a check result. The check result may include a correction (alternative) for the data that is not permitted to be used. The check result is then transmitted from the first checker terminal device 300 and the second checker terminal device 400 to the server device 200.

[0017] Thereafter, the server device 200 performs a secondary judgment process while taking into consideration the result of the primary judgment process and the received check result, and generates an inspection result (total result) that is a response to the inspection request from the requester. The server device 200 then transmits the inspection result to the requester terminal device 100, and the inspection result can be confirmed on the requester terminal device 100. The inspection result may also include corrections (alternatives) for data that is not permitted to be used.

[0018] In this disclosure, examples of the laws and guidelines include various laws and guidelines that regulate inappropriate labeling, such as the "Act on Ensuring Quality, Effectiveness, and Safety of Pharmaceuticals, Medical Devices, etc.", the "Specified Commercial Transactions Act," the "Act on Prevention of Unfair Premiums and Misrepresentations," the "Health Promotion Act," the "Food Labeling Act," the "Food Labeling Standards," and the "Medical Advertising Guidelines." These are not limited to those enacted by national or local governments, but may be enacted by organizations or individuals such as specific industry groups, unions, companies, associations, etc. Furthermore, they do not need to be explicitly called "laws" or "guidelines," and may be any kind of standard, rule, regulation, or rule. Furthermore, they do not need to be explicitly stated and published in a document. Furthermore, the laws and guidelines are not limited to those related to doctors, medical care, or food, and may include, for example, the Copyright Act, the Financial Instruments and Exchange Act, and other laws as long as they are related to evaluating images.

[0019] In addition, in this disclosure, examples of items subject to the regulations of laws, regulations or guidelines include medicines, quasi-drugs, regenerative medicine products, medical devices, health foods, functional foods, cosmetics, and mail-order products. However, these are of course only examples and are not limited to these. For example, when making a judgment regarding an image based on the above-mentioned Copyright Act, items that require attention to the Copyright Act, such as photo books of people and catalogs of artworks, may also be subject to the regulations.

[0020] In addition, in the present disclosure, "content" typically refers to advertising content in the form of characters (text notation), but may be either advertising content in the form of video or advertising content in the form of still images. In addition, as examples of content other than advertising, for example, product packages, product introductions, packaging, company introductions, introductions on SNS, news, newspaper articles, and various other things can be given. In addition, in any format, at least one data (characters, symbols, text, images, or videos) may be included. That is, content in various formats can be used, such as a webpage format including HTML and image data, an image data format consisting of image data only, a standardized electronic file format such as PDF that can be used without depending on a specific OS or hardware, and an electronic file format used for image editing, spreadsheets, and presentations. In addition, these contents may include text such as characters or symbols in a state that can be recognized as text data, or may be integrated with the above image.

[0021] In addition, in this disclosure, "data" simply means something that is contained in the content and has meaning. In other words, "data" includes characters, symbols, sentences, images, videos, and combinations of these. For example, "data" includes words with some meaning, sentences consisting of such words, and predetermined symbols and marks. Furthermore, "data" also includes still images and videos, and their explanations.

[0022] In addition, in the present disclosure, "judgment" is a process of judging whether or not at least one piece of data included in the content can be used in the content based on the regulation by a specific law or guideline. In particular, in the present disclosure, "judgment" is a process performed in the server device 200. In contrast, "check" is a process of confirming that the judgment process is correct, and further includes a process of judging whether or not at least one piece of data included in the content can be used in the content based on the regulation by a specific law or guideline. In particular, in the present disclosure, "check" is a process performed in the first checker terminal device 300 and the second checker terminal device 400. And, "examination" is a process performed by performing the judgment and check. That is, "examination" is a comprehensive final judgment based on the judgment and check, and is a process of judging whether or not the content itself has a problem with respect to the regulation by a specific law or guideline. For this reason, in the present disclosure, "examination result" is positioned as a "comprehensive result" based on the "judgment result" and the "check result". In the following description, the "examination target" is the content, and the "judgment target" is the data included in the content. These terms are merely used for convenience of explanation, and different names may be used as long as the processes can be distinguished from one another.

[0023] Furthermore, "analysis" is the same as "judgment," and is a process of judging whether or not at least one piece of data contained in the content can be used within the content based on the restrictions of a certain law or guideline. However, "analysis" is distinguished from "judgment" as a process that targets content that includes the target data but for which no review request has been received from the client.

[0024] In addition, in this disclosure, the term "user" simply means a user who can use the service provided by the processing system 1. In this disclosure, a person who requests the review of content is called a user.

[0025] In addition, in this disclosure, a "checker" is merely a person who checks the judgment result provided by the server device 200 of the processing system 1. In this disclosure, there are two types of "checkers," a person who is legally authorized to perform the check, such as a lawyer, is referred to as a "first checker," and a person who is assigned to perform the check practically without legal authorization is referred to as a "second checker." Similarly, a terminal device held by a "first checker" is referred to as a first checker terminal device 300, and a terminal device held by a "second checker" is referred to as a second checker terminal device 400. In addition, descriptions such as "first" and "second" may be used, but unless otherwise specified, "first" is described for those related to the "first checker" and "second" is described for those related to the "second checker."

[0026] In this disclosure, even if a description such as "first" or "second" is used, it does not mean that the description is limited to only the two elements to which the description is attached. Naturally, the description may include "third," "fourth," and more elements.

[0027] In addition, although this disclosure uses terms such as "examination," "judgment," "determination," "check," "confirmation," "analysis," and "analysis," these terms are not limited to the meanings defined by specific laws, regulations, etc.

[0028] In the following example, a case will be described in which it is determined whether or not advertising content provided in text form contains expressions regulated by laws, regulations or guidelines, but the present invention is not limited to this example.

[0029] 2. Configuration of Processing System 1 Fig. 2 is a conceptual diagram illustrating a schematic configuration of a processing system 1 according to an embodiment of the present disclosure. According to Fig. 2, the processing system 1 includes a requester terminal device 100 held by a user who requests an examination, a server device 200 that executes a judgment process, and a first checker terminal device 300 and a second checker terminal device 400 that check the result of the judgment by the server device 200. In the processing system 1, each device is connected to each other so as to be able to communicate with each other via a network 500. The network 500 may be configured wirelessly, wired, or a combination thereof. In the processing system 1, each device executes a program stored in a memory to execute the examination process according to this embodiment.

[0030] In the example of FIG. 2, only one requester terminal device 100 is shown, but it is possible to include two or more requester terminal devices 100. In the example of FIG. 2, only one first checker terminal device 300 and one second checker terminal device 400 are shown, but it is possible to include two or more of each device. Furthermore, the server device 200 is shown as a single device, but it is also possible to distribute each component and process of the server device 200 to multiple server devices or cloud server devices. The process according to this embodiment is executed by the processing system 1 including the requester terminal device 100, the server device 200, the first checker terminal device 300, and the second checker terminal device 400, but it is also possible to execute the process of the server device 200 only by the requester terminal device 100, the first checker terminal device 300, or the second checker terminal device 400 without using the server device 200.

[0031] The first checker terminal device 300 and the second checker terminal device 400 may be integrated into one checker terminal device.

[0032] As described above, in the present disclosure, the processing device may be any of the requester terminal device 100, the server device 200, the first checker terminal device 300, and the second checker terminal device 400, so long as it is capable of executing the processing program or the processing method according to the present disclosure. In the following example, unless otherwise specified, a case will be described in which the server device 200 functions as the processing device, but the present disclosure is not limited to this example.

[0033] 3. Configuration of client terminal device 100 Fig. 3 is a block diagram showing an example of a configuration of a client terminal device 100 according to an embodiment of the present disclosure. The client terminal device 100 does not need to include all of the components shown in Fig. 3, and may have a configuration in which some components are omitted, or may include other components.

[0034] An example of the client terminal device 100 is a laptop PC. However, the client terminal device 100 can be suitably applied to any device capable of executing the processing program according to the present disclosure, such as a portable terminal device capable of wireless communication, such as a smartphone, a feature phone, a personal digital assistant, a PDA, a desktop computer, a portable game machine, or a stationary game machine.

[0035] 3, the client terminal device 100 includes an output interface 111, a processor 112, a memory 113 including a RAM, a ROM, a non-volatile memory (HDD in some cases), etc., a communication interface 114 including a communication processing circuit and an antenna, and an input interface 115 including a touch panel, hard keys, etc. These components are electrically connected to each other via control lines and data lines.

[0036] The output interface 111 functions as an output unit that outputs, in response to an instruction from the processor 112, an image captured by a camera (not shown), image information stored in the memory 113, and various displays output by executing a program according to the present disclosure to a device such as a display or a printer. An example of the output interface 111 is a display configured from a liquid crystal display or an organic EL display, but the client terminal device 100 itself does not necessarily need to be equipped with a display. For example, the output interface 111 may be one that outputs information for various displays to a display connected by wire or wirelessly. In other words, the communication interface 114 may function as the output interface 111.

[0037] The processor 112 is composed of a GPU (graphics processing unit) or a CPU (microcomputer), and functions as a control unit that controls other components connected thereto based on various programs stored in the memory 113. Specifically, the processor 112 reads out from the memory 113 a program for executing the processing according to this embodiment and a program for executing the OS, and executes them. In the present disclosure, the processor 112 particularly executes each process shown in the processing sequence of FIG. 8A (details of the process will be described in FIG. 8A). The processor 112 may be composed of a single CPU or multiple CPUs. In addition, other types of processors, such as a GPU specialized for image processing, may be appropriately combined.

[0038] The memory 113 is composed of a ROM, a RAM, a non-volatile memory, a HDD, etc., and functions as a storage unit. The ROM stores instructions for executing the game application and the OS according to this embodiment as a program. The RAM is a memory used for writing and reading data while the program stored in the ROM is being processed by the processor 112. The non-volatile memory is a memory into which data is written and read by the execution of the program, and the data written therein is stored even after the execution of the program is terminated. In particular, it executes each process described in the process flow shown in FIG. 8A (details of the process will be described in FIG. 8A). Although not particularly shown as the memory 113, it may be connected to a removable storage medium, a database, etc. via the input interface 115.

[0039] The communication interface 114 functions as a communication unit that transmits and receives information to and from the server device 200 and other terminal devices installed remotely via a communication processing circuit and an antenna. The communication processing circuit performs processing to receive, from the server device 200, a program for executing the processing according to this embodiment and various information used in the processing, etc., according to the progress of the processing. It also performs processing to transmit various information required for executing the processing to the server device 200. In this embodiment, in particular, content to be subject to review and review results are transmitted and received between the server device 200 and the communication processing circuit.

[0040] The communication processing circuit processes based on a wideband wireless communication system such as the LTE system, but it is also possible to process based on a system related to narrowband wireless communication such as wireless LAN such as IEEE802.11 or Bluetooth (registered trademark). Also, wired communication can be used instead of or in addition to wireless communication.

[0041] The input interface 115 is composed of a touch panel and / or hard keys, and accepts operation input from a user. The touch panel is arranged to cover the display as the output interface 111, and outputs information on position coordinates corresponding to image data displayed on the display to the processor 112. As the touch panel method, known methods such as a resistive film method, a capacitive coupling method, and an ultrasonic surface acoustic wave method can be used. In the present embodiment, as an example, the touch panel detects a tap operation or a swipe operation for selecting or moving an object displayed on the display by a pointer. Note that the touch panel is one example of the input interface 115, and naturally other types can be used. It is also possible to use an input interface (such as a so-called keyboard or mouse) connected wirelessly or by wire to a main body including the processor 112, etc.

[0042] 4. Configuration of Server Device 200 Fig. 4 is a block diagram showing an example of the configuration of a server device 200 according to an embodiment of the present disclosure. The server device 200 does not need to include all of the components shown in Fig. 4, and may have a configuration in which some components are omitted, or other components may be added. In addition, the server device 200 does not need to include the components shown in Fig. 4 in a single housing, and each component and process of the server device 200 may be distributed to multiple server devices or cloud server devices.

[0043] 4, the server device 200 includes a memory 211 including a RAM, a ROM, a non-volatile memory, an HDD, etc., a processor 212 including a GPU or a CPU, etc., a communication interface 213, and an input interface 214. These components are electrically connected to each other via control lines and data lines.

[0044] The memory 211 includes a RAM, a ROM, a non-volatile memory, and a HDD, and functions as a storage unit. The ROM stores instructions for executing the application and the OS according to the present disclosure as a program. Such a program is loaded and executed by the processor 212. The RAM is used to write and read data while the program stored in the ROM is being processed by the processor 212. The non-volatile memory is a memory into which data is written and read by the execution of the program, and the data written therein is stored even after the execution of the program is terminated.

[0045] In the present disclosure, the memory 211 stores the database for examination shown in FIG. 7A and FIG. 7B and the content table shown in FIG. 7C. In addition, in the present disclosure, the memory 211 particularly stores programs for "a process of accepting an input of content including at least one data to be judged from the requester terminal device 100 and acquiring the data from the content," "a process of judging whether or not the data can be used in the content for each of the acquired data based on the regulation by a predetermined law or guideline," "a process of transmitting the judgment result of whether or not the data can be used and a request for the appropriateness of the judgment result to each checker terminal device (the first checker terminal device 300 and the second checker terminal device 400)," and "a process of transmitting the judgment result and a comprehensive result generated based on the check result received from each checker terminal device to the requester terminal device 100." Furthermore, the memory 211 stores a plurality of trained models used in various processes. In this embodiment, the trained models stored include a model for content analysis, a model for sentence decomposition, a model for judgment processing (trained judgment model), and a model for alternative selection. Although not shown in the figure, the examination database stored in the memory 211 may be stored in a database device or the like installed remotely via a network.

[0046] The processor 212 is composed of a GPU (graphics processing unit) or a CPU (microcomputer), and functions as a control unit for controlling other components connected thereto based on various programs stored in the memory 211. Specifically, the processor 212 reads out from the memory 211 a program for executing an application according to the present disclosure and a program for executing an OS, and executes the program.

[0047] In the present disclosure, the processor 212 particularly executes "a process of accepting an input of content including at least one data to be judged from the requester terminal device 100 and acquiring the data from the content," "a process of judging whether or not the data acquired can be used in the content based on the regulations of a predetermined law or guideline," "a process of transmitting the judgment result of whether or not the data can be used and a request for the appropriateness of the judgment result to each checker terminal device (the first checker terminal device 300 and the second checker terminal device 400)," and "a process of transmitting to the requester terminal device 100 a comprehensive result generated based on the judgment result and the check results received from each checker terminal device," etc. The processor 212 may be configured with a single CPU or multiple CPUs.

[0048] The communication interface 213 functions as a communication unit that transmits and receives information between other remotely installed server devices, the requester terminal device 100, the first checker terminal device 300, and the second checker terminal device 400 via a communication processing circuit and an antenna. The communication processing circuit processes programs and various information used in the processing system 1 to transmit and receive information from other server devices, the requester terminal device 100, the first checker terminal device 300, and the second checker terminal device 400 according to the progress of processing. In the present disclosure, particularly, via the communication interface 213, content and an examination request are received from the requester terminal device 100, and examination results are transmitted to the requester terminal device 100. Also, in the present disclosure, particularly, via the communication interface 213, a check request is transmitted to the first checker terminal device 300 and the second checker terminal device 400, and check results are received from the first checker terminal device 300 and the second checker terminal device 400.

[0049] The communication processing circuit processes based on a wideband wireless communication system such as the LTE system, but it is also possible to process based on a system related to narrowband wireless communication or a system related to non-contact wireless communication such as wireless LAN such as IEEE802.11 or Bluetooth (registered trademark). Also, wired communication can be used instead of or in addition to wireless communication.

[0050] The input interface 214 is composed of a touch panel, hard keys, etc., and functions as an input unit that accepts instruction inputs related to the execution of the program according to the present disclosure, operation inputs for registering various information, etc. In the present disclosure, the input interface 214 is used as an information input device for executing model generation, which will be described later.

[0051] 5. Configuration of the First Checker Terminal Device 300 Fig. 5 is a block diagram showing an example of the configuration of the first checker terminal device 300 according to an embodiment of the present disclosure. The first checker terminal device 300 does not need to include all of the components shown in Fig. 5, and may have a configuration in which some components are omitted, or may include other components.

[0052] An example of the first checker terminal device 300 is a laptop PC, but the present invention can also be suitably applied to any other device capable of executing the processing program according to the present disclosure, such as a portable terminal device capable of wireless communication, such as a smartphone, a feature phone, a personal digital assistant, a PDA, a desktop computer, a portable game machine, or a stationary game machine.

[0053] 5, the first checker terminal device 300 includes an output interface 311, a processor 312, a memory 313 including RAM, ROM, non-volatile memory (HDD in some cases), etc., a communication interface 314 including a communication processing circuit and an antenna, and an input interface 315 including a touch panel, hard keys, etc. These components are electrically connected to each other via control lines and data lines.

[0054] The output interface 311 functions as an output unit that outputs, in response to an instruction from the processor 312, an image captured by a camera (not shown) or image information stored in the memory 313, and various displays output by executing the program according to the present disclosure to a device such as a display or a printer. An example of the output interface 311 is a display configured with a liquid crystal display or an organic EL display, but the first checker terminal device 300 itself does not necessarily need to be equipped with a display. For example, the output interface 311 may be one that outputs information for various displays to a display connected by wire or wirelessly. In other words, the communication interface 314 may function as the output interface 311.

[0055] The processor 312 is composed of a GPU (graphics processing unit) or a CPU (microcomputer), and functions as a control unit that controls other components connected thereto based on various programs stored in the memory 313. Specifically, the processor 312 reads out from the memory 313 a program for executing the processing according to the present embodiment and a program for executing the OS, and executes them. In the present disclosure, the processor 312 particularly executes each process shown in the processing sequence of FIG. 8A and FIG. 8B (details of the process will be described in FIG. 8A and FIG. 8B). The processor 312 may be composed of a single CPU or multiple CPUs. In addition, other types of processors, such as a GPU specialized for image processing, may be appropriately combined.

[0056] The memory 313 is composed of a ROM, a RAM, a non-volatile memory, a HDD, etc., and functions as a storage unit. The ROM stores instructions for executing the game application and the OS according to this embodiment as a program. The RAM is a memory used for writing and reading data while the program stored in the ROM is being processed by the processor 312. The non-volatile memory is a memory into which data is written and read by the execution of the program, and the data written therein is stored even after the execution of the program is terminated. In particular, each process described in the process flow shown in FIG. 8A and FIG. 8B is executed (details of the process are explained in FIG. 8A and FIG. 8B). Although not particularly shown as the memory 313, it may be connected to a removable storage medium, a database, etc. via the input interface 315.

[0057] The communication interface 314 functions as a communication unit that transmits and receives information to and from the server device 200 and other terminal devices installed remotely via a communication processing circuit and an antenna. The communication processing circuit performs processing for receiving, from the server device 200, a program for executing the processing according to the embodiment and various information used in the processing, etc., according to the progress of the processing. It also performs processing for transmitting various information required for executing the processing to the server device 200. In this embodiment, in particular, content to be checked, a check request, and check results are transmitted and received between the server device 200 and the communication processing circuit.

[0058] The communication processing circuit processes based on a wideband wireless communication system such as the LTE system, but it is also possible to process based on a system related to narrowband wireless communication such as wireless LAN such as IEEE802.11 or Bluetooth (registered trademark). Also, wired communication can be used instead of or in addition to wireless communication.

[0059] The input interface 315 is composed of a touch panel and / or hard keys, and accepts operation input from a user. The touch panel is arranged to cover the display as the output interface 311, and outputs information on position coordinates corresponding to image data displayed on the display to the processor 312. As the touch panel method, known methods such as a resistive film method, a capacitive coupling method, and an ultrasonic surface acoustic wave method can be used. In the present embodiment, as an example, the touch panel detects a tap operation or a swipe operation for selecting or moving an object displayed on the display by a pointer. Note that the touch panel is an example of the input interface 315, and naturally other types can be used. It is also possible to use an input interface (such as a so-called keyboard or mouse) connected wirelessly or by wire to a main body including the processor 312, etc.

[0060] 6. Configuration of the second checker terminal device 400 Fig. 6 is a block diagram showing an example of the configuration of the second checker terminal device 400 according to an embodiment of the present disclosure. The second checker terminal device 400 does not need to include all of the components shown in Fig. 6, and may have a configuration in which some components are omitted, or may include other components.

[0061] An example of the second checker terminal device 400 is a laptop PC, but the second checker terminal device 400 can also be suitably applied to any other device capable of executing the processing program according to the present disclosure, such as a portable terminal device capable of wireless communication, such as a smartphone, a feature phone, a personal digital assistant, a PDA, a desktop computer, a portable game machine, or a stationary game machine.

[0062] 6, the second checker terminal device 400 includes an output interface 411, a processor 412, a memory 413 including RAM, ROM, non-volatile memory (HDD in some cases), etc., a communication interface 414 including a communication processing circuit and an antenna, and an input interface 415 including a touch panel, hard keys, etc. These components are electrically connected to each other via control lines and data lines.

[0063] The output interface 411 functions as an output unit that outputs, in response to an instruction from the processor 412, an image captured by a camera (not shown) or image information stored in the memory 413, and various displays output by executing the program according to the present disclosure to a device such as a display or a printer. An example of the output interface 411 is a display configured with a liquid crystal display or an organic EL display, but the second checker terminal device 400 itself does not necessarily need to be equipped with a display. For example, the output interface 411 may output information for various displays to a display connected by wire or wirelessly. In other words, the communication interface 414 may function as the output interface 411.

[0064] The processor 412 is composed of a GPU (graphics processing unit) or a CPU (microcomputer), and functions as a control unit that controls other components connected thereto based on various programs stored in the memory 413. Specifically, the processor 412 reads out from the memory 413 a program for executing the processing according to this embodiment and a program for executing the OS, and executes them. In the present disclosure, the processor 412 particularly executes each process shown in the processing sequence of FIG. 8A and FIG. 8B (details of the process will be described in FIG. 8A and FIG. 8B). The processor 412 may be composed of a single CPU or multiple CPUs. In addition, other types of processors, such as a GPU specialized for image processing, may be appropriately combined.

[0065] The memory 413 is composed of a ROM, a RAM, a non-volatile memory, a HDD, etc., and functions as a storage unit. The ROM stores instructions for executing the game application and the OS according to this embodiment as a program. The RAM is a memory used for writing and reading data while the program stored in the ROM is being processed by the processor 412. The non-volatile memory is a memory into which data is written and read by the execution of the program, and the data written therein is stored even after the execution of the program is terminated. In particular, each process described in the process flow shown in FIG. 8A and FIG. 8B is executed (details of the process are explained in FIG. 8A and FIG. 8B). Although not particularly shown as the memory 413, it may be connected to a removable storage medium, a database, etc. via the input interface 415.

[0066] The communication interface 414 functions as a communication unit that transmits and receives information to and from the server device 200 and other terminal devices installed remotely via a communication processing circuit and an antenna. The communication processing circuit performs processing for receiving, from the server device 200, a program for executing the processing according to this embodiment and various information used in the processing, etc., according to the progress of the processing. It also performs processing for transmitting various information required for executing the processing to the server device 200. In this embodiment, in particular, content to be checked, a check request, and check results are transmitted and received between the server device 200 and the communication processing circuit.

[0067] The communication processing circuit processes based on a wideband wireless communication system such as the LTE system, but it is also possible to process based on a system related to narrowband wireless communication such as wireless LAN such as IEEE802.11 or Bluetooth (registered trademark). Also, wired communication can be used instead of or in addition to wireless communication.

[0068] The input interface 415 is composed of a touch panel and / or hard keys, and accepts operation input from a user. The touch panel is arranged to cover the display as the output interface 411, and outputs information on position coordinates corresponding to image data displayed on the display to the processor 412. As the touch panel method, known methods such as a resistive film method, a capacitive coupling method, and an ultrasonic surface acoustic wave method can be used. In the present embodiment, as an example, the touch panel detects a tap operation or a swipe operation for selecting or moving an object displayed on the display by a pointer. Note that the touch panel is one example of an input interface, and naturally other types can be used. It is also possible to use an input interface (such as a so-called keyboard or mouse) connected wirelessly or by wire to a main body including the processor 412, etc.

[0069] 7. Information stored in the memory 211 of the server device 200 7A is a diagram conceptually illustrating a first database for screening stored in the server device 200 according to an embodiment of the present disclosure. Information stored in the first database for screening is updated from time to time in accordance with various laws, regulations, and guidelines.

[0070] According to Fig. 7A, the first database for review stores category information, target information, propriety information, reason information, and alternative information in a mutually associated manner. Generally, the content of a regulation varies depending on the product that is subject to the regulation, so the "category information" is information indicating the category of the product that is subject to the regulation. Note that the product category is an example of category information, and any information for classifying the subject of the regulation may be used.

[0071] "Target information" is data (text in this example) included in the content to be reviewed, and is information on which the usability described below is judged. For example, if "fine lines," "make fine lines caused by dryness less noticeable," "lose weight just by eating," "have beautiful skin like a baby," or "grow taller easily" are included in sentences or frames that divide the content at a specified interval, these words or sentences are registered as targets for judgment. In other words, the words or sentences included in the content are subject to judgment. Note that the specific data in the content to be reviewed is not limited to those mentioned above, and various data (characters, symbols, sentences, images, videos, etc.) will be selected depending on the category of the target product.

[0072] The "possible / inappropriate information" is information indicating the judgment result when the data is used for the product indicated in the category. "Inappropriate" indicates that the expression is subject to regulation by laws and guidelines. "Caution required" indicates that the expression may be subject to regulation depending on how it is expressed, or that it is an expression for which a clear judgment cannot be made at present. "Appropriate" indicates that the expression is in accordance with laws and guidelines, or is not subject to regulation by these. Although three pieces of information, "inappropriate," "caution required," and "appropriate," are stored as the possible / inappropriate information, it is not limited to these pieces of information. For example, it may be only two pieces of information, "inappropriate and appropriate," "inappropriate and caution required," or "appropriate and caution required," or it may be only one of the pieces of information, or it may be further subdivided into a large number of pieces of information, or it may be information of a numerical value or level indicating the probability of being judged as inappropriate or appropriate. In addition, the wording may be any wording, such as NG, gray, OK, O (circle), × (x), △ (triangle), or numbers.

[0073] The "reason information" is information indicating the reason why the content was judged to be inappropriate or to require caution. Examples of the reasons include "inappropriate wording" or "cannot be determined to result in weight loss." The reasons are not limited to those described above, and various reasons are set according to the category of the target product, the subject of review, and the data to be judged. That is, if the subject of review is text such as characters, an example such as that shown in FIG. 7A is given, but for symbols, images, or videos, a specific reason corresponding to each subject of review may be set.

[0074] "Alternative information" is information showing alternatives to avoid receiving such a judgment result for an object judged to be inappropriate or requiring caution. For example, if the judgment object is judged to be inappropriate, information related to "delete or modify" corresponds to the information, and if the judgment object is judged to require caution, information calling for specific caution is set correspondingly.

[0075] The first database for screening can be used for the process of determining the object of determination, the process of determining the reason for the determination, and the process of selecting an alternative. When a trained model is used in each process, the first database for screening can be used as training data for each trained model.

[0076] 7B is a diagram conceptually illustrating a second database for screening stored in the server device 200 according to an embodiment of the present disclosure. Information stored in the second database for screening is updated from time to time in accordance with various laws, regulations, and guidelines.

[0077] According to Fig. 7B, the second database for review stores category information, target information, before information, after information, yes / no information, and alternative information in a mutually associated manner. Generally, the content of a regulation varies depending on the product that is subject to the regulation, so the "category information" is information indicating the category of the product that is subject to the regulation, as in the first database for review. Note that the product category is an example of category information, and any information for classifying the subject of the regulation may be used.

[0078] "Target information" is data contained in the content to be reviewed, similar to the first database for review. For example, if "collagen," "skin," or "oxidation" are contained in sentences or frames that divide the content at a predetermined interval, these words may be subject to judgment. Note that the specific data in the content to be reviewed is not limited to those mentioned above, and various data (characters, symbols, sentences, images, videos, etc.) will be selected depending on the category of the target product.

[0079] The "previous information" is information about a word located before the data (target information) included in the content to be examined. On the other hand, the "following information" is information about a word located after the data (target information) included in the content to be examined. For example, when "oxidation" is the target information, the word "skin" located before the word, the word "of" or "to" located after the word, and the word "reduction" or "prevention" located further after the "of" or "to" are registered. That is, a word or sentence composed of any combination of the target information, the previous information, and the following information is registered as a judgment target. For example, when "skin oxidation", "reduction of skin oxidation", or "prevention of skin oxidation" are included in a sentence or frame obtained by dividing the content at a predetermined interval, these words or sentences can be judgment targets. Similarly, the word "of" or "to" located after "collagen", and the word "production" or "generation" located further after the "of" or "to" are registered, and the word "care" or "clinic" located after "skin" are registered. By making such registrations, it becomes possible to include words and sentences that do not exactly match the words or sentences registered in the first database for review, but which have a similar meaning and are therefore considered to be inappropriate to use, as subject to judgment.

[0080] The "acceptance / rejection information", like the first database for review, is information showing the judgment result when the data shown in the judgment target is used for the product shown in the category. The "alternative information", like the first database for review, is information showing alternatives to avoid receiving such a judgment result when the judgment target is judged to be inappropriate or requiring caution. For example, when the judgment target is judged to be inappropriate data, information related to "deletion or correction" corresponds, and when the judgment target is judged to require caution, information calling for specific cautions is set correspondingly.

[0081] The second database for screening can be used for the process of determining the object of screening, the process of determining the reason for the screening, and the process of selecting an alternative, in the same manner as the first database for screening. When a trained model is used in each process, the second database for screening can be used as training data for each trained model.

[0082] In addition, in the second database for review in Figure 7B, "reason information" is not stored, but as in the first database for review, information indicating the reason why it was judged to be inappropriate or requiring caution may be stored as "reason information."

[0083] As described above, a normalized database used to determine whether or not a specific expression (regular expression) can be used is configured from a first database for screening in which a specific word or phrase is registered as unusable as shown in Fig. 7A, and a second database for screening in which a word or phrase consisting of a specific word and any combination of the words before and after the specific word is registered as unusable as shown in Fig. 7B. Note that, when a still image or a video is included in the subject of the determination, a database in which the usability of the still image or video is registered is also included in the normalized database.

[0084] 7C is a diagram conceptually illustrating a content table stored in the server device 200 according to an embodiment of the present disclosure. The content table is generated every time content is received from the requester terminal device 100, and is updated as needed.

[0085] According to FIG. 7C, the content table stores content ID information, category information, review target information, location information, yes / no information, reason information, alternative information, and the like in a mutually associated manner. The "content ID information" is information for identifying each content to be reviewed, and is information unique to each content. The content ID information is newly generated each time new content is transmitted from the requester terminal device 100 to the server device 200 and stored in the memory 211 of the server device 200. The "category information" is information for identifying the category of the product in which the content to be reviewed is used as an advertisement. The "review target information" is various data (characters, symbols, sentences, images, videos, etc.) included in each content. The data stored here may be, for example, text data itself, or a location where the text data is stored or information identifying the text data. In addition, when image data is included in the content, the data may be the image data itself, or a location where the image data is stored or information identifying the image data. Here, in the embodiment, it is assumed that the content includes at least one piece of data, so that multiple pieces of data may be stored in association with the same content ID information.

[0086] "Location information" is information for specifying the position where each data is placed in the content. The information may be any information that can specify the rough position of each image, such as the coordinates where each image is placed in the content or frame, the number of pixels from the edge of the content or frame, the number of rows or columns, etc. "Possibility information," "reason information," and "alternative information" are all types of judgment result information, and each piece of information explained in FIG. 7A is stored in association with each piece of data. Each time processor 212 performs data judgment processing, judgment processing of the reason for judgment, and alternative selection processing, this information is stored in association with the data that was the subject of the judgment.

[0087] 8. Processing sequence performed by processing system 1 The processing sequence executed in the processing system 1 will be specifically described below with reference to FIGS. 8A and 9. FIG.

[0088] (A) Content input, review of the content and reporting of the results Fig. 8A is a diagram showing a processing sequence executed among the requester terminal device 100, the server device 200, the first checker terminal device 300, and the second checker terminal device 400 according to the embodiment of the present disclosure. Specifically, Fig. 8A is a diagram showing a processing sequence executed among the requester terminal device 100, the server device 200, the first checker terminal device 300, and the second checker terminal device 400 from input of the content to be examined through actual examination (judgment and checking) to returning and storing the final examination result.

[0089] 8A, first, the processor 112 of the requester terminal device 100 accepts an operation input from a user (a person requesting the examination) via the input interface 115, and acquires the content to be examined (S10: Acquire content). After that, the processor 112 of the requester terminal device 100 stores content information related to the content in the memory 113, and transmits the content information to the server device 200 via the communication interface 114 (T11).

[0090] As a more specific process related to content acquisition, when the content screening tool is started in the requester terminal device 100, a content registration screen as shown in FIG. 8B is displayed on the display of the requester terminal device 100. According to FIG. 8B, the content registration screen includes a format selection area 121 for selecting the format of the content to be screened, a storage location input area 122 indicating the storage location of the content to be screened, a category input area 123 for the category of the product advertised by the content to be screened, and a registration start button 124 for sending a registration start instruction to the server device 200. A user who is a user of the requester terminal device 100 selects the format of the content to be judged from the format selection area 121 via the input interface 115. When the selection is made, the corresponding check box is displayed in black. Next, the user selects the storage location input area 122 corresponding to the content to be screened via the input interface 115 and inputs the storage location. Furthermore, the user inputs the category of the product advertised by the content to be screened in the category input area 123 via the input interface 115. Thereafter, when the user selects the registration start button 124, the content information stored in the storage location input in the storage location input area 122 is read out together with a registration start instruction and transmitted to the server device 200.

[0091] Next, the processor 212 of the server device 200 stores the received content information in the memory 211. That is, the processor 212 of the server device 200 registers the content information and generates a content table as shown in Fig. 7C (S11: content registration).

[0092] Next, the processor 112 of the requester terminal device 100 accepts an operational input from the user (requester of the review) via the input interface 115, and sets a review request for the content to be reviewed (S12: review request setting). The subject of the review request setting may be, for example, information on the content to be reviewed, a desired date for the review result reply, and whether or not an alternative proposal has been proposed, and various items related to the review of the content may be set. Thereafter, the processor 112 of the requester terminal device 100 transmits the review request and setting information for the content to the server device 200 via the communication interface 114 (T12).

[0093] Next, the processor 212 of the server device 200 identifies the content to be screened from the content database based on the received screening request and setting information, and executes a primary screening process for the data included in the content (S13). Specifically, the processor 212 reads out and executes a screening program stored in the memory 211, and performs a plurality of primary screening processes. Details of the primary screening process will be described later with reference to FIG. 10A. The processor 212 of the server device 200 then stores the primary screening results in the memory 211.

[0094] Next, the processor 212 of the server device 200 transmits a check request related to the primary judgment result to the first checker terminal device 300 and the second checker terminal device 400 via the communication interface 213 (T13, T14). Here, the check request may include content information including the judged data, information related to the setting of the examination request, and information related to the primary judgment result.

[0095] Next, the processor 312 of the first checker terminal device 300 accepts the check request via the communication interface 314, and acquires the primary judgment result to be checked and the content information to be checked, etc. Then, the processor 312 of the first checker terminal device 300 displays various information related to the check target via the output interface 311, and makes it possible for the first checker who is the operator of the first checker terminal device 300 to confirm and check the information. This enables the first checker, such as a lawyer who is legally authorized to perform checks, to perform the first check process based on the information displayed via the output interface 311 (S14).

[0096] Here, FIG. 8C is a diagram showing an example of a screen output to the first checker terminal device 300 according to the embodiment of the present disclosure. Specifically, FIG. 8C is a screen including the primary judgment result received from the server device 200, for enabling the first checker to perform the check process, and is a check process screen when output to a display via the output interface 311. According to FIG. 8C, the check screen includes a content display area 330 in which the content to be examined is displayed. The content to be examined is displayed in this content display area 330. The content includes five pieces of data that are the subject of the judgment for which the usability of the content in the content has been judged. More specifically, the content includes a data display area 331 set to surround the text T1 with a dashed line, a data display area 332 set to surround the text T2 with a dashed line, a data display area 333 set to surround the text T3 with a dashed line, a data display area 334 set to surround the text T4 with a dashed line, and a data display area 335 set to surround the text T5 with a dashed line. This allows the first checker to visually confirm the data to be checked.

[0097] Also, according to FIG. 8C, a determination result list display area 340 is displayed on the right side of the content display area 330. In this determination result list display area 340, object boxes 341-345 are displayed corresponding to each text data (text T1-T5) that was determined. The object boxes 341-345 contain the availability information, reason information, and alternative information associated with each text data, but for the text data (text T2-T5) that was not selected, only information indicating the explanation of the text data that was determined is displayed in a reduced size. Also, each object box 341-345 is displayed in a manner that makes it possible to identify the result of the determination of the availability of the use based on the availability information associated with each text data. For example, in the example of FIG. 8C, "text T1" is displayed as the availability information, and this object box 341 is displayed in a manner surrounded by a thick line. On the other hand, the other object boxes 342-345 that were not selected are displayed in a manner surrounded by a dashed line.

[0098] A first checkbox 350 for indicating that a check has been checked by a lawyer who is a first checker is provided at the bottom of the content display area 330. A second checkbox 360 for indicating that a check has been checked by a second checker who is a regular person assigned to practically check the contents is provided adjacent to the first checkbox 350.

[0099] Then, the first checker performs a first check process on the screen shown in Fig. 8C regarding whether the text data to be judged can be used or not, taking into consideration the primary judgment result. More specifically, since the primary judgment results are displayed in the object boxes 341-345, the first checker refers to the object boxes 341-345 to confirm the primary judgment result, and checks the judgment result by the first checker. Then, if the judgment result needs to be corrected, the information in the object boxes 341-345 is corrected by operating the first checker via the input interface 315. When checking of all the text data to be judged is completed, the first checker operates the first checker via the input interface 315 to check the first check box 350.

[0100] In addition, if there is text data that has not been determined as a target for judgment in the content display area 330 displayed on the first checker terminal device 300, an additional judgment process may be performed by the first checker using the text data that was not determined as a target as a target for judgment, and additional processing may be performed in which the judgment result is included in the check result.

[0101] Thereafter, when the first checker completes the first check process (S14), the processor 312 of the first checker terminal device 300 transmits the check result to the server device 200 via the communication interface 314 (T15).

[0102] Next, the processor 412 of the second checker terminal device 400 accepts the check request via the communication interface 414 and acquires the primary judgment result to be checked and the content information to be checked, etc. Then, the processor 412 of the second checker terminal device 400 displays various information related to the check target via the output interface 411, enabling the second checker, who is the operator of the second checker terminal device 400, to confirm and check the information. This enables the second checker to perform the second check process based on the information displayed via the output interface 411 (S15).

[0103] The check processing screen displayed on the second checker terminal device 400 is also the same as that shown in Fig. 8C. Therefore, although detailed explanations of the screen and the processing will be omitted, the second checker can perform the check processing in the same manner as the first checker. However, since it is assumed that the second checker is not a lawyer, the second checker will check the second check box 360. Then, when the second checker completes the second check processing (S15), the processor 412 of the second checker terminal device 400 transmits the check result to the server device 200 via the communication interface 414 (T16).

[0104] The first check process by the first checker (S14) and the second check process by the second checker (S15) may be performed in the opposite order, or may be performed simultaneously.

[0105] Next, the processor 212 of the server device 200 stores in the memory 211 the result of the first check received from the first checker terminal device 300 (S16: first check registration). The processor 212 of the server device 200 also stores in the memory 211 the result of the second check received from the second checker terminal device 400 (S17: second check registration). After that, the processor 212 of the server device 200 executes a secondary judgment process for the data included in the content based on each of the registered check results and the result of the primary judgment process (S18). Details of the secondary judgment process will be described later with reference to FIG. 10B.

[0106] Next, the processor 212 of the server device 200 stores the secondary judgment result in the memory 211 as the final review result, and manages the review result (S19). Here, management corresponds to deleting the review result after a certain period of time has passed, performing AI maintenance using the review result, which will be described later, or performing a re-review, etc. After that, the processor 212 of the server device 200 transmits the final review result to the client terminal device 100 via the communication interface 213 (T17). Here, information indicating that the final review result has been checked by a lawyer or the like who is legally authorized to perform a check is added to the final review result.

[0107] Next, the processor 112 of the client terminal device 100 receives the examination result via the communication interface 114, and acquires various information related to the examination result. The processor 112 of the client terminal device 100 then displays the final examination result via the output interface 111, and makes it possible for the person requesting the examination, who is the operator of the client terminal device 100, to confirm it (S20: Output of examination result). This allows the person requesting the examination to know whether the various information included in the content is usable based on the restrictions of predetermined laws and regulations or guidelines.

[0108] Here, FIG. 8D is a diagram showing an example of a screen output to the client terminal device 100 according to the embodiment of the present disclosure. Specifically, FIG. 8D is a screen displaying the examination result received from the server device 200, and is the examination result display screen when output to the display via the output interface 111. According to FIG. 8D, the examination result display screen includes a content display area 130 in which the contents that have been the subject of examination are displayed. The contents that have been the subject of examination are displayed in the content display area 130. The contents include five pieces of data that are the subject of the judgment for whether or not the contents can be used. More specifically, the contents include a data display area 131 that is set to surround the text T1 with a dashed line, a data display area 132 that is set to surround the text T2 with a dashed line, a data display area 133 that is set to surround the text T3 with a dashed line, a data display area 134 that is set to surround the text T4 with a dashed line, and a data display area 135 that is set to surround the text T5 with a dashed line. This allows the person requesting the review to visually confirm the entire content that is the subject of review and each piece of data included in that content.

[0109] Also, according to FIG. 8D, a determination result list display area 140 is displayed on the right side of the content display area 130. In this determination result list display area 140, object boxes 141-145 are displayed corresponding to each text data (text T1-T5) that was determined. The object boxes 141-145 contain the availability information, reason information, and alternative information associated with each text data, but for the text data (text T2-T5) that was not selected, only information indicating the explanation of the text data that was determined is displayed in a reduced size. Also, each object box 141-145 is displayed in a manner that makes it possible to identify the result of the determination of the availability of the use based on the availability information associated with each text data. For example, in the example of FIG. 8D, "text T1" is displayed as the availability information, and this object box 141 is displayed in a manner surrounded by a thick line. On the other hand, the other object boxes 142-145 that were not selected are displayed in a manner surrounded by a dashed line.

[0110] Also, an inappropriate part display area 151 and a caution part display area 152 are displayed at the top of the content display area 130. The number of data items that are judged to be inappropriate in the content that is the subject of review are added up based on the propriety information and displayed in the inappropriate part display area 151. The number of data items that are judged to be caution-requiring in the content that is the subject of review are added up based on the propriety information and displayed in the caution part display area 152.

[0111] Also, at the top of the judgment result list display area 140, an approval display area 160 is displayed, which indicates that a person legally authorized has approved the use of the data related to the content. In other words, the displayed examination results are displayed to indicate that they have been checked by a lawyer or the like who is legally authorized to examine. This makes it clear that the examination is legally authorized. Note that the name and signature of the lawyer who is the first checker may be written in the approval display area 160.

[0112] (B) Maintenance of trained models and databases Fig. 9 is a diagram showing a processing sequence executed among the server device 200, the first checker terminal device 300, and the second checker terminal device 400 according to the embodiment of the present disclosure. Specifically, Fig. 9 is a diagram showing a processing sequence executed among the server device 200, the first checker terminal device 300, and the second checker terminal device 400 with respect to a trained model (AI: artificial intelligence) used to process data included in content to be examined, and the maintenance of the normalized database described above.

[0113] 9, first, the processor 212 of the server device 200 collects content including at least one data to be determined, and registers the collected content in the memory 211 (S21). For example, the processor 212 of the server device 200 may execute a content collection program stored in the memory 211, and collect various content existing on the Internet via the communication interface 213. The processor 212 of the server device 200 may also collect content through input by an operator of the server device 200 or the like via the input interface 214. Then, the processor 212 of the server device 200 generates a content table as shown in FIG. 7C.

[0114] Next, the processor 212 of the server device 200 executes a primary analysis process for the data included in the collected content (S22). Specifically, the processor 212 reads out and executes a program for judgment stored in the memory 211, and performs a plurality of primary analysis processes, similar to the primary judgment process (S13) of FIG. 8A. Here, the primary analysis process has a different name to distinguish it from the primary judgment process, but it is the same process as the primary judgment process, in that it judges whether the data included in the collected content can be used based on the restrictions of a predetermined law or guideline. Therefore, a detailed description thereof will be omitted. The processor 212 of the server device 200 then stores the primary analysis result in the memory 211.

[0115] Next, the processor 212 of the server device 200 transmits a check request related to the primary analysis result to the first checker terminal device 300 and the second checker terminal device 400 via the communication interface 213 (T21, T22). Here, the check request may include content information including the analyzed (determined) data, and information related to the primary analysis result.

[0116] Next, the processor 312 of the first checker terminal device 300 accepts the check request via the communication interface 314 and acquires the primary analysis result and the content information to be checked. Then, the processor 312 of the first checker terminal device 300 displays various information related to the check target via the output interface 311, and makes it possible for the first checker who is the operator of the first checker terminal device 300 to confirm and check the information. This enables the first checker, such as a lawyer who is legally authorized to perform checks, to perform the first check process based on the information displayed via the output interface 311 (S23). Note that the first check process in S23 and the first check process in S14 of FIG. 8A have the same content, so detailed description will be omitted.

[0117] Thereafter, when the first checker completes the first check process (S23), the processor 312 of the first checker terminal device 300 transmits the check result to the server device 200 via the communication interface 314 (T23).

[0118] Next, the processor 412 of the second checker terminal device 400 accepts the check request via the communication interface 414 and acquires the primary analysis result to be checked and the content information to be checked. Then, the processor 412 of the second checker terminal device 400 displays various information related to the check target via the output interface 411, enabling the second checker, who is the operator of the second checker terminal device 400, to confirm and check the information. This enables the second checker to perform the second check process based on the information displayed via the output interface 411 (S24). Note that the second check process in S24 and the second check process in S15 of FIG. 8A have the same content, so detailed description will be omitted.

[0119] Thereafter, when the second checker completes the second check process (S24), the processor 412 of the second checker terminal device 400 transmits the check result to the server device 200 via the communication interface 414 (T24).

[0120] The first check process by the first checker (S23) and the second check process by the second checker (S24) may be performed in the opposite order, or may be performed simultaneously.

[0121] Next, the processor 212 of the server device 200 stores the result of the first check received from the first checker terminal device 300 in the memory 211 (S25: first check registration). The processor 212 of the server device 200 also stores the result of the second check received from the second checker terminal device 400 in the memory 211 (S26: second check registration). After that, the processor 212 of the server device 200 executes a secondary analysis process for the data included in the content based on the registered check results and the result of the primary analysis process (S27). Note that the details of the secondary analysis process are the same as those of the secondary judgment process in S18 of FIG. 8A, and therefore will not be described here.

[0122] Next, the processor 212 of the server device 200 stores the secondary analysis result in the memory 211 as the final analysis result (S28: Accumulation of analysis result). Note that by collecting and registering a plurality of contents in S21, the analysis results for the plurality of contents are stored in the memory 211, and analysis results for various patterns of contents are accumulated.

[0123] Thereafter, the processor 212 of the server device 200 performs maintenance of the trained model using the accumulated analysis results (S29: AI maintenance). Specifically, the processor 212 of the server device 200 performs re-learning of the trained model using the accumulated analysis results as training data, and generates a more optimal model. A more detailed description of the AI ​​maintenance will be given later when describing FIGS. 11A and 11B.

[0124] Then, the processor 212 of the server device 200 performs maintenance of the normalized database using the accumulated analysis results (S30: DB maintenance). Specifically, the processor 212 of the server device 200 adds and registers new data, etc., which are highly likely to be determined as not permitted to be used in the content in the accumulated analysis results, to the normalized database. A more detailed description of the DB maintenance will be given later in the description of FIG. 12.

[0125] 9. Processing flow executed in the server device 200 Below, we will specifically explain the processing flows executed by the server device 200 for the primary judgment processing (primary analysis processing), secondary judgment processing (secondary analysis processing), AI maintenance processing, and DB maintenance processing described in Figures 8A and 9.

[0126] (A) Processing flow for the first judgment process (first analysis process) Fig. 10A is a diagram showing a processing flow executed in the server device 200 according to an embodiment of the present disclosure. Specifically, Fig. 10A is a diagram showing an example of the primary determination processing of S13 in Fig. 8A and the primary analysis processing of S22 in Fig. 9, and is a diagram showing a processing flow using various trained models and normalized databases stored in the memory 211.

[0127] According to Fig. 10A, first, the processor 212 performs content analysis (S111). Specifically, the processor 212 uses a content analysis model stored in the memory 211 to analyze the content to be examined or analyzed. For example, the processor 212 inputs content including data that may be subject to judgment (characters, symbols, text, images, videos, etc.) into the content analysis model, and obtains the extraction result of the data included in the content as output information. As a more specific example, when a web page with textual advertisement is input as content, the text of the advertisement portion is extracted as data.

[0128] Next, processor 212 performs sentence division (S112). Specifically, processor 212 performs processing for dividing the data extracted in S111 into sentences using a sentence division model stored in memory 211. For example, processor 212 inputs the output result in S111 (one of the extracted data) into the sentence division model, and obtains the results of dividing the data into smaller sentences as output information. As a more specific example, when a sentence of the advertisement portion extracted from a web page is input into the sentence division model, the sentence is divided into predetermined sentences by periods, and each divided sentence is output.

[0129] Next, the processor 212 performs an AI judgment process (S113). Specifically, the processor 212 uses the judgment process model stored in the memory 211 to judge whether or not each sentence divided in S112 is usable based on the restrictions of a specific law or guideline. For example, the processor 212 inputs the output result in S112 (the divided sentence) into the judgment process model, and obtains the result of whether or not the sentence complies with the specific law or guideline as output information. As a more specific example, when a part of the text of the advertising portion divided into sentences is input into the judgment process model, words or expressions that do not comply or may not comply with the law or guideline are extracted from the text, and the reason is also output.

[0130] Next, the processor 212 performs DB comparison and judgment processing (S114). Specifically, the processor 212 uses the normalized database stored in the memory 211 to judge whether the data in the content after the AI ​​judgment processing can be used based on the restrictions of a predetermined law or guideline. For example, the processor 212 compares the data in the content with the data registered in the normalized database, and searches for and extracts words or expressions in the data that do not comply or may not comply with the law or guideline. At this time, if there is a word or expression in the data that does not comply or may not comply with the law or guideline, the processor 212 will also read out the reason from the normalized database.

[0131] Then, the processor 212 combines the result of the AI ​​determination process (S113) and the result of the DB comparison determination process (S114), and stores them in the memory 211 as the result of the primary determination process or the primary analysis process.

[0132] (B) Processing flow for secondary judgment processing (secondary analysis processing) Fig. 10B is a diagram showing a processing flow executed in the server device 200 according to the embodiment of the present disclosure. Specifically, Fig. 10B is a diagram showing an example of the secondary determination processing of S18 in Fig. 8A and the secondary analysis processing of S27 in Fig. 9, and is a diagram showing a processing flow using various trained models stored in the memory 211.

[0133] 10B, first, processor 212 compares the result of the primary judgment process with the result of the first check (S121). Specifically, processor 212 reads out and compares the result of the primary judgment process stored in memory 211 with the result of the first check, and extracts a portion in which the result of the primary judgment process has been modified or added by performing the first check. As a more specific example, when data that was determined to be appropriate for use in the primary judgment result is modified to inappropriate by the first check, the judgment related to the data is extracted.

[0134] Next, processor 212 compares the result of the primary judgment process with the result of the second check (S122). Specifically, processor 212 reads out and compares the result of the primary judgment process stored in memory 211 with the result of the second check, and extracts a portion in which the result of the primary judgment process has been modified or added by performing the second check. As a more specific example, when data that was determined to be appropriate for use in the primary judgment result is modified to be inappropriate by the second check, the judgment related to the data is extracted.

[0135] Next, the processor 212 performs a comprehensive judgment process taking into consideration the comparison result in S121, the comparison result in S122, and the result of the primary judgment process (S123: comprehensive judgment process). For example, when two of the results of the primary judgment process, the first check result, and the second check result are the same but the remaining one is different, the processor 212 may preferentially adopt the two results that are the same. Also, when the three results are different, the processor 212 may determine the final judgment result according to a priority order registered beforehand (for example, first checker>second checker>server device 200).

[0136] Next, the processor 212 performs an alternative selection process (S124). Specifically, the processor 212 uses a model for selecting alternatives stored in the memory 211 to select an alternative that can comply with the regulations of a specific law or guideline for data whose use is determined to be inappropriate in the result of the comprehensive judgment in S123. For example, the processor 212 inputs the result of the comprehensive judgment in S123 (information including the content, data, and propriety judgment) into the model for selecting alternatives, and obtains an alternative for making the sentence conform to the specific law or guideline as output information. The processor 212 then adds the alternative to the final judgment result in S123, and stores it in the memory 211 as the final review result for the content.

[0137] (C) Processing flow for AI maintenance FIG. 11A is a diagram conceptually illustrating generation management of a trained model stored in a server device 200 according to an embodiment of the present disclosure. FIG. 11B is a diagram illustrating a process flow executed in the server device 200 according to an embodiment of the present disclosure. Specifically, FIG. 11B illustrates a process flow performed to update a trained model. Here, in this embodiment, a model for content analysis, a model for sentence decomposition, a model for judgment processing, and a model for alternative selection are stored as trained models, and model management and model updates described below are performed for each of these models.

[0138] First, as shown in FIG. 11A, each trained model is based on a primary model (first generation model) generated in an initial stage, and is updated through additional learning using new teacher data. Specifically, the primary model is updated to a 2A model, which is one of the second generation models, through additional learning using teacher data α for the primary model. Furthermore, the 2A model is updated to a 3A model, which is one of the third generation models, through additional learning using teacher data α1 for the 2A model.

[0139] On the other hand, when additional learning is performed on the primary model using teacher data β different from teacher data α, the primary model is updated to the 2B model, which is one of the second generation models. Furthermore, when additional learning is performed on the 2B model using teacher data β1, the 2B model is updated to the 3B-1 model, which is one of the third generation models, and when additional learning is performed using teacher data β2 different from teacher data β1, the 2B model is updated to the 3B-2 model, which is one of the third generation models.

[0140] In this embodiment, each model is stored in the memory 211 of the server device 200 and can be used or updated as appropriate. That is, if the 2A model is updated but an accurate judgment cannot be made with the 2A model, the model can be returned to the primary model and re-learned using different teacher data to regenerate another trained model of the second generation. That is, in this embodiment, re-learning is performed using a model of a previous generation, and it is possible to generate multiple models of the same generation but different models later. Then, each judgment by various trained models can be verified, and the optimal trained model can be selected to perform judgment and examination with higher accuracy.

[0141] Next, how to update a trained model, i.e., regeneration of a trained model, in managing a trained model as shown in FIG. 11A will be described. As shown in FIG. 11B, the processor 212 of the server device 200 reads out the analysis result accumulated in S28 of FIG. 9 from the memory 211 (S211). After that, the processor 212 of the server device 200 extracts the category, additional learning content, additional learning data, review results, and alternatives contained in the analysis result (S212). Here, the category, additional learning content, additional learning data, review results, and alternatives are associated as information related to one content.

[0142] Next, the processor 212 of the server device 200 regenerates the model for content analysis (S213). For example, the processor 212 of the server device 200 uses the extracted additional learning content and additional learning data as teacher data to perform additional learning of the model for content analysis to be re-learned. That is, the model for content analysis is re-learned so that additional learning data is extracted when the additional learning content is input.

[0143] Next, the processor 212 of the server device 200 regenerates the model for sentence decomposition (S214). For example, the processor 212 of the server device 200 uses the extracted additional learning content, additional learning data, and the examination result as teacher data to perform additional learning of the model for sentence decomposition to be re-learned. That is, when a certain additional learning data is input, the model for sentence division is re-learned so that it can be divided into data in units of sentences that are the subject of judgment in the examination result.

[0144] Next, the processor 212 of the server device 200 regenerates the model for the judgment process (S215). For example, the processor 212 of the server device 200 uses the extracted category information, additional learning content, additional learning data, and the examination result as teacher data to perform additional learning of the model for the judgment process to be re-learned. That is, the model for the judgment process is re-learned so that a judgment is made that a predetermined examination result will be obtained when the additional learning content is input in a predetermined category.

[0145] Next, the processor 212 of the server device 200 regenerates the model for selecting alternatives (S216). For example, the processor 212 of the server device 200 uses the extracted category information, additional learning content, additional learning data, screening results, and alternatives as teacher data to perform additional learning of the model for selecting alternatives to be re-learned. That is, in a specific category, the model for selecting alternatives is re-learned so that a specific alternative is proposed for a specific screening result obtained by inputting additional learning content.

[0146] Next, the processor 212 of the server device 200 performs an accuracy analysis of each regenerated model (S217). For example, the processor 212 of the server device 200 may use other analysis results and review results stored in the memory 211 to use each model, analyze whether the output result of each model matches a result already obtained, and perform the accuracy analysis. The accuracy analysis result is then associated with each model and stored in the memory 211. Thereafter, the processor 212 of the server device 200 selects the model with the best accuracy analysis result, and sets it as the trained model to be used in reviewing the content.

[0147] By performing the above process multiple times, multiple models and the evaluations of each model are accumulated. This makes it possible to reselect a model regardless of generation and to further update the model from the reselected model. This allows the accuracy of the trained model to be continuously improved, and even when unprecedented content or data appears, it becomes possible to perform highly accurate screening.

[0148] (D) Processing flow for DB maintenance Fig. 12 is a diagram showing a process flow executed in the server device 200 according to an embodiment of the present disclosure. Specifically, Fig. 12 shows a process flow performed to update the normalized database.

[0149] First, the processor 212 of the server device 200 reads out past screening results stored in the memory 211 (S311). Next, the processor 212 judges whether there is new data and regular expressions that have been confirmed as NG in the past screening results (S312). Here, the new data and regular expressions refer to those that have not been registered in the database at the time of S312. Specifically, the processor 212 searches for data that has been confirmed as not permitted to be used based on the restrictions of a predetermined law or guideline in the past screening results. Then, the processor 212 judges whether the data extracted by the search is registered in the database of FIG. 7A, and extracts new data that has been confirmed as NG. The processor 212 also judges whether the data extracted by the search is registered in the database of FIG. 7B, and judges whether it is registered as a regular expression.

[0150] Next, if it is determined in S312 that new data and regular expressions that are confirmed as NG exist, the processor 212 extracts alternatives for the data from the above review results (S313: Extract alternatives). Then, the processor 212 adds the data extracted in S312 and the alternatives extracted in S313 to the normalized database, and updates the database (S314: Update DB).

[0151] As described above, in this embodiment, the normalized database is updated using the results of the AI ​​judgment process and the subsequent check process. This makes it possible to extract data that is deemed unacceptable for use by the checker in the primary judgment process, even if it is difficult to judge using the AI ​​judgment process.

[0152] 10. Summary of the embodiments of the present disclosure The processing device according to the embodiment of the present disclosure includes at least one processor 212, and is configured to perform processing for accepting an input of content including at least one data to be judged from a requester terminal device 100, acquiring the data from the content, judging for each of the acquired data whether the data can be used in the content based on the regulation by a predetermined law or guideline, transmitting the judgment result of whether the data can be used and a request for the appropriateness of the judgment result to each checker terminal device (the first checker terminal device 300, the second checker terminal device 400), and transmitting to the requester terminal device 100 a comprehensive result generated based on the judgment result and the check results received from each checker terminal device. With such a configuration, it is possible to more accurately judge whether the content of the data included in the content is subject to the regulation by a predetermined law or guideline.

[0153] In the above configuration, the predetermined laws and regulations or guidelines may stipulate regulations on at least any of drugs, quasi-drugs, regenerative medicine products, medical devices, health foods, functional foods, cosmetics, and mail-order products. This makes it possible to determine with high accuracy whether the content of the data included in the content is subject to regulations related to drugs and devices.

[0154] In the above configuration, the check result may be prepared by a person who is permitted under a specific law or guideline to determine whether the data can be used. Here, the person who is permitted under the specific law or guideline to determine whether the data can be used may be a lawyer. This allows the overall result sent to the client terminal device 100 to include a legally recognized confirmation, and thus eliminates any legal issues regarding the overall result.

[0155] In the above configuration, the at least one processor 212 may generate the judgment result by referring to a database in which an output result obtained by inputting the data into the trained judgment model and the usability of the data are associated. In this case, the at least one processor 212 may perform additional learning on the trained judgment model using the check result, and regenerate a next-generation trained judgment model. The at least one processor 212 may also perform accuracy analysis on the trained judgment models before and after regeneration, and select a model according to the result of the accuracy analysis. On the other hand, the at least one processor 212 may store the trained judgment models before and after regeneration in the memory 211, perform additional learning on the trained judgment model before regeneration using teacher data different from the check result, and regenerate another trained judgment model of the same generation as the trained judgment model before regeneration. With such a configuration, it is possible to more accurately determine whether the content of the data included in the content is subject to regulation by a predetermined law or guideline.

[0156] In the above configuration, the determination result may include an alternative obtained by inputting the data into a trained alternative proposal model, thereby making it possible to provide a more effective screening result to the client.

[0157] A processing program according to an embodiment of the present disclosure, when executed by at least one processor 212, causes the at least one processor 212 to function as follows: accepts input of content including at least one data to be judged from the requester terminal device 100, acquires the data from the content, judges for each of the acquired data whether the data can be used in the content based on the regulation by a predetermined law or guideline, transmits the judgment result of whether the data can be used and a request for the appropriateness of the judgment result to each checker terminal device (the first checker terminal device 300, the second checker terminal device 400), and transmits to the requester terminal device 100 a comprehensive result generated based on the judgment result and the check result received from each checker terminal device. With such a configuration, it is possible to more accurately judge whether the content of the data included in the content is subject to the regulation by a predetermined law or guideline.

[0158] The processing method according to the embodiment of the present disclosure is a processing method executed by at least one processor 212, and includes a step of accepting an input of content including at least one data to be judged from a requester terminal device 100 and acquiring the data from the content, a step of judging whether the data can be used in the content for each of the acquired data based on a regulation by a predetermined law or guideline, a step of transmitting a judgment result of whether the data can be used and a request for the appropriateness of the judgment result to each checker terminal device (the first checker terminal device 300, the second checker terminal device 400), and a step of transmitting a comprehensive result generated based on the judgment result and the check results received from each checker terminal device to the requester terminal device 100. With this configuration, it is possible to judge with higher accuracy whether the content of the data included in the content is subject to regulation by a predetermined law or guideline.

[0159] 11. Variations In the above embodiment, a check request is made to two checkers, but the number of checkers may be one or more than two. In either case, it is important that the checkers include people who are legally permitted to perform confirmation processing (check processing), such as lawyers. In the above embodiment, normal checkers who are not legally permitted to perform confirmation processing are provided, but only people who are legally permitted to perform confirmation processing, such as lawyers, may be set as checkers. In the above embodiment, the processor 212 of the server device 200 performs the secondary judgment processing, but the administrator of the server device 200 may perform the secondary judgment processing based on the primary judgment result and each check result in a manner similar to the checker processing. In this case, a check screen like that of the first checker terminal device 300 will be displayed on the server device 200 as well.

[0160] In the above embodiment, when maintaining the AI ​​model, additional learning is performed using analysis results different from the screening results, but this is not limited to this. For example, additional learning of the set model or another model may be performed using the screening results stored in the memory 211 to generate a new model.

[0161] In the above embodiment, the normalized database is updated by comparing the examination result with the existing normalized database, but the present invention is not limited to this. For example, a trained model capable of updating the normalized database may be generated in advance, and the normalized database may be updated using such an AI model.

[0162] In the above embodiment, the primary judgment process is compared with the first check, and the primary judgment process is compared with the second check during the secondary judgment process (FIG. 10B), but the comparison may not be performed and a comprehensive secondary judgment process may be performed based on the results of the first and second checks. Since the results of the primary judgment process are sent to the first checker terminal device 300 and the second checker terminal device 400, each checker may revise the results of the primary judgment process and omit the comparison process in the secondary judgment process.

[0163] In the above embodiment, the output screens in Figures 8C and 8D are examples, and the display location of each area may be changed as appropriate. For example, the judgment or check result is displayed on the right side of the screen, but it may be displayed on the top, bottom, or left side of the screen. Other display areas can be displayed in optimal locations as appropriate, in accordance with the display locations of other display areas.

[0164] The processes and procedures described herein can be realized not only by those explicitly described in this disclosure, but also by software, hardware, or a combination thereof. Specifically, the processes and procedures described herein can be realized by implementing logic corresponding to the processes in a medium such as an integrated circuit, a volatile memory, a non-volatile memory, a magnetic disk, or an optical storage. In addition, the processes and procedures described herein can be implemented as computer programs and executed by various computers including terminal devices and server devices.

[0165] Although the processes and procedures described herein are described as being executed by a single device, software, component, or module, such processes or procedures may be executed by multiple devices, multiple software, multiple components, and / or multiple modules. Furthermore, although the various information described herein is described as being stored in a single memory or storage unit, such information may be stored in multiple memories provided in a single device or multiple memories distributed across multiple devices. Furthermore, the software and hardware elements described herein may be realized by integrating them into fewer components or breaking them down into more components. [Explanation of symbols]

[0166] 100 Client terminal device 200 Server device 300 First checker terminal device 400 Second checker terminal device

Claims

1. at least one processor; the at least one processor: receiving an input of a content including at least one data to be determined from a client terminal device, and acquiring the data from the content; determining whether or not the data obtained can be used in the content based on regulations under a predetermined law or guideline; Transmitting a result of the determination as to whether the data can be used and a request for the suitability of the determination result to a checker terminal device; transmitting to the requester terminal device a comprehensive result generated based on the judgment result and the check result received from the checker terminal device; A processing device configured to process the

2. The specified laws, regulations or guidelines set out regulations on at least one of pharmaceuticals, quasi-drugs, regenerative medicine products, medical devices, health foods, functional foods, cosmetics and mail order products. The processing device of claim 1 .

3. The check results were prepared by a person authorized under the specified laws, regulations or guidelines to determine whether or not the data can be used. The processing device according to claim 2 .

4. The person authorized to determine whether or not the data can be used under the specified laws, regulations or guidelines is a lawyer; The processing device according to claim 3 .

5. The at least one processor generates the judgment result by referring to a database in which an output result obtained by inputting the data into a trained judgment model and whether or not the data can be used are associated with each other. The processing device according to claim 2 .

6. The at least one processor performs additional learning on the trained judgment model using the check result to regenerate a next-generation trained judgment model. The processing device according to claim 5 .

7. The at least one processor performs an accuracy analysis on the trained decision model before and after regeneration, and selects a model according to a result of the accuracy analysis. The processing device of claim 6.

8. The at least one processor stores the trained judgment model before and after regeneration in a memory, performs additional learning on the trained judgment model before regeneration using teacher data different from the check result, and regenerates another trained judgment model of the same generation as the trained judgment model after regeneration. The processing device of claim 6.

9. The determination result includes an alternative obtained by inputting the data into a trained alternative proposal model. The processing device according to claim 2 .

10. When executed by at least one processor, receiving an input of a content including at least one data to be determined from a client terminal device, and acquiring the data from the content; determining whether or not the data obtained can be used in the content based on regulations under a predetermined law or guideline; Transmitting a result of the determination as to whether the data can be used and a request for the suitability of the determination result to a checker terminal device; transmitting to the requester terminal device a comprehensive result generated based on the judgment result and the check result received from the checker terminal device; A processing program that causes the at least one processor to function in such a manner.

11. A method of processing executed by at least one processor, comprising: receiving an input of content including at least one data to be determined from a client terminal device, and acquiring the data from the content; determining whether or not the acquired data can be used in the content based on restrictions imposed by a predetermined law or guideline for each of the acquired data; transmitting a result of the determination as to whether the data can be used and a request for the validity of the determination result to a checker terminal device; transmitting a comprehensive result generated based on the judgment result and the check result received from the checker terminal device to the requester terminal device; A processing method comprising:

Citation Information

Patent Citations

  • Utterance control device and computer program

    JP2011204178A

Cited By

  • Information processing device, information processing method, and program

    JP7830755B1