Processing apparatus, processing program, and processing method

Abstract

To provide a processing device, a program, and a method for outputting a result of determination as to whether or not to be regulated by a predetermined law or guideline in a more understandable manner.SOLUTION: The processing device includes at least one processor, and the at least one processor receives an input of content including one or more texts including at least one word to be determined, acquires the one or more texts from the received content, determines, for each of the acquired one or more texts, whether or not each text is usable in the content based on a regulation by a predetermined law or guideline, and outputs determination result information related to a result of the determination so that the determination result information is displayed together with the content.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] This disclosure is based on the assumption that the text contained in the content is subject to the regulations of the specified laws and regulations or guidelines. The present invention relates to a processing device, a processing program, and a method for determining whether or not to receive a test. [Background technology]

[0002] For example, the Act on Ensuring the Quality, Efficacy and Safety of Pharmaceuticals, Medical Devices, etc. There is a "law" that, depending on how it is advertised, could have a major impact on health and hygiene. Therefore, there is a lack of appropriateness regarding pharmaceuticals, quasi-drugs, cosmetics, medical devices, or regenerative medicine products. It was known that advertising that was prohibited by the law was subject to restrictions. Patent Document 1 describes a database of NG expressions in which NG sentences and NG words are registered in advance. It is equipped with a system that compares text-based speech data with the NG expression database. The paper describes a system for determining whether a test is NG or not and reporting the results. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-204178 Summary of the Invention [Problem to be solved by the invention]

[0004] Therefore, in light of the above-mentioned technology, the present disclosure is based on the following: Processing device and processing program for outputting the results of the determination of whether or not a product is subject to regulation in an easier-to-understand manner The present invention provides a method of processing. [Means for solving the problem]

[0005] According to one aspect of the present disclosure, "at least one processor, The processor includes one or more texts that include at least one word to be determined. and receiving an input of content to be included in the one or more and for each of the one or more texts obtained, The use of each text in the content is subject to regulations by law or guidelines. The content is displayed together with the content. and outputting the determination result information so that the determination result information is "Placement" is provided.

[0006] According to one aspect of the present disclosure, "executing by at least one processor Receives input of content containing text that includes at least one word to be judged. and receiving the received content, acquiring the text from the received content, and In response to the strike, the content will be subject to the regulations of the prescribed laws and regulations or guidelines. and determining whether or not the text can be used in the and outputting the determination result information so that it is displayed together with the content. A processing program that causes another processor to function is provided.

[0007] According to one aspect of the present disclosure, "a processing method executed by at least one processor" The input of the content contains text that includes at least one word to be judged. Accepting a force and obtaining the text from the accepted content. The acquired text is then processed based on the regulations of the prescribed laws and guidelines. determining whether or not the text can be used in the content; The determination result information related to the result is displayed together with the content. and outputting the result. [Effects of the Invention]

[0008] According to an embodiment of the present disclosure, whether or not the device is subject to regulation by a predetermined law or guideline A processing device, a processing program, or a processing method for outputting judgment results in a more understandable manner can be provided.

[0009] It should be noted that the above effects are merely illustrative for the sake of convenience and are not limiting. In addition to or instead of the above advantages, any of the advantages or effects described in this disclosure may be obtained. It is also possible to achieve effects that would be obvious to those skilled in the art. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a schematic diagram of a processing system 1 according to one embodiment of the present disclosure. [Figure 2] FIG. 2 is a block diagram showing a configuration of the terminal device 100 according to an embodiment of the present disclosure. [Figure 3] FIG. 3 is a block diagram showing a configuration of the server device 200 according to an embodiment of the present disclosure. [Figure 4A] FIG. 4A is a diagram conceptually illustrating a determination database stored in the server device 200 according to an embodiment of the present disclosure. [Figure 4B] FIG. 4B is a diagram conceptually illustrating a content table stored in the server device 200 according to an embodiment of the present disclosure. [Figure 5]FIG. 5 is a diagram showing a processing flow executed in the processing device 100 according to an embodiment of the present disclosure. [Figure 6] FIG. 6 is a diagram showing a processing flow executed in the server device 200 according to an embodiment of the present disclosure. [Figure 7] FIG. 7 is a diagram showing a processing flow executed in the server device 200 according to an embodiment of the present disclosure. [Figure 8] FIG. 8 is a diagram illustrating a processing flow for generating a trained model according to an embodiment of the present disclosure. [Figure 9] FIG. 9 is a diagram illustrating a processing flow for generating a trained model according to an embodiment of the present disclosure. [Figure 10] FIG. 10 is a diagram illustrating a processing flow for generating a trained model according to an embodiment of the present disclosure. [Figure 11] FIG. 11 is a diagram showing a processing flow executed in the server device 200 according to an embodiment of the present disclosure. [Figure 12] FIG. 12 is a diagram showing a processing flow executed in the terminal device 100 according to an embodiment of the present disclosure. [Figure 13] FIG. 13 is a diagram showing an example of a screen output to the terminal device 100 according to an embodiment of the present disclosure. [Figure 14] FIG. 14 is a diagram illustrating an example of content according to an embodiment of the present disclosure. [Figure 15] FIG. 15 is a diagram illustrating an example of a screen output to the terminal device 100 according to an embodiment of the present disclosure. [Figure 16] FIG. 16 is a diagram showing an example of a screen output to the terminal device 100 according to an embodiment of the present disclosure. [Figure 17] FIG. 17 is a diagram showing an example of a screen output to the terminal device 100 according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] An embodiment of the present disclosure will be described with reference to the accompanying drawings. The elements are given the same reference numerals.

[0012] 1. Overview of Processing System 1 The processing system 1 according to the present disclosure is mainly for checking the quality, efficacy, and safety of pharmaceuticals, medical devices, etc. The content of advertisements and other media includes wording and expressions that are regulated by laws, regulations, guidelines, etc. This is used to determine whether a particular item is being used in the content. When the target content is input and the processing system 1 receives the content, The processing system 1 then acquires text from the content. We will determine whether the content of the document contains any wording or expressions that are restricted by the above laws, regulations, or guidelines. The result is output.

[0013] In this disclosure, laws, regulations, or guidelines may not include the following: "Act on Ensuring Effectiveness and Safety," "Specified Commercial Transactions Act," "Unfair Premiums and Misleading Representations" Prevention Act, Health Promotion Act, Food Labeling Act, Food Labeling Standards, Medical Advertising Guidelines Examples include various laws and guidelines that regulate inappropriate labeling, such as "food labeling." These are not limited to those enacted by national or local governments, but also by specific industry groups. It may be established by an organization such as a body, union, company, association, or individual. It does not necessarily have to be something explicitly called a "law" or "guideline," but rather a standard, rule, or regulation. It can be any kind of document, regulation, etc. It can also be anything that is specifically stated in a document and made public. It does not have to be something that is

[0014] In addition, in this disclosure, items subject to regulation by laws, regulations or guidelines include medicines, Quasi-drugs, regenerative medicine products, medical devices, health foods, functional foods, cosmetics and mail order However, these are of course just examples, and It is not limited to only

[0015] In the present disclosure, the content typically includes the above-mentioned advertising content. However, it is not limited to this. For example, content can be other than advertisements, such as product information. Packaging, product introductions, packaging, company introductions, comments and messages on social media, news There are various examples, such as newspaper articles. Any content containing text containing the two words is acceptable. Web page format containing ML and image data, image data consisting of image data only Standardized data formats such as PDF that can be used without depending on a specific OS or hardware electronic document file formats, electronic files used for document editing, spreadsheets, and presentations Content is available in various formats, including file formats. The text does not need to be recognizable as text data. For example, the image data (that is, the subject) may include text.

[0016] In the following examples, advertising content provided in web page format is subject to laws and regulations. This section explains how to determine whether a content contains expressions restricted by the guidelines. Of course, the present invention is not limited to this example.

[0017] 2. Configuration of Processing System 1 FIG. 1 is a schematic diagram of a processing system 1 according to an embodiment of the present disclosure. The management system 1 includes a terminal device 100 held by a user and a network The system 1 includes a server device 200 connected to the network 300 so as to be able to communicate with the server device 200. In this example, the server device 200 and the terminal device 100 execute a program stored in memory. The server device 200 and the terminal device 100 execute the process of this embodiment. , and communicate with each other as needed to exchange various information (for example, FIG. 4A and FIG. 4B) and processes necessary for the progress of the process. Send and receive programs, etc.

[0018] In the example of FIG. 2, only one terminal device 100 is shown, but of course, two or more terminal devices may be used. It is also possible to include the terminal device 100. Also, the server device 200 can be used as a single However, the components and processes of the server device 200 are not necessarily shared among multiple server devices or cloud computing devices. It is also possible to distribute the data to the server device 200 and the terminal device 10. The processing according to this embodiment is executed by a system 1 including the server device 200. It is also possible to execute the process only on the terminal device 100 without using the present disclosure. In the present disclosure, the processing device is capable of executing the processing program and the processing method. For example, it may be either the terminal device 100 or the server device 200. Unless otherwise specified, the server device 200 functions as a processing device. However, the present invention is not limited to this example.

[0019] 3. Configuration of Terminal Device 100 FIG. 2 is a block diagram showing a configuration of the terminal device 100 according to an embodiment of the present disclosure. The terminal device 100 does not need to include all of the components shown in FIG. 2, and may have a configuration in which some of the components are omitted. It is possible to remove other components and add other components.

[0020] An example of the terminal device 100 is a laptop PC. In particular, mobile terminal devices capable of wireless communication, such as smartphones, feature phones, The processing program according to the present disclosure is used in a mobile information terminal, a PDA, a desktop computer, or the like. Any device capable of carrying out the method can be suitably applied.

[0021] 2, the terminal device 100 includes an output interface 111, a processor 112, Memory 113, including RAM, ROM, non-volatile memory (in some cases, HDD), etc. A communication interface 114 including a communication processing circuit and an antenna, a touch panel 116, and a hardware The input interface 115 includes a keyboard 117 and other components. are electrically connected to each other via control lines and data lines.

[0022] The processor 112 is composed of a CPU (microcomputer) and a memory. Based on various programs stored in the controller 113, the controller 113 controls other connected components. Specifically, the processor 112 executes the processing according to this embodiment. and a program for executing the OS are read from the memory 113 and executed. In this embodiment, the processor 112 receives the following through the input interface 115: and selects the content to be judged based on the instruction input received from the user. The processed and selected content is sent to the server device 200 via the communication interface 114. The server device 200 transmits the determination result information via the communication interface 114. The received determination result information is stored in the memory 113 in association with the content ID information. The stored judgment result information is output together with the content that was the subject of judgment. The processor 112 executes processing such as outputting the data via the interface 111. It may be configured with one CPU, or multiple CPUs. Other types of processors, such as a GPU with a built-in processor, may also be combined as appropriate.

[0023] The memory 113 is composed of a ROM, a RAM, a nonvolatile memory, a HDD, etc., and serves as a storage unit. The ROM functions as a memory for executing the game application and the OS according to this embodiment. The RAM stores instructions and commands for the program. To write and read data while it is being processed by the processor 112 Non-volatile memory is memory that is used for writing data by executing the program. A memory in which data is written and read, and data written to this memory is read from and written to the processor. In this embodiment, the memory 113 stores: Determination based on instruction input received from the user via the input interface 115 A process of selecting target content, and transmitting the selected content to a communication interface 1. 14 to the server device 200, and a process of receiving the determination result information via the device 114, and transmitting the received determination result information to the content I D information and storing it in the memory 113; The processor 111 outputs the content together with the output interface 111. Although not specifically shown in the figure, the memory 113 may be an input interface. The interface 115 may be connected to a removable storage medium, a database, etc. stomach.

[0024] The communication interface 114 is a remotely installed It also functions as a communication unit that transmits and receives information to and from the server device 200 and other terminal devices. The communication processing circuit stores a program for executing the processing according to this embodiment and The various information to be used in the process is received from the server device 200 as the process progresses. In addition, various information required for executing the processing is transmitted to the server device 200. In this embodiment, the processing is performed in particular for the communication between the server device 200 and the It transmits and receives content and judgment result information.

[0025] The communication processing circuit processes data based on wideband wireless communication methods such as the LTE method. However, wireless LANs such as IEEE802.11 and Bluetooth ( It is also possible to process based on a method relating to narrowband wireless communication such as Also, wired communication may be used instead of or in addition to wireless communication.

[0026] The input interface 115 is a touch panel 116 and / or a hard key 117. The touch panel 116 is configured to receive operation inputs from the user. The display is disposed so as to cover the surface 111 of the display. The information on the position coordinates corresponding to the image data displayed is output to the processor 112. The sensor type is a known type such as a resistive film type, a capacitive coupling type, or an ultrasonic surface acoustic wave type. In this embodiment, as an example, the touch panel 116 is a tap to select or move an object displayed on the display using a pointer. The touch panel 116 is one of the input interfaces. It is only an example, and of course other configurations can be used. Wireless or wired input interface (such as a keyboard or mouse) It is also possible to use an input interface.

[0027] The output interface 111 receives instructions from the processor 112 and stores the data in the memory 113. The image information stored in the memory is read out and various tables generated by executing the processing according to this embodiment are displayed. It functions as an output unit that outputs the display (for example, Figs. 13 and 15 to 17). An example of an interface 111 is a display that consists of a liquid crystal display or an organic EL display. However, the terminal device 100 itself does not necessarily have a display. For example, the output interface 111 may be connected by wire or wirelessly. It may also be a device that outputs various information for display on a connected display. The communication interface 114 may also function as the output interface 111. do.

[0028] When the terminal device 100 functions as a processing device, the processor 112 The input of content containing one or more texts containing at least one word of interest If the accepted content contains text, Extracting text from image data, extracting one or more texts from accepted content A process of acquiring text, and applying a predetermined law to each of the acquired text or texts. Or, based on the restrictions of the guidelines, whether or not each text can be used in the content The judgment process and judgment result information related to the judgment result are now displayed along with the content. Executes processing such as outputting determination result information.

[0029] Similarly, when the terminal device 100 functions as a processing device, the memory 113 Content containing one or more pieces of text that contain at least one word to be judged If the received content contains text, A process of extracting text from the image data, one or more from the accepted content A process of acquiring text, and a predetermined process for each of the acquired one or more texts. The use of each text in the content is subject to restrictions by law or guidelines. The process of determining whether or not the content is acceptable, and the judgment result information related to the judgment result are displayed together with the content. The storage device stores a program for executing a process for outputting the determination result information as described above.

[0030] 4. Configuration of Server Device 200 FIG. 3 is a block diagram showing a 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. It is possible to use an abbreviated configuration, or to add other components.

[0031] 3, the server device 200 includes RAM, ROM, nonvolatile memory, HDD, etc. a memory 211 including a processor 212 including a CPU and a communication interface These components are connected to each other via control lines and data lines. are electrically connected to each other.

[0032] The memory 211 includes RAM, ROM, non-volatile memory, and HDD, and functions as a storage unit. The memory 211 stores instructions and commands for executing the processes and OS according to this embodiment. Such programs are loaded by the processor 212. The memory 211 (particularly RAM) stores the program. It is used temporarily to write and read data while the The memory 211 also stores the determination database shown in FIG. 4A and the control data shown in FIG. In addition to the content table, the drawing information used to output the screens shown in FIGS. 13 and 15 to 17 is also included. Furthermore, the memory 211 stores information such as at least one word to be judged. A process of accepting input of content containing one or more texts, If the content contains text, the text is extracted from the image data. processing, retrieving one or more texts from the received content, Or, for each of multiple texts, based on the regulations of the prescribed laws or guidelines The process of determining whether each text can be used in the content, and the results of the determination Executes processing such as outputting the judgment result information so that it can be displayed together with the content The memory 212 also stores a program for performing various determination processes. The trained text decision model, trained reason decision model, and trained alternative decision model are Although not shown in the figure, the determination database stored in the memory 211 The data may be stored in a database device or the like installed remotely via a network. good.

[0033] The processor 212 is composed of a CPU (microcomputer) and a memory. The controller 211 controls other connected components based on various programs stored in the controller 211. In this embodiment, the processor 212 functions as a control unit for determining whether or not a Input of content containing one or more pieces of text that contain at least one target word If the accepted content contains text, Extracting text from image data, extracting one or more texts from the received content A process of obtaining a text, and then processing each of the obtained texts in accordance with a predetermined law or regulation. determines whether or not each text can be used within the content based on the restrictions set out in the guidelines. The process to be performed and the judgment result information related to the judgment result are displayed together with the content. The processor 212 is configured as a single CPU. Alternatively, it may be configured with multiple CPUs.

[0034] The communication interface 213 is, for example, a communication interface between each terminal device 100 and the network 300. or via a network 300 with other server devices or database devices. In order to transmit and receive programs for executing the processes according to the embodiment, various information, etc., modulation and demodulation are performed. The communication interface 213 is a communication interface that can be used with the above wireless communication method or a known wired communication method. In this embodiment, the terminal device communicates with other server devices according to the communication method. In particular, the terminal device 100 transmits and receives the content to be judged and the judgment result information. .

[0035] 5. Information stored in memory FIG. 4A is a determination database stored in the server device 200 according to an embodiment of the present disclosure. The information stored in the determination database is based on various laws and regulations. It is updated as needed to reflect the current line.

[0036] As shown in Figure 4A, the decision database contains category information, target information, propriety information, reason information, and The category information and the alternative information are stored in association with each other. The content of the regulations differs depending on the product subject to the regulations. This product category is an example of category information, and is Any information is acceptable as long as it is used to classify the subject of the control. For example, if the word "fine wrinkles" is included in the content, If the product is in a category, it can be judged. This information indicates the judgment result when the text shown in the specified target is used. The word "Caution" refers to expressions that are subject to regulation by the relevant laws and guidelines. Therefore, it may be subject to regulation, or it may be an expression for which a clear judgment cannot be made at this time. "Appropriate" does not mean expressions that comply with laws and guidelines or are subject to regulation by these. "Reason information" indicates the reason why the content was judged to be inappropriate or requiring caution. "Alternative information" is information that indicates the reason for a decision that has been made that is inappropriate or requires caution. , and is information indicating alternatives to avoid being judged in such a manner.

[0037] In addition, three types of information, inappropriate, caution required, and appropriate, are stored as the appropriateness information. Of course, the information is not limited to these. For example, "inappropriate and appropriate" and "inappropriate and caution required" It is acceptable to have only two pieces of information, "Appropriate" or "Appropriate and Caution Required", or only one of them. It may be possible to divide it into more detailed information and make it into a large number of pieces of information, or it may be possible to determine whether the information is inappropriate or appropriate. It may be a numerical value or level information showing the rate. The wording may also be inappropriate, caution required, appropriate, etc. Instead of expressions such as "NG", "gray", "OK", "〇 (circle)", "× (cross)", "△ (triangle)", etc. It can be anything, such as a letter, symbol, or number.

[0038] In addition, this judgment database includes a text judgment process, a judgment process for the judgment reason, and an alternative In addition, the trained decision model can be used for each decision. When using the data, it must be used as training data for each trained decision model. is also possible.

[0039] FIG. 4B shows a content stored in the server device 200 according to an embodiment of the present disclosure. 1 is a diagram conceptually illustrating a content table. It is generated and updated as needed.

[0040] As shown in FIG. 4B, the content table includes content ID information, category information, and text. Store information, location information, character count information, propriety information, reason information, alternative information, etc. are all associated with each other. The "content ID information" is used to identify each piece of content to be judged. The content ID information is unique to each content. New content is transmitted from 100 to the server device 200, and the memory of the server device 200 The "category information" is newly generated each time it is stored in 211. This information identifies the category of product for which the content is advertised. The "text" information identifies the text contained in and retrieved from each piece of content. The information may be the text data itself, or It may also be information for identifying the location where the text data is stored or the text data. In this case, the content may contain one or more texts depending on the content. A plurality of pieces of text may be stored in association with one piece of content ID information. "Location information" is information for identifying the position where each text is placed in the content. The information includes the coordinates where each piece of text is placed in the content, the edge of the content, The rough position of each text, such as the number of pixels from the image, the number of rows and columns, the number of characters, etc. Any information can be used as long as it can be determined. This information indicates the number of characters that make up the text. Each of the information explained in FIG. 4A is included in each text. This information is stored in association with the text. Each time the determination process, the determination reason process, and the alternative plan determination process are performed, The text is stored in association with the entered text.

[0041] 6. Processing flow performed by the terminal device 100 (content output processing) FIG. 5 shows a processing flow executed in the terminal device 100 according to an embodiment of the present disclosure. Specifically, FIG. 5 shows the process performed when the advertisement determination tool is started in the terminal device 100. 1 shows a process flow performed until the content to be set is transmitted to the server device 200. The processing flow is mainly executed by the processor 112 of the terminal device 100. This is done by reading and executing the program.

[0042] Referring to FIG. 5, the processor 112 receives an advertisement decision tool via the input interface 115. The advertisement determination tool is started upon receiving a user's access to the advertisement determination tool (S111 ) Such an advertisement determination tool is typically installed in advance in the terminal device 100. Examples include tools provided by installed application software. Other examples of such tools include, but are not limited to, web browsers. Examples of such tools include those provided via browser application software.

[0043] Here, FIG. 13 shows an example of a screen output to the terminal device 100 according to an embodiment of the present disclosure. Specifically, FIG. 13 shows the process in which the advertisement determination tool is started in S111 of FIG. After that, the selection of the judgment object is output to the display via the output interface 111. 13 is a diagram showing an example of a screen. According to FIG. 13, the judgment target selection screen displays the components to be judged. The format selection area 11 is used to select the format of the content, and the storage location of the content to be judged is and a storage location input area 12 for indicating the storage location and a judgment area for transmitting a judgment start instruction to the server device 200. A start button 13 is displayed. The user inputs the judgment object via the input interface 115. Select the type of content to be judged from the selection area 12. Once the selection is made, The corresponding check box will be blacked out. Then, the user can enter the The user selects a storage location input area corresponding to the content to be judged via the interface 115. Then, when the user selects the judgment start button 13, The content stored in the storage location entered in the storage location input area will be displayed along with a command to start judgment. The data is read and sent to the server device 200 .

[0044] 14 is a diagram showing an example of content according to an embodiment of the present disclosure. Figure 14 shows an example of a web page containing HTML and image data. 14 shows an example of advertising content for cosmetics provided in the form of a A web page 31 is shown, which is advertising content for cosmetics. is a hypertext written in markup language that says "You can improve your skin" Text 32a: "Maybe??" and text 32b: "It will make fine lines caused by dryness less noticeable." The web page 31 also includes a human-shaped object as image data. Object 33a, "You can get beautiful skin like a baby!" Image object 33b, and The image object 33c includes the text "Say goodbye to fine lines." Although it includes text and image data, the explanation of these will be omitted.

[0045] In this embodiment, the content shown in FIG. 14 is stored in a predetermined storage area of ​​the memory 113. The data is stored in a location and used in the processing described below.

[0046] Returning to FIG. 5 again, the processor 112 receives a format selection signal via the input interface 115. The user's input to area 11 (Fig. 13) is displayed. In the example of FIG. 14, the advertisement content in the form of a web page is selected (S112). Since the content is a web page, "web page" is selected as the content format. The processor 112 enables the storage location input area 12 (FIG. 13) corresponding to the selected format. The storage location input area 12 (see FIG. 1) is enabled via the input interface 115. 13), the processor 112 receives the user's input and determines the content to be judged. Next, the user selects a storage location for the content (S113). Upon receiving an input from the user to the judgment start button 13 (FIG. 13), the processor 112 The memory 113 is referred to, and the contents stored in the storage location input in the storage location input area 12 are The processor 112 then reads out the content (S114). The result is sent to the server device via the interface 114 together with an instruction to start the determination (S115). That is, in the example of FIG. 14, a web page format consisting of HTML and image data is used. The content is sent to the server device 200 as a target for determination. End the row.

[0047] In Figure 13, the category of the product advertised by the content to be judged is The selection of the library is not specifically illustrated, but the steps S111 to S115 are During the process, the processor 112 receives input from the user via the input interface 115. By accepting the instruction input, the card advertised by the content to be judged is A category is selected and transmitted to the server device 200 together with the content in S115.

[0048] 7. Processing flow (determination processing) performed by the server device 200 FIG. 6 shows a processing flow executed in the server device 200 according to an embodiment of the present disclosure. Specifically, FIG. 6 shows a diagram illustrating a content to be determined received from the terminal device 100. This shows the process flow from receiving the data to outputting the judgment result. The processor 212 of the server device 200 reads the program stored in the memory 211. This is done by executing the

[0049] 6, the processor 212 communicates with the terminal device 1 via the communication interface 213. 00, and receive the judgment start command, the content and category to be judged. Then, the processor 112 receives the input of the content (S211). When you acquire content, the content ID information to identify the acquired content is and associates the content ID information with the content and category information. The acquired content is stored in the memory 211. In addition, the acquired content is converted into image data, and the converted data is used as the content to be judged. and stored.

[0050] Next, the processor 212 reads the acquired content from the memory 211 and It is determined whether there are any parts that have not yet undergone various determination processes such as the determination process of the stock (S2 12). If there is a part for which the determination process has not been performed, the processor 212 Determine whether text data is included in the part where content is not judged (S 213). If the result of the determination is that text data is included, the processor 212 The text data is acquired and stored in the memory 211 without performing text extraction processing. The content ID information is stored in the content table in association with the content ID information (S215). If the text data is not included, the processor 212 uses optical character recognition technology such as OCR. Using this technique, text data is extracted from the content to be judged (S214). The processor 212 then retrieves the extracted text data and stores it in the memory 211. The content ID information is stored in the content table in association with the content ID information (S215).

[0051] Specifically, the processor 212 reads the contents of, for example, FIG. 14 and If the judgment is not performed on the part of the image object 33c that says "Goodbye to the slow motion" The processor 212 determines that the image object 33c is image data. The process proceeds to 214, where text data is extracted from the image object. The image object 33c is then displayed as a text message saying "Say goodbye to fine lines." The data is extracted and stored in memory 211 in association with the content ID information.

[0052] Here, each acquired text is composed of one or more words. The processor 212 reads the stored text from the memory 211 and The text is decomposed and analyzed into individual words (S216). As an example, dictionary data in which various words are stored in advance is stored in the memory 211 or the database. The data is stored in the database and a check is made to see if the read text contains any words stored in the dictionary data. Another example of this process is to check the read text. This is done by morphological analysis of the text using a morphological analysis program. Morphological analysis is the process of dividing text into the smallest units that cannot be further broken down to make sense. This is a method of breaking down a word into morphemes using dictionary data. For example, the processor 212 When the text "Say goodbye to fine wrinkles" in Figure 14 is read, the morphological analysis "Say goodbye to fine wrinkles" (" / " indicates the division of each word after it has been broken down into morphemes) ) and the text is broken down into "fine lines", "to", "goodbye", "fine lines", "to It is analyzed that the song is composed of "Yonara" and "Sayonara to Fine Wrinkles". 2 is the text "Say goodbye to fine lines" as text information in the content table. The analysis results are also stored. Not only the smallest units of a sentence, such as "fine lines," "to," and "goodbye," but also the words before and after Sentences with elements added: "Say goodbye to fine lines," "Say goodbye to fine lines," and "Say goodbye to fine lines" Therefore, the decision described below is not made simply based on the smallest unit, the word. Instead, it becomes possible to make a judgment based on the context before and after.

[0053] At this time, the processor 212 counts the number of characters that make up each text, The number of characters is stored in the content table of the memory 211 in association with the text.

[0054] The processor 212 then converts the parsed text into one or more words into a statute or guideline. It is determined whether the vehicle is subject to regulation by the roadmap (S217). If the text is neither "inappropriate" nor "warning" as a result of the assessment, The processor 212 stores the result in the contents of the memory 211 in association with the determined text. The text is stored in the table, and the process returns to step S212 (S218). If the result is determined to be "inappropriate" or "requires caution," the processor 212 is stored in the content table of the memory 211 in association with the text to be judged, The process proceeds to the determination of the reason for the determination (S218). 12 determines the reason for the above judgment and provides the reason for the judgment in the text that was the subject of the judgment. The processor 212 then stores the information in the memory 211 in association with the To avoid receiving judgments about text that has been judged as appropriate or requiring caution. The alternatives for the text are identified and associated with the identified alternatives. The data is stored in the memory 211 (S220). Details of the processes in S217 to S220 will be explained later. More details will be given later.

[0055] The processor 212 performs the processes of S212 to S220 as part of the text determination process, etc. Execute the various judgment processes until there are no more parts that have not yet been processed. Then, the processor 212 reads the content as image data stored in S211, Determination result information including propriety information, reason information, and alternative information stored in association with the text is output to the terminal device 100 via the communication interface 213 (S221). Then, the processing flow is terminated.

[0056] In this way, in the determination processes of S217, S219, and S220 in FIG. The text to be processed is the smallest unit of text, such as "fine wrinkles," "to," and "goodbye." Not only the word itself, but also the words "koji ni," "ni sayonara," and "koji" with the preceding and following elements added. This includes the sentence "Goodbye to the crocodile." Therefore, it is not possible to simply base it on the smallest unit, the word. These judgment processes are not performed based on the context, but rather on the context before and after. It is possible.

[0057] As shown in FIG. 14, the processor 212 determines whether the content to be determined (image data) For each point (data), position coordinates are set on a two-dimensional coordinate system consisting of the x-axis and y-axis. That is, the processor 212 determines whether each text acquired in S215 is a target of the determination. The position information of the text is acquired and stored in the memory 2 in association with each text. The processor 212 then stores the information in the table of contents 11 based on the information. The object information for identifying whether or not the object can be used is generated, and the object information is used to identify whether or not the object can be used. The location information of the content that was judged is also included. When outputting the determination result information in S221, the processor 212 The object information indicating whether or not the object can be used and its location information are also included in the determination result. The information is output to the terminal device 100.

[0058] Also, as mentioned above, when decomposing and analyzing the text in S216, The processor 212 counts the number of characters in each text and associates the result with the text. The processor 212 stores the table of contents in the memory 211. In the case of outputting the judgment result information in 11, the computer associated with the judgment result information is The character count information in the content table is referenced based on the content ID information, and the content ID information is The total number of characters associated with the information is output as character count information and included in the judgment result information. Note that this total is an example of character count information, and the average, maximum, minimum, etc. Any information related to the above is acceptable.

[0059] 7. Processing flow performed by the server device 200 (details of determination processing using a trained model) ) FIG. 7 shows a processing flow executed in the server device 200 according to an embodiment of the present disclosure. Specifically, FIG. 7 shows the process performed in steps S217 to S220 in FIG. The processing flow is mainly performed by the processor 212 of the server device 200. This is done by reading and executing the program stored in 1.

[0060] According to FIG. 7, the processor 212 performs word-by-word decomposition and analysis in S216 of FIG. The text (including the text itself and each element after analysis and decomposition) and the text The category associated with the item is read from the memory 211 (S311). The memory 212 inputs the text and the trained text determination model stored in the memory 211. The category is input (S312). The processor 212 uses the trained text determination model The text entered is restricted in use by law or guidelines. The processor 212 outputs the information indicating whether the determination result is correct or not. As the result information, the content table of the memory 211 is stored in association with the text to be judged. Store it in Bull (S313).

[0061] Next, the processor 212 refers to the table of contents and It is determined whether "inappropriate" or "requires caution" is stored in the information on whether the request is acceptable or not (S314 ) If the result is neither "inappropriate" nor "requires caution," Processor 2 12 executes the process of S112 shown in FIG. 6. On the other hand, if the In this case, the processor 212 stores the text and category read in S311 in the memory. The reason determination model stored in the processor 211 is input (S315). The processor uses the learned reason determination model to output reason information indicating the reason for the determination. 212 outputs the output reason information as judgment result information to the text to be judged. The contents are stored in the corresponding content table in the memory 211 (S316).

[0062] Next, the processor 212 stores the text and category read in S311 in the memory 2. The result is input to the trained alternative decision model stored in the processor 212 (S317). uses a trained alternative decision model to avoid being judged as inappropriate or requiring caution. The processor 212 outputs the alternative proposal information as the determination result information. The result is stored in the content table of the memory 211 in association with the text to be judged. (S318).

[0063] Here, FIG. 8 shows a processing flow for generating a trained model according to an embodiment of the present disclosure. Specifically, FIG. 8 shows the generation of the trained text determination model in S312 of FIG. The processing flow is a diagram showing a process flow relating to the creation of the processor 100. 11 or may be executed by a processor in another processing device. .

[0064] According to FIG. 8, the processor performs the process shown in steps S213 to S216 of FIG. The study text and the content containing the study text are selected from the content prepared for study. The step of acquiring the learning category of the content to be included is executed (S411). The text has undergone the same analysis process as in S216 in Figure 6. In the case of the text "Say goodbye to wrinkles," not only "Say goodbye to wrinkles," but also "Fine wrinkles," The elements of "ni", "sayonara", "fine lines" and "ni sayonara" are also used as learning texts. The processor also uses the learning text and the learning Based on the classification category, the classification database shown in FIG. 4A is referenced to determine the classification of the training text. The processor then executes a step of determining whether the device is available for use (S412). The result of the evaluation is stored as a correct answer label in association with the learning text. The backup is executed (S413).

[0065] Once the learning text, learning category, and corresponding information are obtained, The processor then performs a step of performing machine learning of a text determination pattern using the data. The machine learning is performed (S414). For example, the machine learning is performed by combining neurons. These sets of information are given to the neural network, and the output from the neural network is The parameters of each neuron are adjusted so that the correct label is the correct / incorrect information. The processor then repeats the learning process to generate the learned text decision. The step of acquiring the model and storing it in memory is executed (S415). The text determination model is stored in the memory 211 of the server device 200 or in a memory connected to the server device 200 via a wired or It may also be stored in another device connected via a wireless network.

[0066] FIG. 9 shows a processing flow for generating a trained model according to an embodiment of the present disclosure. Specifically, FIG. 9 is a diagram showing the process for generating a trained reason determination model in S315 of FIG. The processing flow is a diagram showing a processing flow performed by the processor 111 of the processing device 100. The processing may be performed by a processor in another processing device.

[0067] According to FIG. 9, the processor may generate a learning image by using a method similar to that of S411 in FIG. Study texts from the prepared content and content that includes the study texts In addition, the processor executes a step of acquiring a training category (S421). The result of the determination as to whether or not the learning text can be used is obtained by the same method as in S412 of FIG. Then, the processor executes a step of acquiring information (S421). Based on the acquired learning text, learning category, and valid / invalid information, the judgment shown in FIG. 4A is performed. The reason why the use of the learning textbook was judged to be inappropriate or requiring caution, based on the standard database. Then, the processor executes a step of determining whether the and storing the reason information as a correct answer label in association with the learning text. (S423).

[0068] Study text, study category and approval / disapproval information, and reason information associated with them Once the information is obtained, the processor uses it to perform machine learning of the judgment patterns of the judgment reasons. The machine learning is performed by combining neurons (S424). These sets of information are given to the neural network that has been combined, and the neural network The parameters of each neuron are adjusted so that the output from the network is the same as the reason why the label is correct. The learning is done by repeatedly adjusting the parameters. The reason determination model is acquired and stored in memory (S425). The learned reason determination model is stored in the memory 211 of the server device 200 and is effectively connected to the server device 200. It may also be stored in another device connected via a wired or wireless network.

[0069] FIG. 10 shows a processing flow for generating a trained model according to an embodiment of the present disclosure. Specifically, FIG. 10 shows the process of generating the trained alternatives determination model in S317 of FIG. The processing flow is a diagram showing a process flow relating to the creation of the processor 100. 11 or may be executed by a processor in another processing device. .

[0070] 10, the processor generates training data by, for example, a method similar to that of S411 in FIG. Select the study text and the content containing the study text from the content provided on In addition, the processor executes a step of acquiring a learning category of the object (S431). , the result of the determination of whether or not the learning text can be used is determined by the same method as in S412 of FIG. The processor executes a step of acquiring information (S431). Based on the acquired learning text, learning category, and validity information, the following is performed: The use of the learning text is not judged as inappropriate or requiring caution by referring to the judgment database. Next, the processor executes a step of determining an alternative for making the process easier (S432). The alternative information indicating the determined alternative is associated with the training text as a correct answer label. A storing step is executed (S433).

[0071] Study texts, study categories, availability information, and associated alternatives Once the information is available, the processor uses it to machine-learn decision patterns for alternatives. The machine learning is performed by combining neurons (S434). These sets of information are given to the neural network that has been combined, and the neural network The parameters of each neuron are adjusted so that the output from the network is the same as the correct label. The learning process is carried out by repeatedly adjusting the data. A step of acquiring and storing the alternative decision model in memory is executed (S435). The trained alternative decision model is stored in the memory 211 of the server device 200 or the 0 may be stored in another device connected to the device via a wired or wireless network.

[0072] 8 to 10, the correct labels are calculated by processing the decision database shown in FIG. 4A with the processor. However, this is not limited to this, and any user may use the learning text We judge whether text is acceptable or not in accordance with laws or guidelines corresponding to the educational category, The reasons for this may be determined and alternatives may be determined.

[0073] 8. Processing flow performed by the server device 200 (details of the determination process using the determination database) fine) FIG. 11 is a process flow executed in the server device 200 according to an embodiment of the present disclosure. Specifically, the process performed in steps S217 to S220 in FIG. In the example of Figure 7, a judgment was made using a trained model, but in Figure 11, a trained model was used. This shows the case where a judgment is made using a judgment database instead of a model. The processor 212 of the server device 200 reads the program stored in the memory 211. This is done by executing the

[0074] 11, the processor 212 performs word-by-word decomposition and analysis in S216 of FIG. The analyzed text and the category associated with the text are read from the memory 211. (S311). The processor 212 refers to the determination database in the memory 211 and reads The entire text provided, each word and each combination of words that make up the text, The text is checked to see if it matches the text registered in the target information of the fixed database. If there is a matched text, a word, or a combination of words, the result is judged (S512). If so, the processor 212 determines that the target information in the determination database matches. The judgment result information is used to determine whether the text is correct or not. The content is stored in the content table in association with the object (S513).

[0075] Specifically, for example, after analysis, in the category of "cosmetics," The text "Wa / ni / Sayonara" (" / " separates each word after it has been broken down into morphemes) Next, the processor 212 reads out the category of the judgment database. The target information in which "cosmetics" is registered is extracted by referring to the target information. The text that makes up the sentences "Fine lines", "Goodbye", "Goodbye to fine lines", "Goodbye to " and "Say goodbye to fine wrinkles" match the extracted target text. In this case, since "fine wrinkles" is registered in the target information, the processor The step S212 refers to the information on whether or not the fine lines are present in the determination database. Then, the processor 212 associates the phrase "fine wrinkles / say goodbye" with the phrase to be judged. , and stores "inappropriate" in the propriety information of the content table.

[0076] Next, the processor 212 refers to the table of contents and It is determined whether "inappropriate" or "requires caution" is stored in the information on whether the request is acceptable or not (S514 ) If the result is neither "inappropriate" nor "requires caution," Processor 2 12 executes the process of S112 shown in FIG. 6. On the other hand, if the In this case, the processor 212 refers to the determination database and determines whether the target information is a The associated "reason information" is extracted (S515). The judgment extracted in the "reason information" of the content table in correspondence with the target text The reason is stored (S516). Similarly, the processor 212 stores the determination database. Then, the "alternative information" associated with the target information is extracted (S517). The processor 212 determines whether the text to be determined is a "substitute" in the table of contents. The extracted alternative plan is stored in "alternative plan information" (S516).

[0077] Specifically, for example, in the text to be judged, "Fine wrinkles / Sayonara / " If the determination is "inappropriate," the processor 212 sets "inappropriate" in the determination database. The processor then refers to the reason information associated with the "fine wrinkles" that have been determined to be "off." 212 is the reason why the judged "Fine Wrinkles / Sayonara" is "inappropriate" The reason for the judgment, "The phrase 'fine wrinkles' is inappropriate," is extracted, and the target of judgment, "fine wrinkles / / Goodbye, and the reason information in the content table is inappropriate. The processor 212 stores the reason for the determination, "OFF." Then, the processor 212 refers to the alternative information associated with "fine wrinkles." As the alternative information for the target item "Fine Wrinkles / Goodbye / ", the alternative "Delete" was extracted. and associate it with the target "Fine Wrinkles / Goodbye" and create a table instead of the contents table. "Delete" is stored in the alternative information.

[0078] 9. Processing flow performed by the terminal device 100 (output processing of determination result information) FIG. 12 shows a processing flow executed in the terminal device 100 according to an embodiment of the present disclosure. Specifically, FIG. 12 shows the result of the determination from the server device 200 in S221 of FIG. After the determination result information is output to the terminal device 100, the processing flow to be performed in the terminal device 100 is The processing flow mainly involves the processor 112 of the terminal device 100 processing data in the memory 113. This is done by reading and executing the program stored in the memory.

[0079] First, the processor 112 receives a message from the server device 200 via the communication interface 114. When the determination result information is received, the received determination result information is stored in the memory 113. The processor 112 reads out the stored judgment result information and outputs it to the output interface 11. The judgment result information is output to a display or the like via the display 1. , , ,the ,possibility ,reason ,information ,alternative ,information ,usage ,associated ,with ,each ,text ,that ,was ,determined. Possibility identification object information indicating whether or not the object is possible, and an arrangement indicating the arrangement position of the possibility identification object information. It includes position information and character count information.

[0080] Here, FIG. 15 shows an example of a screen output to the terminal device 100 according to an embodiment of the present disclosure. Specifically, FIG. 15 shows the determination result information received from the server device 200. The judgment result display screen when output to the display via the output interface 111 As shown in Figure 15, the judgment result display screen displays the content that has been judged. The content display area 14 includes a content display area 14 for determining whether or not the content is to be displayed. The content (for example, FIG. 14) is displayed, and the content is superimposed on the content. As shown in FIG. 1, the possible / probable identification object 15, the possible / probable identification object 16, and the possible / probable identification object 17 are Specifically, each text included in the content that was the subject of the judgment is displayed. Based on the placement position information set at the position corresponding to the position information of each object, The following items 15 to 17 are displayed so as to be superimposed on the content. The text "Maybe it can improve your skin?" in the test contains information on whether it can improve your skin. "Caution" is written in the text "You can have beautiful skin like a baby!" and "Say goodbye to fine wrinkles!" The text "Nara" has "inappropriate" stored as its appropriateness information. As shown in 5, the corresponding text is displayed at the position where the object is needed to identify whether it is needed or not. The text is displayed differently depending on whether it is intended or not.

[0081] Moreover, a judgment result list display area 21 is displayed on the right side of the content display area 14. In this judgment result list display area, there are The text boxes 21a to 21f are displayed. In Fig. 1f, reason information and alternative information associated with each text are listed. In example 15, these are displayed in a reduced size, and only the information indicating the text that was judged is displayed. In addition, each of the text boxes 21a to 21f is associated with a text. Based on the availability information, the result of the determination of availability is displayed in a recognizable manner. For example, in the example in Figure 15, the text "You can have beautiful skin like a baby!" and "Fine wrinkles" are displayed. The text "Goodbye to" has "inappropriate" stored as the appropriateness information, but The text boxes 21a and 21b are displayed surrounded by a thick line. The text "Maybe it can improve your skin?" has the message "Caution" as the validity information. However, this text box 21c is displayed surrounded by a dashed line.

[0082] In addition, at the top of the content display area 14, there are a character count display area 22 and an inappropriate part display area 23 and a cautionary part display area 24 are displayed. The character count display area 22 displays the judgment result information. The inappropriate part display area 23 displays the information on the number of characters received as the inappropriate part. The number of texts that are judged to be inappropriate in the content that has been submitted is added based on the appropriateness information. The critical part display area 24 displays the critical part of the content that has been judged. The number of texts that are judged to require caution is added up based on the information on whether or not the text is relevant, and is displayed. In this embodiment, the number of characters, the number of texts determined to be inappropriate, and the number of texts determined to be inappropriate are stored. The number of texts displayed is now the cumulative value of each. The information may be other than a cumulative value as long as it is related to each numerical value. For example, the number of characters may be a cumulative value. In addition to product values ​​(total values), information related to the number of characters, such as average values, maximum values, and minimum values, can also be used. Either is fine.

[0083] In addition, at the top of the judgment result list display area 21, a download icon 19 and a print icon are displayed. The icon 20 is displayed. The pointer 18 is operated via the input interface 115. When one of the icons is selected, the corresponding process is executed. When an operation input is made to the download icon 19, each text that is the subject of judgment and The information on whether or not each text is acceptable, the reason for its availability, and the alternatives are listed. Download in standardized electronic formats such as spreadsheets and PDFs. In addition, when an operation input is made to the print icon 20, the determination result You can output the display screen to paper, or convert it to standardized electronic file formats such as PDF or image data. Although not shown in the figure, it is possible to download it as a shared file. An icon is displayed, and when an operation input is made to the icon, the icon is assigned to another desired user. You can share the results in any format you like. It is also possible to display other function icons in addition to the icon 9 and the print icon 20.

[0084] Returning to FIG. 12 again, after the judgment result information is output to the judgment result display screen, the processor 11 2 is input via the input interface 115 to the judgment result list display area 21. Determine whether an instruction to expand has been received for any of the text boxes Specifically, the processor 112 receives the following from the input interface 115 (S611). , an interrupt indicating that a click operation was performed at a position corresponding to one of the text boxes. If it is determined that the command input has been accepted, The processor 212 then selects the item in the determination result list display area for which an instruction to expand has been given. The text box is judged so that the information on whether it is OK, the reason, and the alternatives are displayed. The result list display area is updated and displayed (S612).

[0085] Here, FIG. 16 shows an example of a screen output to the terminal device 100 according to an embodiment of the present disclosure. Specifically, FIG. 16 shows the determination result list display area displayed in step S612 of FIG. 16 shows the screen after the area 21 is updated. The pointer 18 is moved via the interface 115 to the desired text box (here, the text box As a result, the text box 21a is As a result, for example, in Figure 15, the image is displayed in a reduced size. The text box 21a is expanded, and the reason information and alternatives associated with the text are displayed. The alternative information is displayed in a visually recognizable manner. As a result, the other text boxes 21b to 21f are expanded to the judgment result list display area. It is displayed by sliding downward within area 21.

[0086] Returning to FIG. 12 again, the processor 112 receives the computer program via the input interface 115. An instruction to select any text in the content display area was accepted. Specifically, the processor 112 determines whether the input interface If the pointer is detected at a position corresponding to any of the texts for a certain period of time from the sensor 115, It is then determined whether an interrupt signal indicating that an instruction input has been received has been received. If it is determined that the determination result list has been selected, the processor 212 displays the selected The text box corresponding to the text is positioned at the top of the judgment result list display area. The judgment result list display area is configured so that the information on whether the decision is accepted or not, the reason, and the alternative plan information are displayed. The area is updated and displayed (S614).

[0087] Here, FIG. 17 shows an example of a screen output to the terminal device 100 according to an embodiment of the present disclosure. Specifically, FIG. 17 shows the determination result list display area displayed in step S614 of FIG. 17 shows the screen after the area 21 is updated. Point the pointer 18 through the interface 115 to the desired text (here, "Improve your skin" "Maybe it's possible?") and stop there for a certain period of time. The position of the text box 21c corresponding to the text is displayed in the determination result list display area 21. The text box 21c is displayed at the top of the As a result, for example, in FIG. 15, the determination result list display area 21 The text box 21c that was displayed in a reduced size in the middle row is expanded at the top, and the text The reason information and alternative information associated with each of the items are displayed in a visually recognizable manner. The text box 21c is positioned at the top and expanded vertically, and the other The text boxes 21a and 2b to 21f are displayed below in the judgment result list display area 21. It slides to be displayed.

[0088] Returning to FIG. 12 again, the processor 112 receives, via the input interface 115, It is determined whether or not an instruction input to select the download icon has been accepted (S615 Specifically, the processor 112 receives the downloaded data from the input interface 115. An interrupt signal indicating that a click operation of the pointer was detected at the position corresponding to the icon is sent. If it is determined that the instruction input has been accepted, the process The processor 212 receives each text to be judged, and the corresponding information on whether or not the text is acceptable. List the reasons and alternatives and standardize the file format such as spreadsheet or PDF. Download the file in the format specified by the user to a storage location (S 616).

[0089] The processor 112 selects a print icon via the input interface 115. It is determined whether or not an instruction input to that effect has been received (S617). 112, from the input interface 115, the pointer is input at a position corresponding to the print icon. It is determined whether an interrupt signal indicating that a click operation has been detected is received. When it is determined that the instruction input has been accepted, the processor 212 displays the determination result display screen on paper. It can be output to media, or sent to users as standardized electronic file formats such as PDF or image data. The data is then downloaded to the storage location designated by the user (S618). The flow ends.

[0090] 9. Variations In the example of FIG. 16, the user moves the pointer 18 to a desired position via the input interface 115. Move the cursor over the text box and click to move the text box vertically. The text box is expanded and displayed as an extension. When a selection operation is detected, a pop-up window containing reason information and alternative information is displayed. When the selection operation is detected, a previously set position may be displayed. The reason information and the alternative information may be displayed in the display area. Each text box may be displayed expanded so that the information is always displayed.

[0091] In the example of FIG. 17, the user moves the pointer 18 to a desired position via the input interface 115. By moving the cursor to the text and pausing for a certain period of time, the corresponding text box will move. The case where the image is moved and expanded vertically is explained. By moving the cursor 18 to the desired text and stopping it for a certain period of time, the text corresponding to the text is displayed. You can also expand the text box in place without moving it. By moving 18 to the desired text and stopping it for a certain period of time, the text corresponding to the text will be displayed. The pointer may be displayed in an emphasized manner, for example, by blinking the check box. By moving 18 to the desired text and stopping it for a certain period of time, the text corresponding to the text will be displayed. Alternatively, you can just change the order in which the text boxes are displayed. By moving the text and stopping it for a certain period of time, reason information and alternatives are displayed at the position corresponding to the text. A pop-up window containing the information may be displayed.

[0092] In the examples of FIGS. 15 to 17, a permission / prohibition identification object indicating whether the text to be judged can be used is used. However, the case where the object is formed by a star-shaped decorative object has been explained. The shape of the object is not limited to this, and it can be any shape, such as a circle or a square. In the examples of FIGS. 15 to 17, the validity determination is performed at a position that does not overlap the text to be determined. Another object is displayed, but it is displayed behind or above the text. In the examples of FIGS. 15 to 17, the possibility identification object may be in a stationary state. Dynamic or color-changing objects are displayed. In the examples of FIGS. 15 to 17, the text to be judged and The object for identifying whether or not the text is displayed separately, but the text is displayed in bold, underlined, The display may be emphasized by highlighting or the like.

[0093] In the judgment result list display area of ​​Figures 15 to 17, the frame of each text box can be changed. The above explanation is for cases where the corresponding text can be used or not. , highlight the text in the text box, change the color of the text box, Use text by overlaying decorative objects on text boxes. It may also indicate whether or not

[0094] In the examples of Figures 15 to 17, the text and text boxes are selected using the pointer 18. However, this is not limited to this, and input interfaces such as touch panels may also be used. The user can select items by touching them with a finger or a stylus. You can do this.

[0095] In the examples of FIGS. 15 to 17, the judgment result display screen displays the content and the validity identification object. The case where the content display area 14 including the judgment result list display area 21 is included will be described. However, the determination result display area is not limited to this, and the content display area 14 may be included in the determination result display area. Alternatively, only the content and determination result list display area 21 may be displayed. In addition, the judgment result display area does not include the judgment result list display area 21. Alternatively, only the main display area 21 may be displayed.

[0096] In the examples of FIGS. 15 to 17, only the judgment result information of inappropriate and caution-required items is displayed on the judgment result display screen. However, depending on the received judgment result information, and appropriate" or "Appropriate and caution required" and then display only one of the following two or three pieces of information. In addition, it is also possible to set the classification to correspond to more detailed classifications according to the received judgment result information. Alternatively, the information indicating the numerical value or level indicating the probability of being judged as inappropriate or appropriate may be displayed. It may be shown.

[0097] Regarding the determination processes in S217, S219, and S220 in FIG. However, the present invention is not limited to this, and the determination itself may be performed by the The judgment is made by a user or other person using a base, and the process is carried out based on the memory of the judgment result. Alternatively, the sensor 212 may perform this.

[0098] In the examples of FIGS. 1 to 17, the server device 200 functions as a processing device. 6 to 11. However, the present invention is not limited to this. The terminal device 100 functions as a processing device, and the processor 112 of the terminal device 100 functions as a processing device. The process according to 1 may be executed.

[0099] As described above, in this embodiment, whether or not the product is subject to regulation by a predetermined law or guideline Processing device, processing program, and processing method capable of outputting the judgment result in a more understandable manner It is possible to provide a method for processing.

[0100] The processes and procedures described herein are in accordance with those explicitly described in the embodiments. It can be realized not only by computer but also by software, hardware or a combination of these. In particular, the processes and procedures described herein may be implemented in integrated circuits, volatile memories, , non-volatile memory, magnetic disk, optical storage, etc., and the logic corresponding to the processing The processes and procedures described in this specification are realized by implementing the These processes and procedures are implemented as computer programs, and include terminal devices and server devices. It can be executed on various computers.

[0101] The processes and procedures described herein may be implemented as a single device, software, component, Even if described as being performed by a module, such processes or procedures may Number of devices, software, components, and / or modules In addition, various information described in this specification can be Even if it is described as being stored in a single memory or storage unit, such information may be stored in a single Distributed across multiple memories installed in a device or distributed across multiple devices Furthermore, the software described herein may be stored as The software and hardware elements may be integrated into fewer components or This can be achieved by breaking it down into smaller components. [Explanation of symbols]

[0102] 100 Terminal Device 200 Server device 300 Network

Claims

1. Includes at least one processor, The at least one processor, The system accepts input content that includes one or more texts containing at least one word that is subject to evaluation. The system obtains one or more texts from the received content, and also obtains location information of the one or more texts in the content. For each of the one or more texts obtained, the permissibility of using each text within the content is determined based on laws or guidelines for determining whether the use of each text in the content is restricted. The determination result information includes at least permission information indicating whether each of the above texts can be used and reason information indicating the reason for the permission decision, and the permission information is output so as to be displayed superimposed on the content at the position within the content corresponding to the position information. A processing device configured to perform a process for that purpose.

2. The apparatus according to claim 1, wherein at least one processor is configured to output the determination result information from the determination result information as a first display superimposed on the content at a position corresponding to the position information, and to output the reason information as a second display in the form of a list provided in a display area separate from the area in which the content is displayed.

3. The content includes image data, The at least one processor is configured to perform a process for obtaining at least a portion of the one or more texts by performing optical character recognition processing on the image data in the content. The apparatus according to claim 1 or 2.

4. The at least one processor is By performing morphological analysis on the acquired one or more texts, the one or more texts are broken down into word or phrase units. The above determination is made based on each of the decomposed words or phrases. An apparatus according to any one of claims 1 to 3, configured to perform processing for the purpose of

5. The apparatus according to any one of claims 1 to 4, wherein the approval / disapproval information and the reason information are obtained by referring to a database in which approval / disapproval information and reason information are associated with each text in accordance with the laws or guidelines.

6. The approval / disapproval information is obtained by inputting the one or more texts into a trained text judgment model for determining whether each of the one or more texts contained in the content is permissible to use based on the laws or guidelines, The reason information is obtained by inputting the one or more texts into a trained reason determination model for determining whether each of the one or more texts contained in the content is permissible to use based on the laws or guidelines. The apparatus according to any one of claims 1 to 4.

7. The determination result information further includes alternative information indicating alternatives for texts that were not determined to be usable in the determination, The alternative information is obtained by inputting the one or more texts into a trained alternative determination model for determining alternatives to the one or more texts included in the content that were not determined to be usable. The apparatus according to any one of claims 1 to 6.

8. The processing apparatus according to any one of claims 1 to 7, wherein the at least one processor is configured to perform a process for outputting aggregate information related to at least one of the following, calculated based on the approval / disapproval information: the total number of characters in the content, the number of texts that were not deemed acceptable in the approval / disapproval, and the number of texts that were deemed to require attention.

9. The processing apparatus according to any one of claims 1 to 8, wherein the content is a web page containing HTML and image data.

10. The apparatus according to any one of claims 1 to 9, wherein the aforementioned laws or guidelines stipulate regulations concerning at least one of the following: pharmaceuticals, quasi-drugs, regenerative medicine products, medical devices, health foods, functional foods, cosmetics, and mail-order products.

11. By being executed by at least one processor, The text contains one or more words that are subject to evaluation. Accepting content input, The system obtains one or more texts from the received content, and also obtains location information of the one or more texts in the content. For each of the one or more texts obtained, the permissibility of using each text within the content is determined based on laws or guidelines for determining whether the use of each text in the content is restricted. The determination result information includes at least permission information indicating whether each of the above texts can be used and reason information indicating the reason for the permission decision, and the permission information is output so as to be displayed superimposed on the content at the position within the content corresponding to the position information. A processing program that causes the aforementioned at least one processor to function in this manner.

12. A processing method that is performed by at least one processor, A step of receiving input content that includes one or more texts containing at least one word to be judged, The steps include obtaining one or more texts from the received content and obtaining location information of the one or more texts in the content, For each of the one or more texts obtained, the step of determining whether the use of each text in the content is permitted based on laws or guidelines for determining whether the use of each text in the content is restricted, The steps include outputting judgment result information which includes at least permission information indicating whether each of the above texts can be used and reason information indicating the reason for the permission decision, such that the permission information is displayed superimposed on the content at a position within the content corresponding to the position information, A processing method that includes this.

Images