Information processing system, information processing method, and program

Abstract

To provide an information processing system, an information processing method, and a program for dealing with slander on the Internet.SOLUTION: In an information processing system 1000 in which a client device 110, a server device 100, and an AI system 120 are communicatively connected via a network 150, the client device 110 has at least one or more control units. The control units execute processing of non-displaying a newly posted message on a social networking service. The control units generate an instruction sentence requesting output of a confirmation result as to whether the message does not contain inappropriate or problematic discourse. The instruction sentence is input into a predetermined large-scale natural language model. If the confirmation result output from the predetermined large-scale natural language model indicates that the message does not contain inappropriate or problematic discourse, the control units execute processing of changing the message from non-display to display.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing system, an information processing method, and a program. [Background technology]

[0002] Patent Document 1 discloses a labeling technology related to a learning system that detects fraudulent behavior such as posting slanderous comments against others. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7302107 Summary of the Invention [Problem to be solved by the invention]

[0004] Preparing datasets for machine learning requires a lot of time and money. There is a need for new technologies to address these issues and combat defamation on the Internet. [Means for solving the problem]

[0005] According to one aspect of the present invention, an information processing system is provided. The information processing system has at least one or more control units. The control unit executes a process of hiding a newly posted message on a social networking service. The control unit generates an instruction message requesting output of a check result as to whether the message contains inappropriate or problematic discourse. The instruction message is input to a predetermined large-scale natural language model. If the check result output from the predetermined large-scale natural language model indicates that the message does not contain inappropriate or problematic discourse, the control unit executes a process of changing the message from hidden to visible. [Brief description of the drawings]

[0006] [Figure 1] FIG. 1 is a diagram illustrating an example of a system configuration of an information processing system. [Diagram 2] FIG. 2 is a diagram illustrating an example of a hardware configuration of the client device. [Diagram 3] FIG. 3 is a diagram illustrating an example of a hardware configuration of the server device. [Figure 4] FIG. 4 is a flowchart showing an example of information processing in the client device. [Diagram 5] FIG. 5 is an example of a command requesting output of the results of checking whether a message contains inappropriate or problematic language. [Figure 6] FIG. 6 is a diagram showing an example of a screen for selecting a category related to messages to be hidden. [Figure 7] FIG. 7 shows an example of a screen that allows users to select whether to hide messages when a policy of a company that provides AI such as ChatGPT is violated. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. Various characteristic features shown in the following embodiments can be combined with each other.

[0008] In particular, in this specification, the term "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In addition, while various information is handled in the first embodiment, communication and calculation can be performed on a circuit in the broad sense regardless of whether the information is represented by a high or low signal value as a binary bit collection consisting of 0 or 1, represented by a physical numerical value of a signal value, or represented by a quantum superposition.

[0009] In addition, a circuit in the broad sense is a circuit realized by at least appropriately combining a circuit, circuitry, a processor, a memory, etc. In other words, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0010] In addition, the program for realizing the software appearing in the embodiments may be implemented in a form that is downloadable from a server, the program may be executed on a cloud computer, or the program may be stored in a non-volatile or volatile non-transitory storage medium and distributed.

[0011] <Embodiment 1> 1. System Configuration Fig. 1 is a diagram showing an example of a system configuration of an information processing system 1000. As shown in Fig. 1, the information processing system 1000 includes, as a system configuration, a client device 110, a server device 100, and an AI system 120. The client device 110 and the server device 100 are communicatively connected via a network 150. The server device 100 and the AI ​​system 120 are communicatively connected via the network 150.

[0012] In response to a request from the client device 110, the server device 100 transmits to the client device 110 a program that provides an extended function to a predetermined web browser of the client device 110. The client device 110 is a device operated by a user, logs into a social networking service (SNS), posts messages on the SNS, and displays messages posted by other users. The AI ​​system 120 is a system that provides the function of a large-scale natural language model. In the specification, ChatGPT is used as an example of a large-scale natural language model. However, this does not limit the embodiment.

[0013] 1, for simplicity of explanation, only one client device 110 is illustrated in the information processing system 1000. However, the information processing system 1000 may include as many client devices 110 as there are users. 1 shows a PC (Personal Computer) as an example of the client device 110, but is not limited to a PC (Personal Computer) and may be a smartphone or a tablet computer. In other words, the client device 110 may be any device that has a predetermined web browser installed and can post and display messages on an SNS.

[0014] Here, the information processing system described in the claims may be composed of multiple devices or may be composed of one device. When the information processing system described in the claims is composed of one device, an example of the device is client device 110.

[0015] 2. Hardware Configuration (1) Hardware Configuration of the Client Device 110 FIG. 2 is a diagram showing an example of a hardware configuration of the client device 110. The client device 110 includes, as a hardware configuration, a control unit 210, a storage unit 220, an input unit 230, an output unit 240, and a communication unit 250. The control unit 210 is a CPU (Central Processing Unit) or the like, and controls the entire client device 110. The storage unit 220 is any one of an HDD (Hard Disk Drive), a ROM (Read Only Memory), a RAM (Random Access Memory), and an SSD (Solid State Drive), or any combination thereof, and stores a program and data used when the control unit 210 executes processing based on the program. The control unit 210 executes processing based on a program that provides an extension function of a predetermined Web browser provided by the server device 100, thereby realizing the processing of the client device 110 and the function of the client device 110 in the flowchart shown in FIG. 4 described later. In the following, for the sake of simplicity, the description will be given assuming that the control unit 210 executes the processing of the specification. The program provided from the server device 100 is stored in, for example, the storage unit 220. The input unit 230 is a keyboard and / or a mouse, and inputs user operations to the control unit 210. The input unit 230 may further be a microphone and a speaker, and input the user's voice to the control unit 210. The output unit 240 is a display, and displays the results of information processing by the control unit 210 and / or data acquired from other devices. The communication unit 250 is a NIC (Network Interface Card), and connects the client device 110 to the network 150 and controls communication with other devices. (2) Hardware Configuration of Server Device 100 FIG. 3 is a diagram showing an example of a hardware configuration of the server device 100. The server device 100 includes a control unit 310, a storage unit 320, and a communication unit 330 as a hardware configuration. The control unit 310 is a CPU or the like, and controls the entire server device 100. The storage unit 320 is any one of a HDD, a ROM, a RAM, an SSD, or the like, or any combination thereof, and stores a program and data used when the control unit 310 executes a process based on the program. The control unit 310 executes a process based on a program stored in the storage unit 320, thereby realizing functions of the server device 100. The communication unit 330 is a NIC or the like, and connects the server device 100 to the network 150 and manages communication with other devices. In this embodiment, the data used by the control unit 310 when executing processing based on a program is described as being stored in the memory unit 320, but the data may also be stored in a memory unit or the like of another device with which the server device 100 can communicate.

[0016] 3. Information Processing The information processing of this embodiment will be described below. The control unit 210 executes a process of hiding a newly posted message on the social networking service. The control unit 210 generates an instruction sentence (a so-called prompt) requesting output of the result of checking whether the message contains any inappropriate or problematic statements. The control unit 210 inputs the instruction text into ChatGPT via an API (Application Programming Interface). If the confirmation result output from ChatGPT indicates that the message does not contain any inappropriate or problematic statements, the control unit 210 executes a process of changing the message from hidden to visible.

[0017] If the verification result output from ChatGPT indicates that the message contains inappropriate or problematic language, the control unit 210 leaves the message hidden.

[0018] The control unit 210 accepts an input of a category via the screen. The category is, for example, a category indicating the content of a message on an SNS, such as sexual, hate, harassment, violence, etc. The control unit 210 generates an instruction message requesting output of the result of checking whether the message contains any inappropriate or problematic statements belonging to a category.

[0019] The control unit 210 executes a process of hiding a newly posted message on the timeline of a social networking service or a message addressed to the user.

[0020] When the control unit 210 receives a setting to turn on a predetermined operation mode, the control unit 210 executes a process to hide a newly posted message on the social networking service. The control unit 210 generates an instruction requesting output of the results of checking whether the message contains any inappropriate or problematic statements. The control unit 210 inputs the instruction text into ChatGPT. If the confirmation result output from ChatGPT indicates that the message does not contain any inappropriate or problematic statements, the control unit 210 executes a process of changing the message from hidden to visible.

[0021] The control unit 210 accepts a setting to turn on a predetermined operation mode via the screen. The control unit 210 is a screen on which a timeline of a social networking service is displayed.

[0022] FIG. 4 is a flowchart showing an example of information processing in the client device 110. The first step is to connect to the API of the selected social networking site and temporarily hide newly posted Tweets at an initial stage. New Tweets are kept hidden from users until they are evaluated via the Moderation API, which is based on the OpenAI API. Then, individual processing of Tweets begins. Each hidden Tweet is processed individually for evaluation by the Moderation API. This API analyzes the content of Tweets to check whether they contain inappropriate or problematic statements. Based on the response from the Moderation API, the process branches into two paths: if problematic statements are detected, the Tweet will continue to remain hidden, or if it is deemed OK, the Tweet will be made public and available for users to view. Finally, the generation of new tweets is monitored, and if a new tweet is detected, the process returns to the initial filtering phase and the above-mentioned process is repeated.

[0023] FIG. 5 is an example of a command requesting output of the results of checking whether a message contains inappropriate or problematic language. The prompt shown in Figure 5 has the following structure: (1) Clearly explain the task. (2) Define the meaning of "Toxic" in context. (3) Present the text to be evaluated. (4) Ask the model to make a judgment based on the explanations and definitions you provide.

[0024] Fig. 6 is a diagram showing an example of a screen for selecting a category related to messages to be hidden. When a predetermined operation is performed on the screen of a user's SNS account, the control unit 210 displays a screen as shown in Fig. 6. The same applies to Fig. 7 described later.

[0025] FIG. 7 shows an example of a screen that allows users to select whether to hide messages when a policy of a company that provides AI such as ChatGPT is violated.

[0026] According to this embodiment, it is possible to secure the right to remain ignorant of libel and slander, as well as freedom of expression, while using a large-scale natural language model provided by an external system such as ChatGPT. In addition, according to this embodiment, it is possible to provide a new technology for dealing with libel and slander on the Internet. Since it is not necessary to create a trained model from scratch, it is possible to save time and money.

[0027] <Additional Notes> The invention may be provided in each of the following aspects:

[0028] (1) An information processing system having at least one control unit, the control unit executing a process of hiding a newly posted message on a social networking service, generating an instruction statement requesting output of a confirmation result as to whether the message contains any inappropriate or problematic language, inputting the instruction statement into a predetermined large-scale natural language model, and, if the confirmation result output from the predetermined large-scale natural language model indicates that the message does not contain any inappropriate or problematic language, executing a process of changing the message from hidden to displayed.

[0029] (2) In the information processing device described in (1) above, the control unit keeps the message hidden if the confirmation result output from the specified large-scale natural language model indicates that the message contains inappropriate or problematic discourse.

[0030] (3) In the information processing system described in (1) or (2) above, the control unit accepts input of a category via a screen and generates an instruction statement requesting output of a confirmation result as to whether the message contains inappropriate or problematic discourse belonging to the category.

[0031] (4) In the information processing system described in any one of (1) to (3) above, the control unit executes a process of hiding a newly posted message on a timeline of a social networking service or a message addressed to a user.

[0032] (5) In the information processing system described in any one of (1) to (4) above, when the control unit receives a setting to turn on a predetermined operating mode, it executes a process of hiding a newly posted message on a social networking service, generates an instruction statement requesting output of a confirmation result as to whether the message contains any inappropriate or problematic discourse, inputs the instruction statement into a predetermined large-scale natural language model, and, when the confirmation result output from the predetermined large-scale natural language model indicates that the message does not contain any inappropriate or problematic discourse, executes a process of changing the message from hidden to displayed.

[0033] (6) In the information processing system described in (5) above, the control unit accepts a setting to turn on the specified operating mode via a screen, and the screen is a screen on which a timeline of a social networking service is displayed.

[0034] (7) An information processing method executed by an information processing system, comprising: executing a process of hiding a newly posted message on a social networking service; generating an instruction statement requesting output of a confirmation result as to whether the message contains any inappropriate or problematic language; inputting the instruction statement into a predetermined large-scale natural language model; and, if the confirmation result output from the predetermined large-scale natural language model indicates that the message does not contain any inappropriate or problematic language, executing a process of changing the message from hidden to displayed.

[0035] (8) A program for causing a computer to function as the information processing system according to any one of (1) to (6) above. Of course, this is not the case.

[0036] For example, the above-mentioned program may be provided as a computer-readable non-transitory storage medium that stores the program.

[0037] Finally, although various embodiments of the present invention have been described, these are presented as examples and are not intended to limit the scope of the invention. The new embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments and their modifications are within the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0038] 100: Server device 110: Client device 120: AI system 150: Network 210: Control section 220: Storage section 230: Input section 240: Output section 250: Communications Department 310: Control section 320: Storage section 330: Communications Department 1000: Information processing systems

Claims

1. An information processing system, At least one control unit is included, The control unit is Execute a process to hide newly posted messages on a social networking service, generating an instruction requesting output of a result of checking whether the message contains inappropriate or problematic language; inputting said instruction into a predetermined large-scale natural language model; If the confirmation result output from the predetermined large-scale natural language model indicates that the message does not contain inappropriate or problematic language, a process of changing the message from hidden to displayed is executed. Information processing system.

2. 2. The information processing device according to claim 1, The control unit is if the validation results output from the predetermined large-scale natural language model indicate that the message contains inappropriate or problematic language, then leaving the message hidden; Information processing system.

3. 2. The information processing system according to claim 1, The control unit is Accepts category input via the screen, generating an instruction sentence for requesting output of a confirmation result as to whether the message contains inappropriate or problematic speech belonging to the category; Information processing system.

4. 2. The information processing system according to claim 1, The control unit is Execute a process to hide newly posted messages on the timeline of a social networking service or messages addressed to the user; Information processing system.

5. 2. The information processing system according to claim 1, The control unit is When the setting to turn on a specific operation mode is accepted, Execute a process to hide newly posted messages on a social networking service, generating an instruction requesting output of a result of checking whether the message contains inappropriate or problematic language; inputting said instruction into a predetermined large-scale natural language model; If the confirmation result output from the predetermined large-scale natural language model indicates that the message does not contain inappropriate or problematic language, a process of changing the message from hidden to displayed is executed. Information processing system.

6. 6. The information processing system according to claim 5, The control unit is accepting a setting to turn on the predetermined operation mode via a screen; The screen is a screen that displays a timeline of a social networking service. Information processing system.

7. An information processing method executed by an information processing system, comprising: Execute a process to hide newly posted messages on a social networking service, generating an instruction requesting output of a result of checking whether the message contains inappropriate or problematic language; inputting said instruction into a predetermined large-scale natural language model; If the confirmation result output from the predetermined large-scale natural language model indicates that the message does not contain inappropriate or problematic language, a process of changing the message from hidden to displayed is executed. Information processing methods.

8. A program, Computer, A program for causing the information processing system according to any one of claims 1 to 6 to function as such.

Images