Apparatus and method for blocking harmful expressions

Abstract

The present disclosure relates to an apparatus and a method for blocking harmful expressions, wherein, if content is received, situation information is identified; a harmfulness level database corresponding to the situation information is identified; it is confirmed, with reference to the harmfulness level database, whether the content includes harmful information; if the content includes harmful information, it is determined whether to remove the harmful information from the content or to change same; and according to the determination, the harmful information may be removed from the content or changed through a harmful content processing engine.

Description

Device and method for blocking harmful expressions

[0001] The following embodiments relate to a device and method for blocking harmful expressions.

[0002] Modern smartphones are becoming increasingly useful due to their integration with various input / output devices. Through connectivity methods such as Bluetooth, Wi-Fi, and USB, smartphones can connect to a variety of shared devices, including public speakers, displays, car navigation systems, and voice recognition devices. While these connections offer users convenience, they can also lead to the output of unwanted, profane language and other harmful information via voice or text on shared devices. This can lead to serious social problems, especially in public spaces and work environments.

[0003] When connected to a public device, a smartphone can perform various functions through voice commands or text input. However, if audio or text files stored on the smartphone, or messages transmitted in real time, contain harmful information, this information may be displayed on the public device, causing discomfort. For example, when a smartphone is connected to a public display and a document is displayed, text containing harmful information, such as profanity, may appear.

[0004] Problems can also arise when using smartphone voice recognition. When a smartphone recognizes a user's voice and converts it to text or executes a voice command, if harmful information is recognized and output, it can lead to inappropriate situations in public places or work environments. These issues can lead to a decline in trust among users of public devices and social discomfort, necessitating technological solutions to address them.

[0005] There have been several attempts to address this issue to date, but most solutions are limited to simple filtering without considering the user's context, making them difficult to address in a variety of situations.

[0006] According to one embodiment of the present disclosure, a device and method for blocking harmful expressions can be provided.

[0007] A method for blocking harmful expressions according to one embodiment may include: receiving content; checking context information; checking a harmful level database corresponding to the context information; checking whether harmful information exists in the content based on the harmful level database; determining whether to remove or change the harmful information from the content if the harmful information exists in the content; and removing or changing the harmful information from the content through a harmful processing engine based on the determination.

[0008] According to one embodiment, a computer-readable recording medium stores commands, which, when executed by one or more processors, may perform the following actions: receiving content; checking context information; checking a harmful level database corresponding to the context information; checking whether harmful information exists in the content based on the harmful level database; determining whether to remove or change the harmful information from the content if the harmful information exists in the content; and, based on the determination, removing or changing the harmful information from the content through a harmful processing engine.

[0009] According to one embodiment, a device for blocking harmful expressions includes one or more processors and a memory storing commands, wherein the commands, when executed by the one or more processors, cause the device to perform the following operations: receiving content; checking context information; checking a harmful level database corresponding to the context information; checking whether harmful information exists in the content based on the harmful level database; determining whether to remove or change the harmful information from the content if the harmful information exists in the content; and removing or changing the harmful information from the content through a harmful processing engine based on the determination.

[0010] FIG. 1 is a diagram illustrating a configuration of a device for blocking harmful expressions according to one embodiment.

[0011] Figure 2 is a flowchart illustrating a flow for blocking harmful expressions according to one embodiment.

[0012] FIG. 3 is a diagram illustrating an example of a user interface provided to a user to block harmful expressions based on contextual information according to one embodiment.

[0013] FIG. 4 is a diagram illustrating an example of a harmful level database according to one embodiment.

[0014] FIG. 5 is a diagram illustrating an example of blocking harmful expressions according to age in a harmful expression blocking device according to one embodiment.

[0015] FIG. 6 is a drawing illustrating an example of blocking harmful expressions according to location in a harmful expression blocking device according to one embodiment.

[0016] FIG. 7 is a diagram illustrating an example of blocking harmful expressions according to a connection device in a harmful expression blocking device according to one embodiment.

[0017] FIG. 8 is a schematic diagram illustrating the structure of an electronic device according to one embodiment.

[0018] FIG. 9 is a diagram illustrating an electronic device within a network environment according to one embodiment.

[0019] Hereinafter, embodiments are described in detail with reference to the attached drawings. However, the embodiments may be modified in various ways, and the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, or alternatives to the embodiments are included within the scope of the patent application.

[0020] The terms used in the examples are for illustrative purposes only and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "comprise" or "have" are intended to indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but should be understood to not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0021] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiments pertain. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and shall not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0022] In addition, when describing with reference to the attached drawings, identical components will be assigned the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted. When describing embodiments, if a detailed description of a related known technology is judged to unnecessarily obscure the gist of the embodiment, the detailed description will be omitted.

[0023] Additionally, terms such as first, second, A, B, (a), (b), etc. may be used to describe components of the embodiments. These terms are only intended to distinguish the components from other components, and the nature, order, or sequence of the components are not limited by the terms. When a component is described as being "connected," "coupled," or "connected" to another component, it should be understood that the component may be directly connected or connected to the other component, but another component may also be "connected," "coupled," or "connected" between each component.

[0024] Components included in one embodiment and components with common functions will be described using the same names in other embodiments. Unless otherwise stated, the descriptions given in one embodiment may also apply to other embodiments, and detailed descriptions will be omitted to the extent of overlap.

[0025]

[0026] Hereinafter, a harmful expression blocking device and method according to an embodiment of the present invention will be described in detail with reference to the attached FIGS. 1 to 13.

[0027] FIG. 1 is a diagram illustrating a configuration of an electronic device for blocking harmful expressions according to one embodiment.

[0028] Referring to FIG. 1, an electronic device (100) may be configured to include a processor (110), a memory (120), and a large language model (130).

[0029] The processor (110) can perform the operations of an input filter (112), a harmful searcher (114), a harmful regulator (116), and an output filter (118) to block harmful expressions.

[0030] When the input filter (112) receives content, it checks whether the content contains harmful information through a harmful search device (114), and if the content contains harmful information, it removes or changes the harmful information through a harmful control device (116), and transmits content without harmful information or content with harmful information removed or changed to an output filter (118).

[0031] At this time, the content may be text, voice, image, or video. Furthermore, the content may include context. The context may include status information of the electronic device (100). More specifically, the context may include information about the current electronic device (100), its software version and model name, screen size information, Bluetooth device connection information, and silent status information.

[0032] The harmful search engine (114) can check the situation information, check the harmful level database (122) corresponding to the situation information in the memory (120), and check whether harmful information exists in the content based on the harmful level database (122).

[0033] The harmful search engine (114) can check situational information through the context included in the content. The context may include information about the current electronic device (100), software version and model name, screen size information, Bluetooth device connection information, and silent status information.

[0034] Contextual information may include at least one of the user's age, current location, set routines, and connected devices.

[0035] The age of the user of the context information can be determined by whether the user is connected to a preset mode (e.g., Kids Mode, etc.) or through the account information the user is connected to, and can be categorized into preset age groups (e.g., children, teenagers, 15 years old or younger, 19 years old or younger, adults, etc.).

[0036] The current location of the situation information can be confirmed through GPS information, connected Wi-Fi, and connected devices, and can be classified into a location set by the user.

[0037] The connection device of the situation information is a device connected to an electronic device (100) that provides harmful expression blocking, and may be a wireless connection device or a wired connection device, and may be classified into a personal device and a public device.

[0038] A set routine for context information can be a function that automatically performs certain actions when certain conditions are met, based on user settings.

[0039] The harmful level database (122) may exist for each situation information, or may be composed of a harmful level database (122) including multiple situations information. The harmful level database (122) may store a harmful blocking level corresponding to each harmful category. The harmful blocking level may be composed of several levels, for example, off, low, mid, and high. The harmful blocking level for each harmful category may be set by the user, and if the user does not make a separate setting, it may be automatically set to a preset level. In addition, the harmful category may include at least one of, for example, Illegal, Sexual, Child Abuse, Violent, Derogatory, Toxic, Insult, Profanity, Sensitive, Self-harm, and Generation of malware.

[0040] More specifically, the harmful search engine (114) can check whether harmful information exists in the content according to the harmful blocking level defined for each harmful category included in the harmful level database (122) through the harmful search engine. In addition, the harmful search engine can be implemented using at least one of a rule-based engine, a deep model-based engine, a small large language model (sLLM), and a large language model (LLM).

[0041] The harmful content search engine (114) can detect the presence of harmful information by detecting major keywords or patterns through a rule-based harmful content search engine, and can utilize a deep model-based harmful content search engine that considers context and circumstances rather than a rule-based engine to determine harmfulness. In addition, the harmful content search engine (114) can also determine whether received content contains harmful information through a model such as a small large language model (sLLM) or a separate large language model (LLM).

[0042] The harmful information regulator (116) determines whether to remove or change harmful information from the content, and based on the determination, the harmful information can be removed or changed from the content through a large language model (LLM) (130).

[0043] At this time, the decision to remove or modify harmful information from content may be determined based on contextual information, or based on whether the harmful information has exceeded the harmful blocking level of a specific category among the harmful categories included in the harmful level database (122). For example, the harmful information regulator (116) may remove harmful information if the user's age is a child based on contextual information, or if the electronic device (100) is located in a company and connected to a company conference room speaker.

[0044] The harmful information regulator (116) can remove harmful information from the content if it is determined to remove harmful information from the content.

[0045] When the harmful information regulator (116) decides to change the harmful information, it can change the harmful information contained in the text, images, and voices included in the content using a large language model (LLM) (130) according to the corresponding harmful blocking level for each harmful category included in the harmful level database (122).

[0046] In the present disclosure, the large language model (LLM) (130) is located outside the processor (110), but may be implemented in a form located inside the processor (110) or may be implemented outside the electronic device (100).

[0047] In addition, the harmful information regulator (116) changes the harmful information of the content using a large language model (LLM) (130), but may change it using another harmful processing engine. In this case, the harmful processing engine may be implemented using at least one of a rule-based engine, a deep model-based engine, a small large language model (sLLM), and a large language model (LLM).

[0048] When the harmful modifier (116) utilizes a rule-based harmful processing engine, voice and text can be changed into purified expressions mapped to specific keywords and patterns, and in the case of images, if they fall into a harmful category, they can be returned as statically fixed image information. In addition, when the harmful modifier (116) utilizes a deep model-based harmful processing engine, it can be changed into purified expressions that take into account the context and situation rather than a rule-based harmful processing engine. The harmful modifier (116) can also regenerate content by changing the received content into an appropriate purified expression through a generative model, such as a small large language model (sLLM) or a separate large language model (LLM). The large language model (130) can use a generally used public large language model (public LLM) or an LLM specialized in fine-tuning by a separate company.

[0049] A large language model (130) is an artificial intelligence model learned based on a large amount of text data and can be used for various language processing tasks such as sentence generation, translation, summarization, and question-answering.

[0050] The output filter (118) checks whether the content received from the outside or the changed content contains harmful information through the harmful search unit (114) of the content received from the outside or the content changed through the large language model (LLM) (130), and if the content received from the outside or the content changed contains harmful information, removes or changes the harmful information through the harmful control unit (116), and outputs the content without harmful information or the content with the harmful information removed or changed. At this time, the content received from the outside can be, for example, messages, messages or data (images or videos) transmitted by other users through applications (e.g., KakaoTalk), and responses generated from services.

[0051] The memory (120) can store various data used by at least one component of the electronic device (100). The data can include, for example, input data or output data for software and commands related thereto. The memory (120) can include volatile memory or non-volatile memory. According to the present disclosure, the memory (120) can store a hazard level database (122).

[0052]

[0053] Hereinafter, the method according to the present disclosure configured as above will be described with reference to the drawings below.

[0054] Figure 2 is a flowchart illustrating a flow for blocking harmful expressions according to one embodiment.

[0055] Referring to FIG. 2, in operation 210, the electronic device (100) may receive content. The content may be text, voice, image, or video. Furthermore, the content may include context. The context may include status information of the electronic device (100). More specifically, the context may include information about the current electronic device (100), software version and model name, screen size information, Bluetooth device connection information, and silent status information.

[0056] In operation 212, the electronic device (100) can check contextual information. At this time, the electronic device (100) can check contextual information through the context included in the content. The contextual information may include at least one of the user's age, current location, set routine, and connected device.

[0057] The age of the user of the context information can be determined by whether the user is connected to a preset mode (e.g., Kids Mode, etc.) or through the account information the user is connected to, and can be categorized into preset age groups (e.g., children, teenagers, 15 years old or younger, 19 years old or younger, adults, etc.).

[0058] The current location of the situation information can be confirmed through GPS information, connected Wi-Fi, and connected devices, and can be classified into a location set by the user.

[0059] The connection device of the situation information is a device connected to an electronic device (100) that provides harmful expression blocking, and may be a wireless connection device or a wired connection device, and may be classified into a personal device and a public device.

[0060] A set routine for context information can be a function that automatically performs certain actions when certain conditions are met, based on user settings.

[0061] The electronic device (100) can set a harmful blocking level according to situational information in advance.

[0062] FIG. 3 is a diagram illustrating an example of a user interface provided to a user to block harmful expressions based on contextual information according to one embodiment.

[0063] Referring to FIG. 3, the electronic device (100) provides a user interface such as a harmful blocking custom setting screen (310), a device-specific custom setting screen (320), and a device-specific custom setting screen (330) to the user so that the user can set a harmful blocking level for each harmful category according to the situation.

[0064] The harmful blocking custom settings screen (310) is a screen where you can select child settings that set the user's age, settings by routine, and settings by device.

[0065] The device-specific customization screen (320) is a screen for adding or selecting connected devices.

[0066] The device-specific custom settings screen (330) is a screen for setting the harmful blocking level for each harmful category of the selected device.

[0067] Returning to the description of FIG. 2, in operation 214, the electronic device (100) can check the harmful level database corresponding to the situation information. At this time, the harmful level database may exist for each situation information, or may be composed of a harmful level database including multiple situations information.

[0068] FIG. 4 is a diagram illustrating an example of a harmful level database according to one embodiment.

[0069] Referring to FIG. 4, the harmful level database includes harmful categories, and it can be confirmed whether the harmful blocking level corresponding to each harmful category is preset (auto) or set by the user (manual).

[0070] Additionally, the harmful level database can store the corresponding harmful blocking level for each harmful category. The harmful blocking level can be composed of multiple levels, for example, off, low, mid, and high. The harmful blocking level for each harmful category can be set by the user, and if the user does not make a separate setting, it can be automatically set to a preset level. In addition, the harmful category can include at least one of, for example, Illegal, Sexual, Child Abuse, Violent, Derogatory, Toxic, Insult, Profanity, Sensitive, Self-harm, and Generation of malware.

[0071] Returning to the description of FIG. 2, in operation 220, the electronic device (100) can check whether harmful information exists in the content based on the harmful level database.

[0072] At this time, the electronic device (100) can check whether harmful information exists in the content based on the harmful blocking level defined for each harmful category included in the harmful level database through the harmful search engine. In addition, the harmful search engine can be implemented using at least one of a rule-based engine, a deep model-based engine, a small large language model (sLLM), and a large language model (LLM).

[0073] More specifically, the electronic device (100) can confirm the presence of harmful information by detecting major keywords or patterns through a rule-based harmful search engine, and can utilize a deep model-based harmful search engine that considers context and situation rather than a rule-based engine to determine harmfulness. In addition, the electronic device (100) can also determine whether received content contains harmful information through a model such as a small large language model (sLLM) or a separate large language model (LLM).

[0074] If the result of the confirmation of operation 220 shows that there is no harmful information in the content, the electronic device (100) can output the received content as is in operation 222.

[0075] If, as a result of the confirmation of operation 220, harmful information exists in the content, in operation 230, the electronic device (100) can decide whether to remove or change the harmful information from the content.

[0076] If it is determined as a result of the confirmation of operation 230 to change the harmful information in the content, in operation 232, the electronic device (100) can change the harmful information contained in the text, images, and voices contained in the content using the harmful processing engine according to the harmful blocking level corresponding to each harmful category contained in the harmful level database.

[0077] At this time, the harmful processing engine can be implemented using at least one of a rule-based engine, a deep model-based engine, a small large language model (sLLM), and a large language model (LLM).

[0078] When the electronic device (100) utilizes a rule-based harmful processing engine, voice and text can be changed into purified expressions mapped to specific keywords and patterns, and in the case of images, if they fall into a harmful category, they can be returned as statically fixed image information. In addition, when the electronic device (100) utilizes a deep model-based harmful processing engine, it is possible to change into purified expressions that take into account the context and situation rather than the rule-based harmful processing engine. The electronic device (100) can also regenerate content by changing the received content into an appropriate purified expression through a generative model, such as a small large language model (sLLM) or a separate large language model (LLM). The large language model can use a generally used public large language model (LLM) or an LLM specialized in fine-tuning by a separate company.

[0079] In operation 240, the electronic device (100) can re-examine whether harmful information exists in the content changed in operation 232.

[0080] If the changed content contains harmful information as a result of the confirmation of operation 240, the electronic device (100) can return to operation 230 and repeat a series of operations.

[0081] In operation 230, if the electronic device (100) determines that harmful information exists in the content that has been changed a preset number of times even though the harmful information has been changed, it may decide to remove the harmful information.

[0082] If the result of the confirmation of operation 240 shows that there is no harmful information in the changed content, the electronic device (100) can output the changed content in operation 242.

[0083] If it is determined as a result of the verification of operation 230 to remove harmful information from the content, the electronic device (100) can remove the harmful information from the content in operation 250.

[0084] And, in operation 252, the electronic device (100) can output content with enhanced harmful information. At this time, the electronic device (100) can indicate that the harmful information has been removed.

[0085]

[0086] Then, examples of blocking harmful expressions are explained below with reference to Figures 5 to 7.

[0087] FIG. 5 is a diagram illustrating an example of blocking harmful expressions according to age in a harmful expression blocking device according to one embodiment.

[0088] Referring to FIG. 5, when comparing the result content (514) according to the request (512) of an adult (510) saying “Listen to the daily briefing today” and the result content (524) according to the request (522) of a child (520) saying “Listen to the daily briefing today”, it can be confirmed that in the result content (524) according to the request (522) of the child (520), the harmful information “murder case” is changed to “violent crime” and provided.

[0089]

[0090] FIG. 6 is a drawing illustrating an example of blocking harmful expressions according to location in a harmful expression blocking device according to one embodiment.

[0091] Referring to FIG. 6, a comparison of the resulting content according to the request for "read the text message I just received" can be confirmed when the location of the electronic device (100) is a company (620) and when it is not a company (610). It can be confirmed that when the location of the electronic device (100) is a company (620), the electronic device (100) changes the expression "ㅅㅂ, you're so good", which is a vulgar expression, to "You're really good" and provides it.

[0092]

[0093] FIG. 7 is a diagram illustrating an example of blocking harmful expressions according to a connection device in a harmful expression blocking device according to one embodiment.

[0094] Referring to FIG. 7, a comparison of contents (752, 754, 756) according to a request (740) to “tell me about recent incidents and accidents” can be confirmed in cases where the device connected to the electronic device (100) is a speaker in a small room (710), in cases where the device connected to the electronic device (100) is a speaker in a living room (720), and in cases where the device connected to the electronic device (100) is a speaker in a bedroom (730).

[0095] In the case of a speaker in a small room where the harmful blocking level of the harmful category Violent is set to low (710), it can be confirmed that content (752) displaying harmful information related to violence without modification is output.

[0096] In the case of a speaker in a living room where the harmful blocking level for violence among the harmful categories is set to high (720), it can be confirmed that content (754) displayed as murder information with harmful information related to violence as refined as possible is output.

[0097] In the case of a speaker in the living room where the harmful blocking level for violence among the harmful categories is set to mid (730), it can be confirmed that content (756) displaying harmful information related to violence in a slightly toned-down form is output.

[0098]

[0099] Meanwhile, the electronic device (100) for blocking harmful expressions of FIG. 1 may be configured in the form of an electronic device (800) as in FIG. 8 below, or may be configured in the form of an electronic device (901) in a network environment as in FIG. 9.

[0100] FIG. 8 is a schematic diagram illustrating the structure of an electronic device according to one embodiment.

[0101] Referring to FIG. 8, an electronic device (800) may be configured to include a processor (810) and a memory (820).

[0102] The memory (820) can store various data used by at least one component (e.g., the processor (810)) of the electronic device (800). The data can include, for example, input data or output data for software and commands related thereto. The memory (820) can include volatile memory or non-volatile memory. In this case, the memory (820) can have a configuration corresponding to the memory (120) of FIG. 1.

[0103] The processor (810) can control the overall operation of the electronic device (800). When receiving content, the processor (810) verifies contextual information, verifies a harmful level database corresponding to the contextual information, verifies whether harmful information exists in the content based on the harmful level database, determines whether the harmful information exists in the content, determines whether to remove or change the harmful information from the content, and, based on the determination, controls the harmful information to be removed or changed from the content through a harmful processing engine. At this time, the processor (810) may be configured with multiple processors. At this time, the processor (810) may have a configuration corresponding to the processor (110) of FIG. 1 .

[0104]

[0105] FIG. 9 is a diagram illustrating an electronic device within a network environment according to one embodiment.

[0106] Referring to FIG. 9, in a network environment (900), an electronic device (901) may communicate with an electronic device (902) via a first network (998) (e.g., a short-range wireless communication network), or may communicate with an electronic device (904) or a server (908) via a second network (999) (e.g., a long-range wireless communication network). In one embodiment, the electronic device (901) may communicate with the electronic device (904) via the server (908). According to one embodiment, the electronic device (901) may include a processor (920), a memory (930), an input module (950), an audio output module (955), a display module (960), an audio module (970), a sensor module (976), an interface (977), a connection terminal (978), a haptic module (979), a camera module (980), a power management module (988), a battery (989), a communication module (990), a subscriber identification module (996), or an antenna module (997). In some embodiments, the electronic device (901) may omit at least one of these components (e.g., the connection terminal (978)), or may have one or more other components added. In some embodiments, some of these components (e.g., the sensor module (976), the camera module (980), or the antenna module (997)) may be integrated into one component (e.g., the display module (960)).

[0107] The processor (920) may, for example, execute software (e.g., a program (940)) to control at least one other component (e.g., a hardware or software component) of the electronic device (901) connected to the processor (920) and perform various data processing or operations. According to one embodiment, as at least a part of the data processing or operations, the processor (920) may store commands or data received from other components (e.g., a sensor module (976) or a communication module (990)) in a volatile memory (932), process the commands or data stored in the volatile memory (932), and store result data in a non-volatile memory (934). At this time, the processor (920) may have a configuration corresponding to the processor (110) of FIG. 1.

[0108] According to one embodiment, the processor (920) may include a main processor (921) (e.g., a central processing unit or an application processor) or an auxiliary processor (923) (e.g., a graphics processing unit, a neural processing unit (NPU), an image signal processor, a sensor hub processor, or a communication processor) that can operate independently or together with the main processor (921). For example, when the electronic device (901) includes the main processor (921) and the auxiliary processor (923), the auxiliary processor (923) may be configured to use less power than the main processor (921) or to be specialized for a given function. The auxiliary processor (923) may be implemented separately from the main processor (921) or as a part thereof.

[0109] The auxiliary processor (923) may control at least a portion of functions or states associated with at least one component (e.g., a display module (960), a sensor module (976), or a communication module (990)) of the electronic device (901), for example, on behalf of the main processor (921) while the main processor (921) is in an inactive (e.g., sleep) state, or together with the main processor (921) while the main processor (921) is in an active (e.g., application execution) state. In one embodiment, the auxiliary processor (923) (e.g., an image signal processor or a communication processor) may be implemented as a part of another functionally related component (e.g., a camera module (980) or a communication module (990)). In one embodiment, the auxiliary processor (923) (e.g., a neural network processing unit) may include a hardware structure specialized for processing artificial intelligence models. The artificial intelligence models may be generated through machine learning. This learning can be performed, for example, on the electronic device (901) itself where the artificial intelligence is performed, or can be performed through a separate server (e.g., server (908)). The learning algorithm can include, for example, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but is not limited to the examples described above. The artificial intelligence model can include multiple artificial neural network layers.The artificial neural network may be one of a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a restricted Boltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), a deep Q-network, or a combination of two or more of the above, but is not limited to the examples described above. In addition to, or alternatively to, a hardware structure, an artificial intelligence model may include a software structure.

[0110] The memory (930) can store various data used by at least one component (e.g., the processor (920) or the sensor module (976)) of the electronic device (901). The data can include, for example, software (e.g., the program (940)) and input data or output data for commands related thereto. The memory (930) can include a volatile memory (932) or a non-volatile memory (934). In this case, the memory (930) can have a configuration corresponding to the memory (120) of FIG. 1.

[0111] The program (940) may be stored as software in the memory (930) and may include, for example, an operating system (942), middleware (944), or an application (946).

[0112] The input module (950) can receive commands or data to be used in a component of the electronic device (901) (e.g., a processor (920)) from an external source (e.g., a user) of the electronic device (901). The input module (950) can include, for example, a microphone, a mouse, a keyboard, a key (e.g., a button), or a digital pen (e.g., a stylus pen).

[0113] The audio output module (955) can output audio signals to the outside of the electronic device (901). The audio output module (955) can include, for example, a speaker or a receiver. The speaker can be used for general purposes, such as multimedia playback or recording playback. The receiver can be used to receive incoming calls. In one embodiment, the receiver can be implemented separately from the speaker or as part of the speaker.

[0114] The display module (960) can visually provide information to an external party (e.g., a user) of the electronic device (901). The display module (960) may include, for example, a display, a holographic device, or a projector and a control circuit for controlling the device. In one embodiment, the display module (960) may include a touch sensor configured to detect a touch, or a pressure sensor configured to measure the intensity of a force generated by the touch.

[0115] The audio module (970) can convert sound into an electrical signal, or vice versa, convert an electrical signal into sound. According to one embodiment, the audio module (970) can acquire sound through the input module (950), output sound through the sound output module (955), or an external electronic device (e.g., electronic device (902)) (e.g., speaker or headphone) directly or wirelessly connected to the electronic device (901).

[0116] The sensor module (976) can detect the operating status (e.g., power or temperature) of the electronic device (901) or the external environmental status (e.g., user status) and generate an electrical signal or data value corresponding to the detected status. According to one embodiment, the sensor module (976) can include, for example, a gesture sensor, a gyro sensor, a barometric pressure sensor, a magnetic sensor, an acceleration sensor, a grip sensor, a proximity sensor, a color sensor, an IR (infrared) sensor, a biometric sensor, a temperature sensor, a humidity sensor, or an illuminance sensor.

[0117] The interface (977) may support one or more designated protocols that may be used to directly or wirelessly connect the electronic device (901) with an external electronic device (e.g., the electronic device (902)). In one embodiment, the interface (977) may include, for example, a high definition multimedia interface (HDMI), a universal serial bus (USB) interface, an SD card interface, or an audio interface.

[0118] The connection terminal (978) may include a connector through which the electronic device (901) may be physically connected to an external electronic device (e.g., the electronic device (902)). In one embodiment, the connection terminal (978) may include, for example, an HDMI connector, a USB connector, an SD card connector, or an audio connector (e.g., a headphone connector).

[0119] The haptic module (979) can convert electrical signals into mechanical stimuli (e.g., vibration or movement) or electrical stimuli that a user can perceive through tactile or kinesthetic sensations. According to one embodiment, the haptic module (979) can include, for example, a motor, a piezoelectric element, or an electrical stimulation device.

[0120] The camera module (980) can capture still images and videos. According to one embodiment, the camera module (980) may include one or more lenses, image sensors, image signal processors, or flashes.

[0121] The power management module (988) can manage the power supplied to the electronic device (901). According to one embodiment, the power management module (988) can be implemented as, for example, at least a part of a power management integrated circuit (PMIC).

[0122] A battery (989) may power at least one component of the electronic device (901). In one embodiment, the battery (989) may include, for example, a non-rechargeable primary battery, a rechargeable secondary battery, or a fuel cell.

[0123] The communication module (990) may support the establishment of a direct (e.g., wired) communication channel or a wireless communication channel between the electronic device (901) and an external electronic device (e.g., electronic device (902), electronic device (904), or server (908)), and the performance of communication through the established communication channel. The communication module (990) may operate independently from the processor (920) (e.g., application processor) and may include one or more communication processors that support direct (e.g., wired) communication or wireless communication. According to one embodiment, the communication module (990) may include a wireless communication module (992) (e.g., a cellular communication module, a short-range wireless communication module, or a global navigation satellite system (GNSS) communication module) or a wired communication module (994) (e.g., a local area network (LAN) communication module, or a power line communication module). Any of these communication modules may communicate with an external electronic device (904) via a first network (998) (e.g., a short-range communication network such as Bluetooth, wireless fidelity (WiFi) direct, or infrared data association (IrDA)) or a second network (999) (e.g., a long-range communication network such as a legacy cellular network, a 5G network, a next-generation communication network, the Internet, or a computer network (e.g., a LAN or WAN)). These various types of communication modules may be integrated into a single component (e.g., a single chip) or implemented as multiple separate components (e.g., multiple chips). The wireless communication module (992) may use subscriber information (e.g., an international mobile subscriber identity (IMSI)) stored in the subscriber identification module (996) to identify or authenticate the electronic device (901) within a communication network such as the first network (998) or the second network (999).

[0124] The wireless communication module (992) can support 5G networks and next-generation communication technologies following the 4G network, such as NR access technology (new radio access technology). The NR access technology can support high-speed transmission of high-capacity data (eMBB (enhanced mobile broadband)), minimization of terminal power and connection of multiple terminals (mMTC (massive machine type communications)), or high reliability and low latency communications (URLLC (ultra-reliable and low-latency communications)). The wireless communication module (992) can support, for example, a high-frequency band (e.g., mmWave band) to achieve a high data transmission rate. The wireless communication module (992) can support various technologies for securing performance in high-frequency bands, such as beamforming, massive multiple-input and multiple-output (MIMO), full dimensional MIMO (FD-MIMO), array antenna, analog beam-forming, or large scale antenna. The wireless communication module (992) can support various requirements specified in the electronic device (901), an external electronic device (e.g., the electronic device (904)), or a network system (e.g., the second network (999)). According to one embodiment, the wireless communication module (992) can support a peak data rate (e.g., 20 Gbps or more) for eMBB realization, a loss coverage (e.g., 164 dB or less) for mMTC realization, or a U-plane latency (e.g., 0.5 ms or less for downlink (DL) and uplink (UL), or 1 ms or less for round trip) for URLLC realization.

[0125] The antenna module (997) can transmit or receive signals or power to or from an external device (e.g., an external electronic device). In one embodiment, the antenna module (997) may include an antenna including a radiator formed of a conductor or a conductive pattern formed on a substrate (e.g., a PCB). In one embodiment, the antenna module (997) may include a plurality of antennas (e.g., an array antenna). In this case, at least one antenna suitable for a communication method used in a communication network, such as the first network (998) or the second network (999), may be selected from the plurality of antennas, for example, by the communication module (990). A signal or power may be transmitted or received between the communication module (990) and the external electronic device via the selected at least one antenna. In some embodiments, in addition to the radiator, another component (e.g., a radio frequency integrated circuit (RFIC)) may be additionally formed as a part of the antenna module (997).

[0126] According to various embodiments, the antenna module (997) may form a mmWave antenna module. In one embodiment, the mmWave antenna module may include a printed circuit board, an RFIC disposed on or adjacent a first side (e.g., a bottom side) of the printed circuit board and capable of supporting a designated high frequency band (e.g., a mmWave band), and a plurality of antennas (e.g., an array antenna) disposed on or adjacent a second side (e.g., a top side or a side side) of the printed circuit board and capable of transmitting or receiving signals in the designated high frequency band.

[0127] At least some of the above components can be interconnected and exchange signals (e.g., commands or data) with each other via a communication method between peripheral devices (e.g., a bus, GPIO (general purpose input and output), SPI (serial peripheral interface), or MIPI (mobile industry processor interface)).

[0128] According to one embodiment, commands or data may be transmitted or received between the electronic device (901) and an external electronic device (904) via a server (908) connected to a second network (999). Each of the external electronic devices (902 or 904) may be the same or a different type of device as the electronic device (901). According to one embodiment, all or part of the operations executed in the electronic device (901) may be executed in one or more of the external electronic devices (902, 904, or 908). For example, when the electronic device (901) is to perform a certain function or service automatically or in response to a request from a user or another device, the electronic device (901) may, instead of or in addition to executing the function or service itself, request one or more external electronic devices to perform the function or at least part of the service. One or more external electronic devices that receive the request may execute at least a portion of the requested function or service, or an additional function or service related to the request, and transmit the result of the execution to the electronic device (901). The electronic device (901) may process the result as is or additionally and provide it as at least a portion of a response to the request. For this purpose, cloud computing, distributed computing, mobile edge computing (MEC), or client-server computing technology may be used, for example. The electronic device (901) may provide an ultra-low latency service by using distributed computing or mobile edge computing, for example. In another embodiment, the external electronic device (904) may include an Internet of Things (IoT) device. The server (908) may be an intelligent server utilizing machine learning and / or a neural network. According to one embodiment, the external electronic device (904) or the server (908) may be included in the second network (999).The electronic device (901) can be applied to intelligent services (e.g., smart home, smart city, smart car, or healthcare) based on 5G communication technology and IoT-related technology.

[0129]

[0130] According to one embodiment, a method for blocking harmful expressions may include: receiving content; checking context information; checking a harmful level database corresponding to the context information; checking whether harmful information exists in the content based on the harmful level database; determining whether to remove or change the harmful information from the content if the harmful information exists in the content; and removing or changing the harmful information from the content through a harmful processing engine based on the determination.

[0131] According to one embodiment, the harmful processing engine may be implemented using at least one of a rule-based engine, a deep model-based engine, a small large language model (sLLM), and a large language model (LLM).

[0132] According to one embodiment, the method for blocking harmful expressions may further include, before outputting the changed content, an operation of checking whether harmful information exists in the changed content based on the harmful level database; an operation of outputting the changed content if harmful information does not exist in the changed content; and an operation of removing the harmful information from the changed content if harmful information exists in the changed content or changing the harmful information in the changed content through the harmful processing engine until no harmful information exists in the changed content.

[0133] In one embodiment, the contextual information may include at least one of the user's age, current location, set routine, and connected device.

[0134] According to one embodiment, the age of the user may be determined by whether the user is connected to a preset mode or through account information connected to the user, and the user may be classified into a preset age group.

[0135] According to one embodiment, the current location can be identified through GPS information, connected Wi-Fi, and connected devices, and can be classified as a location set by the user.

[0136] According to one embodiment, the connection device is a device connected to an electronic device that provides the harmful expression blocking, and may be a wireless connection device or a wired connection device, and may be classified into a personal device and a public device.

[0137] According to one embodiment, the operation of checking the situation information may include an operation of checking the situation information through a context included in the content.

[0138] According to one embodiment, the operation of checking whether harmful information exists in the content based on the harmful level database may include an operation of checking whether harmful information exists in the content through a harmful search engine based on a harmful blocking level defined for each harmful category included in the harmful level database.

[0139] According to one embodiment, the harmful search engine may be implemented using at least one of a rule-based engine, a deep model-based engine, a small large language model (sLLM), and a large language model (LLM).

[0140] According to one embodiment, the harmful level database can store harmful blocking levels corresponding to each harmful category.

[0141] According to one embodiment, the harmful blocking level may be preset or set by the user.

[0142] According to one embodiment, the harmful blocking level may be comprised of off, low, mid, and high.

[0143] According to one embodiment, the harmful category may include at least one of Illegal, Sexual, Child Abuse, Violent, Derogatory, Toxic, Insult, Profanity, Sensitive, Self-harm, and Generation of malware.

[0144] According to one embodiment, the operation of removing or changing the harmful information from the content through the harmful processing engine according to the above decision may include, if it is determined to remove the harmful information, removing the harmful information from the content and indicating that the harmful information has been removed.

[0145] According to one embodiment, the operation of removing or changing the harmful information from the content through the harmful processing engine according to the above decision may include an operation of changing the harmful information included in the text, image, and voice included in the content using the harmful processing engine according to the corresponding harmful blocking level for each harmful category included in the harmful level database, if it is determined to change the harmful information.

[0146] According to one embodiment, a computer-readable recording medium stores instructions, which, when executed by one or more processors, may perform the following actions: receiving content; checking context information; checking a harmful level database corresponding to the context information; checking whether harmful information exists in the content based on the harmful level database; determining whether to remove or change the harmful information from the content if the harmful information exists in the content; and, based on the determination, removing or changing the harmful information from the content through a harmful processing engine.

[0147] According to one embodiment, a device for blocking harmful expressions includes one or more processors and a memory storing instructions, wherein the instructions, when executed by the one or more processors, cause the device to perform the following operations: receiving content; checking context information; checking a harmful level database corresponding to the context information; checking whether harmful information exists in the content based on the harmful level database; determining whether to remove or change the harmful information from the content if the harmful information exists in the content; and removing or changing the harmful information from the content through a harmful processing engine based on the determination.

[0148] According to one embodiment, the instructions, when executed by the one or more processors, may further cause the device to perform the following operations: checking whether harmful information exists in the changed content based on the harmful level database before outputting the changed content; if harmful information does not exist in the changed content, outputting the changed content; and if harmful information exists in the changed content, removing the harmful information from the changed content or changing the harmful information in the changed content through the harmful processing engine until no harmful information exists in the changed content.

[0149] In one embodiment, the contextual information may include at least one of the user's age, current location, set routine, and connected device.

[0150]

[0151] The method according to the embodiment may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may store program commands, data files, data structures, etc., singly or in combination. The program commands recorded on the medium may be those specially designed and configured for the embodiment or may be those known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program commands, such as ROMs, RAMs, and flash memories. Examples of the program commands include not only machine language codes generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiment, and vice versa.

[0152] Software may include a computer program, code, instructions, or a combination of one or more of these, and may configure a processing device to perform a desired operation or, independently or collectively, command the processing device. The software and / or data may be stored on any type of machine, component, physical device, virtual equipment, computer storage medium, or device, for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over networked computer systems, and stored or executed in a distributed manner. The software and data may be stored on one or more computer-readable recording media.

[0153] Although the embodiments described above have been described with limited drawings, those skilled in the art will appreciate that various technical modifications and variations can be applied based on the above. For example, appropriate results can still be achieved even if the described techniques are performed in a different order than described, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different manner than described, or are replaced or substituted with other components or equivalents.

[0154] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims described below.

Claims

1. In a device that blocks harmful expressions, one or more processors, and Memory that stores commands Including, The above instructions, when executed by the one or more processors, cause the device to: The act of receiving content; Action to check situation information; An action to check the harmful level database corresponding to the above situation information; An action to check whether harmful information exists in the content based on the harmful level database; If the harmful information exists in the content, an operation of determining whether to remove or change the harmful information from the content; and According to the above decision, an action is taken to remove or change the harmful information from the content through the harmful processing engine. A device that enables you to perform a task.

2. In paragraph 1, The above harmful processing engine, Implemented using at least one of a rule-based engine, a deep model-based engine, a small large language model (sLLM), and a large language model (LLM). method.

3. In any one of paragraphs 1 and 2, The above instructions, when executed by the one or more processors, cause the device to: Before outputting the changed content, an action of checking whether harmful information exists in the changed content based on the harmful level database; If there is no harmful information in the changed content, an action of outputting the changed content; and If harmful information exists in the changed content, an action is taken to remove the harmful information from the changed content or change the harmful information in the changed content through the harmful processing engine until no harmful information exists in the changed content. A device that allows you to perform more.

4. In any one of paragraphs 1 to 3, The above situation information is, Including at least one of the user's age, current location, set routines, and connected devices. device.

5. In any one of paragraphs 1 to 4, The age of the above user is Whether you are connected to a preset mode or through the account information of the user connected above, and are classified into a preset age group device.

6. In any one of paragraphs 1 to 5, The current location above is, It is identified through GPS information, connected Wi-Fi, and connected devices, and is classified as the location set by the user. device.

7. In any one of paragraphs 1 to 6, The above connecting device, A device connected to an electronic device that provides the above harmful expression blocking, It can be a wireless connection device or a wired connection device, Classified into personal devices and public devices device.

8. In any one of paragraphs 1 to 7, The action to check the above situation information is: An action to check the above situation information through the context included in the above content. A device comprising:

9. In any one of paragraphs 1 to 8, The operation of checking whether harmful information exists in the content based on the above harmful level database is as follows: An action to check whether harmful information exists in the content through a harmful search engine according to the harmful blocking level defined for each harmful category included in the above harmful level database. A device comprising:

10. In any one of paragraphs 1 to 9, The above harmful level database is, Stores the corresponding harmful blocking level for each harmful category above. method.

11. In any one of paragraphs 1 to 10, The above harmful blocking level is, Either preset or set by the user, Consists of off, low, mid, and high device.

12. In any one of paragraphs 1 to 11, The above hazardous categories are: Contains at least one of the following: Illegal, Sexual, Child Abuse, Violent, Derogatory, Toxic, Insult, Profanity, Sensitive, Self-harm, Generation of malware method.

13. In any one of paragraphs 1 to 12, According to the above decision, the action of removing or changing the harmful information from the content through the harmful processing engine is: If it is decided to remove the above harmful information, an action is taken to remove the above harmful information from the above content and indicate that the above harmful information has been removed. How to include.

14. In any one of paragraphs 1 to 13, According to the above decision, the action of removing or changing the harmful information from the content through the harmful processing engine is: If it is decided to change the above harmful information, an action is taken to change the harmful information contained in the text, images, and voices included in the content using the harmful processing engine according to the corresponding harmful blocking level for each harmful category included in the harmful level database. How to include.

15. In the method of blocking harmful expressions, The act of receiving content; Action to check situation information; An action to check the harmful level database corresponding to the above situation information; An action to check whether harmful information exists in the content based on the harmful level database; If the harmful information exists in the content, an operation of determining whether to remove or change the harmful information from the content; and According to the above decision, an action is taken to remove or change the harmful information from the content through the harmful processing engine. How to include.

Images