Electronic device and operation method thereof

Abstract

The present disclosure relates to an artificial intelligence (AI) system utilizing a machine learning algorithm, and an application thereof. An electronic device according to the present disclosure comprises: a memory for storing one or more instructions; and one or more processors for executing the one or more instructions stored in the memory, wherein the one or more processors can: transmit, to a server, request information including text information about an input conversation type acquired from situation information and / or metadata corresponding to content; and receive, from the server, a recommendation result acquired on the basis of the request information.

Description

Electronic device and method of operation thereof

[0001] The disclosed various embodiments relate to electronic devices and their operating methods, and more specifically, to electronic devices combined with artificial intelligence technology and operating methods using the electronic devices.

[0002] Hyper-large-scale AI is an AI model trained on large-scale data. Hyper-large-scale AI can process and analyze large amounts of data, demonstrating high accuracy and performance.

[0003] The emergence of hyper-large-scale artificial intelligence has led to a dramatic leap in the performance of generative AI. Unlike existing AI systems designed to recognize and predict patterns, generative AI is an AI algorithm that generates new data based on existing data.

[0004] An electronic device according to an embodiment may include a memory storing one or more instructions and one or more processors executing the one or more instructions stored in the memory.

[0005] In an embodiment, the one or more processors may transmit request information to the server, the request information including text information in the form of an input dialogue obtained from at least one of contextual information and metadata corresponding to the content.

[0006] In an embodiment, the one or more processors may receive a recommendation result based on the request information from the server.

[0007] FIG. 1 is a diagram illustrating an electronic device transmitting request information to a server and receiving a recommendation result from the server, according to an embodiment.

[0008] Figure 2 is a block diagram of an electronic device according to an embodiment.

[0009] Figure 3 illustrates a block diagram of a processor according to an embodiment.

[0010] Fig. 4 is a block diagram of a situation information acquisition unit according to an embodiment.

[0011] FIG. 5 illustrates an electronic device outputting a recommendation result received from a server as a video signal, according to an embodiment.

[0012] FIG. 6 illustrates an example in which an electronic device outputs a recommendation result received from a server as an audio signal.

[0013] FIG. 7 illustrates an example in which an electronic device outputs a recommendation result received from a server as a video signal.

[0014] FIG. 8 illustrates an example in which an electronic device outputs a recommendation result received from a server as a video signal.

[0015] Figure 9 is a block diagram of an electronic device according to an embodiment.

[0016] Figure 10 is a block diagram of a server according to an embodiment.

[0017] Figure 11 is a block diagram of a server according to an embodiment.

[0018] Figure 12 illustrates that, according to an embodiment, when a server obtains a recommendation result, it also obtains a recommendation reason.

[0019] FIG. 13 is a diagram illustrating how a server generates recommendation results in a consistent manner, according to an embodiment.

[0020] FIG. 14 is a diagram illustrating how a server generates recommendation results in a consistent manner, according to an embodiment.

[0021] FIG. 15 is a diagram showing, according to an embodiment, a server obtaining recommendation results using multiple theme-specific neural networks.

[0022] FIG. 16 illustrates a recommendation result output by an electronic device according to an embodiment.

[0023] FIG. 17 is a flowchart illustrating an operation method of an electronic device and a server according to an embodiment.

[0024] FIG. 18 is a flowchart illustrating a method in which an electronic device and a server operate according to an embodiment.

[0025] A server according to an embodiment may include a memory storing one or more instructions and one or more processors executing the one or more instructions stored in the memory.

[0026] In an embodiment, when request information including text information in the form of an input dialogue is received from an electronic device, the one or more processors may transmit a recommendation result including a recommendation reason corresponding to the request information, obtained based on one or more neural networks, to the electronic device.

[0027] A method of operating an electronic device according to an embodiment may include transmitting request information including text information in the form of an input dialogue, obtained from at least one of contextual information and metadata corresponding to content, to a server.

[0028] In an embodiment, the method of operating an electronic device may include receiving a recommendation result based on the request information from the server.

[0029] A method of operating a server according to an embodiment may include, when request information including text information in the form of an input dialogue is received from an electronic device, transmitting a recommendation result including a recommendation reason corresponding to the request information, obtained based on one or more neural networks, to the electronic device.

[0030] In this disclosure, the expression “at least one of a, b or c” may refer to “a”, “b”, “c”, “a and b”, “a and c”, “b and c”, “all of a, b and c”, or variations thereof.

[0031] Below, embodiments of the present disclosure are described in detail with reference to the attached drawings so that those skilled in the art can easily implement the present disclosure. However, the present disclosure may be implemented in various different forms and is not limited to the embodiments described herein.

[0032] The terms used in this disclosure are described as currently common terms, taking into account the functions mentioned herein. However, these terms may mean various other terms depending on the intentions of those skilled in the art, precedents, the emergence of new technologies, etc. Therefore, the terms used in this disclosure should not be interpreted solely based on their names, but rather based on the meanings of the terms and the overall content of this disclosure.

[0033] Additionally, the terminology used in this disclosure is for the purpose of describing specific embodiments only and is not intended to limit the present disclosure.

[0034] Throughout the specification, when a part is said to be "connected" to another part, this includes not only the cases where it is "directly connected" but also the cases where it is "electrically connected" with another element in between.

[0035] As used herein, and particularly in the claims, the terms "above" and "above" and similar referents may refer to both the singular and the plural. Furthermore, unless the order of steps in a method according to the present disclosure is explicitly specified, the steps described may be performed in any appropriate order. The present disclosure is not limited by the order in which the steps are described.

[0036] The appearances of phrases such as “in some embodiments” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment.

[0037] Some embodiments of the present disclosure may be represented by functional block configurations and various processing steps. Some or all of these functional blocks may be implemented by various hardware and / or software configurations that perform specific functions. For example, the functional blocks of the present disclosure may be implemented by one or more microprocessors or by circuit configurations for a given function. Furthermore, for example, the functional blocks of the present disclosure may be implemented in various programming or scripting languages. The functional blocks may be implemented by algorithms that execute on one or more processors. Furthermore, the present disclosure may employ conventional techniques for electronic configuration, signal processing, and / or data processing. Terms such as “mechanism,” “element,” “means,” and “configuration” may be used broadly and are not limited to mechanical and physical configurations.

[0038] Additionally, the connecting lines or connecting members between components depicted in the drawings are merely exemplary representations of functional connections and / or physical or circuit connections. In an actual device, connections between components may be represented by various functional connections, physical connections, or circuit connections that may be replaced or added.

[0039] Additionally, terms such as “part”, “module”, etc. described in the specification mean a unit that processes at least one function or operation, which may be implemented as hardware or software, or a combination of hardware and software.

[0040] Additionally, the term “user” in the specification means a person who uses an electronic device or server, and may include a consumer, evaluator, viewer, administrator, or installer.

[0041] The present disclosure will be described in detail with reference to the attached drawings below.

[0042] The present disclosure may relate to artificial intelligence (AI) systems and / or applications of AI systems utilizing machine learning algorithms.

[0043] FIG. 1 is a diagram illustrating an electronic device (100) transmitting request information to a server (200) and receiving a recommendation result from the server (200), according to an embodiment.

[0044] Referring to FIG. 1, the electronic device (100) may be an electronic device capable of outputting images. For example, the electronic device (100) may be implemented as various types of electronic devices including a display. The electronic device (100) may be fixed or mobile, and may be a digital TV capable of receiving digital broadcasts, but is not limited thereto.

[0045] The electronic device (100) can output various types of content provided by content providers. Content may include video, such as still images and moving images, audio, subtitles, and other additional information. A content provider may refer to a terrestrial broadcaster, cable broadcaster, satellite broadcaster, or IPTV (Internet Protocol Television) service provider or OTT (Over the Top) service provider that provides various types of content to consumers.

[0046] In an embodiment, the electronic device (100) can receive and output various contents generated by a content provider through an external device (not shown). For example, the external device can be implemented as a variety of source devices such as a PC, a set-top box, a Blu-ray disc player, a mobile phone, a game console, a home theater, an audio player, a USB, etc. The external device can be connected to the electronic device (100) through a wired communication network such as HDMI or a wireless communication network and provide various contents to the electronic device (100). The electronic device (100) can receive and output VOD (Video On Demand) contents provided by an IPTV service provider or an OTT service provider through a set-top box. The VOD service is a service that provides a desired video to a user at a desired time through a communication network connection, and can refer to various forms of contents provided by an OTT service provider or an IPTV service provider. An IPTV service provider or an OTT service provider can provide not only VOD contents but also real-time broadcasting programs.

[0047] In an embodiment, the electronic device (100) may be a smart TV. A smart TV may refer to a digital TV equipped with an operating system (OS) and internet access capabilities. A smart TV may also be referred to as an internet TV, a connected TV, or a hybrid TV.

[0048] In an embodiment, the electronic device (100) can stream and output various types of VOD content created by OTT service providers such as YouTube™ or Netflix™ using an operating system installed therein in addition to real-time broadcast programs.

[0049] In an embodiment, the electronic device (100) can connect to the Internet and provide web surfing services, social network services, etc. In addition, the electronic device (100) can function as a communication center that can check news, weather, email, etc. in real time.

[0050] In an embodiment, the electronic device (100) can execute various types of applications. The electronic device (100) may have various types of apps (applications) installed by default. Alternatively, the electronic device (100) may, under user control, access the Internet, search for apps requested by the user, and install them. The electronic device (100) can execute apps to provide various services.

[0051] In an embodiment, the electronic device (100) can wirelessly connect to an external device, such as a set-top box, through a wireless network that follows a communication standard, such as Bluetooth, WLAN (Wireless LAN) (Wi-Fi), Wibro (Wireless broadband), Wimax (World Interoperability for Microwave Access), CDMA, or WCDMA, and receive a video signal from the external device.

[0052] In an embodiment, the electronic device (100) may be connected to an external device via a wired cable to receive a video signal from the external device or transmit a video signal to the external device. The wired cable (140) may include a port capable of simultaneously transmitting a video signal and an audio signal, such as a High-Definition Multimedia Interface (HDMI), a Universal Serial Bus (USB), a Display Port (DP), or Thunderbolt™. Alternatively, the wired cable may include a port for separately transmitting a video signal and an audio signal.

[0053] In an embodiment, the electronic device (100) may be implemented as an electronic device that does not include a video display device such as a display. For example, if the electronic device (100) is an external device itself, such as a set-top box, a satellite broadcast receiving device, an Internet receiving device that receives content from an OTT (Over The Top) service provider, the electronic device (100) may be of a form that does not include a video display device.

[0054] In this case, the electronic device (100) can be connected to the video display device via a wire, and transmit a signal input from an external source to the video display device. For example, the electronic device (100) can be connected to the video display device using a port that can simultaneously transmit video signals and audio signals, such as a Universal Serial Bus (USB), a High-Definition Multimedia Interface (HDMI), a Display Port (DP), Thunderbolt™, etc. Alternatively, the electronic device (100) can be connected to the video display device using a port that separately transmits video signals and audio signals.

[0055] In an embodiment, the electronic device (100) may be controlled by a control device (50). In an embodiment, the control device (50) may be a device used to control the electronic device (100), such as a remote controller. A user may control various functions of the electronic device (100) using the control device (50).

[0056] In an embodiment, the control device (50) may be a dedicated control device for the electronic device (100). Alternatively, in an embodiment, the control device (500) may be an electronic device that performs other operations as its main function in addition to the operation of controlling the electronic device (100), such as a smartphone or an AI speaker. In this case, a user may control the electronic device (100) using the control device (50) by installing a remote control app or the like on the control device (50). In this case, the control device (50) may include a Wi-Fi, Bluetooth, or infrared communication module. The control device (50) may transmit and receive data with the electronic device (100) using the Wi-Fi, Bluetooth, or infrared communication module or the like.

[0057] In an embodiment, the control device (50) may have an input unit. The input unit may receive user input for controlling the electronic device (100). The input unit included in the control device (50) may include a plurality of keys. The keys may be in various forms, such as physical buttons that receive a user's push operation, a jog shuttle, a touch button displayed on a touchpad that detects touch, etc. The user may control various functions of the electronic device (100) using the plurality of keys provided in the control device (50).

[0058] A plurality of keys included in the control device (50) can be used to control various functions of the electronic device (100).

[0059] However, the present invention is not limited thereto, and in an embodiment, at least one of the electronic device (100) or the control device (50) may include a microphone capable of receiving a user's voice. If the control device (50) is equipped with a microphone, the microphone may receive the user's analog voice signal, digitize it, and transmit it to the electronic device (100). In an embodiment, the control device (50) may receive the user's voice signal through the microphone, digitize it, and transmit the digitized voice signal to the electronic device (100) using a data transmission communication method such as Bluetooth or Wi-Fi. In an embodiment, if the electronic device (100) is equipped with a microphone, the microphone may collect the user's voice signal, digitize it, and transmit it to the processor.

[0060] In an embodiment, if the server (200) can perform a voice recognition operation, the electronic device (100) can transmit a voice signal to the server (200). The server (200) can perform a STT (Speech-to-Text) operation that converts the voice signal received from the electronic device (100) into text.

[0061] Alternatively, if both the electronic device (100) and the server (200) are unable to perform the STT operation, the electronic device (100) may transmit the voice signal to an STT server that performs an operation of converting the voice signal into text, receive the text from the STT server, and transmit it to the server (200). Alternatively, the STT server may transmit the text directly to the server (200).

[0062] In an embodiment, the electronic device (400) can execute a voice recognition function by performing the functions of a voice recognition engine without interfacing with a separate server. For example, in an embodiment, if the electronic device (100) can provide a voice recognition service in an on-device form, the electronic device (400) can perform an operation of interpreting spoken language and converting its content into text data. In an embodiment, if the electronic device (100) can perform an STT operation on a voice signal, the electronic device (100) can convert the voice signal into text and transmit the converted text to the server (200).

[0063] In an embodiment, the electronic device (100) may include a camera or motion sensor that can recognize a user's gesture as input.

[0064] In the past, remote controls were of the type that exposed all the various types of keys used to control electronic devices (100) on the front. However, recently, simple remote controls have been developed and are being used, minimizing the number of rarely used buttons and including only buttons for functions frequently used by users. For example, remote controls may include a small number of keys, such as channel up or down keys, volume up or down keys, four-way keys, a key to move to the main screen, and a search key. Accordingly, users operate electronic devices (100) using a small number of buttons included in the remote control, i.e., the control device (50).

[0065] Systems utilizing generative AI technology are primarily implemented for use on the web or in apps. Therefore, it is difficult for users to experience generative AI technology using electronic devices (100), such as televisions, controlled by control devices (50) such as remote controls. In other words, in order to utilize a generative AI system, the electronic device (100) must transmit conversational questions to the generative AI. However, if the remote control only includes a small number of keys, it is difficult for users to input conversational questions into the electronic device (100) using the remote control.

[0066] The present disclosure is intended to solve such problems, and an electronic device (100) according to an embodiment can automatically generate requested information desired by a user based on metadata corresponding to contextual information or content, and transmit the same to a generating artificial intelligence model to request a recommendation result, without the user having to directly input a conversation using a control device (50).

[0067] In an embodiment, the electronic device (100) may obtain contextual information. In an embodiment, the contextual information may be information indicating the context of the electronic device (100) itself, the context of a user using the electronic device (100), or information indicating the surrounding context of the electronic device (100). For example, the contextual information may include at least one of contextual information of content currently output by the electronic device (100), user information, characteristic information, and surrounding contextual information.

[0068] In an embodiment, the electronic device (100) may obtain metadata corresponding to the content. The metadata corresponding to the content is structured data regarding the content, and may be data that describes the properties of the content or the content.

[0069] In an embodiment, metadata corresponding to content may include at least one of metadata corresponding to content currently being output by the electronic device (100), metadata corresponding to content that the electronic device (100) can output, and composition information.

[0070] In an embodiment, the electronic device (100) can generate request information from at least one of contextual information and metadata corresponding to the content.

[0071] In an embodiment, the request information may include information that the electronic device (100) requests from the server (200), requesting a recommendation for content or a channel, etc. In an embodiment, the request information may include text information in the form of an input dialogue.

[0072] In an embodiment, the electronic device (100) may generate request information based on receiving a specific control signal from the control device (50). For example, the electronic device (100) may generate request information when, under user control, a control signal corresponding to the execution of one of a search function, a channel change function, and a function of moving to a main screen is received from the control device (50).

[0073] For example, if there is content that the user wants to watch, the user can use the control device (50) to request the electronic device (100) to execute a search function, or request a channel change to watch the desired channel. Alternatively, if the content that the user wants to watch is content that can be received and output from an external device other than the external device currently connected to the electronic device (100), or content that is available through another web or app, the user can control the electronic device (100) to move to the main screen or home screen by selecting a button for moving to the main screen provided on the control device (50), and to select another web, another app, or another external device from the home screen.

[0074] In an embodiment, when a user requests to execute one of the functions of a search function, a channel change function, or a function of moving to the main screen by using a control device (50), the electronic device (100) may generate request information requesting a content search based on information about content that the user has previously viewed, channels that the user frequently views, or content that has recently become popular with the public.

[0075] In an embodiment, the electronic device (100) may continuously, or at regular time intervals, acquire context information of content that the electronic device (100) is currently displaying on the screen. If the electronic device (100) does not receive a specific control signal from the control device (50), the electronic device (100) may repeat the process of discarding context information of previously acquired content.

[0076] In an embodiment, when the electronic device (100) receives a specific control signal from the control device (50), it may generate request information based on context information acquired during a predetermined period of time prior to receiving the specific control signal. For example, when the electronic device (100) receives a control signal requesting execution of a channel change function from the control device (50) at a time point t, the electronic device (100) may generate request information based on context information acquired from a time point t-30 seconds.

[0077] In an embodiment, the electronic device (100) may transmit request information to the server (200) based on receiving a specific control signal from the control device (50).

[0078] In an embodiment, the electronic device (100) may be connected to the server (200) via a wired or wireless communication network.

[0079] In an embodiment, the server (200) may be a server that provides recommendation results using a super-large generative artificial intelligence model.

[0080] In an embodiment, the server (200) may be a server utilizing one or more neural networks. In an embodiment, the neural network utilized by the server (200) may be a large-scale conversational generative artificial intelligence model.

[0081] In an embodiment, the server (200) may receive request information in the form of a conversation from the electronic device (100) and input it into a neural network, i.e., an artificial intelligence model, to obtain a result. In an embodiment, the recommendation result obtained by the server (200) using the artificial intelligence model may be information in the form of an output conversation.

[0082] In an embodiment, the server (200) may transmit recommendation results obtained from an artificial intelligence model to an electronic device (100). In an embodiment, the server (200) may obtain recommendation reasons along with the recommendation results using the artificial intelligence model. In an embodiment, the server (200) may transmit the recommendation results and recommendation reasons to the electronic device (100), thereby allowing a user to utilize the recommendation results and recommendation reasons together.

[0083] In an embodiment, the server (200) may obtain additional information. In an embodiment, the additional information may be information collected by the server (200) to provide more accurate recommendation results to the electronic device (100).

[0084] In an embodiment, the additional information may include at least one of metadata corresponding to the content, popular content, curated content, content selected based on a specific theme, and Key Performance Index (KPI) information.

[0085] In an embodiment, the server (200) may obtain user information. User information is information generated according to the account of a specific user or a specific electronic device (100), and may include setting information of a specific user or a user of a specific electronic device (100), user viewing history information, i.e., user channel or content usage history information, channel switching history information, user profile information, etc.

[0086] In an embodiment, user information may be stored and utilized in a user information database. Additionally, user information may be updated based on information provided by users.

[0087] In an embodiment, the server (200) may input at least one of additional information and user information together with recommendation request information into one or more neural networks, thereby obtaining recommendation results from one or more neural networks.

[0088] In an embodiment, the server (200) may input candidate information that may serve as a reason for recommendation into one or more neural networks. Candidate information that may serve as a reason for recommendation may include user preferences, user viewing history, user settings, and information on popular content or programs.

[0089] In an embodiment, one or more neural networks utilized by the server (200) may include multiple theme-specific neural networks. The theme-specific neural networks may refer to artificial intelligence models trained with learning data of different themes.

[0090] In an embodiment, the server (200) may input request information received from the electronic device (100) into a plurality of theme-specific neural networks. In an embodiment, the server (200) may obtain a recommendation result from at least one of the plurality of theme-specific neural networks.

[0091] In an embodiment, the server (200) may generate a recommendation result including at least one of information corresponding to an executable action, an outputtable channel or content, and a recommendation reason. The server (200) may transmit the recommendation result to the electronic device (100).

[0092] In an embodiment, the electronic device (100) may receive a recommendation result from the server (200) and output at least one of an audio signal and a video signal corresponding to the recommendation result. In an embodiment, the video signal corresponding to the recommendation result may include text information in the form of a dialogue. In an embodiment, the electronic device (100) may output the recommendation result on a multi-view screen. The multi-view screen may be a screen that includes channels or content according to the recommendation result received from the server (200) in the form of images or metadata.

[0093] However, the present invention is not limited thereto, and in an embodiment, the electronic device (100) may be in a form in which the above-mentioned artificial intelligence model is combined. That is, the electronic device (100) may be in a form including a generative AI model. The electronic device (100) may also directly obtain a recommendation result from the request information using on-device AI technology. That is, the electronic device (100) may quickly obtain a recommendation result by collecting and calculating information on its own without going through a cloud server. The electronic device (100) may input at least one of additional information and user information together with the request information into one or more neural networks, thereby directly obtaining a recommendation result from one or more neural networks.

[0094] In this way, according to an embodiment, when a user controls an electronic device (100) using a control device with limited input, such as a remote control, the electronic device (100) can automatically generate request information in the form of an input dialogue without the user having to individually input the surrounding circumstances or the user's preferences. The electronic device (100) can output recommendation results obtained based on the request information, thereby allowing the user to have an experience similar to having a conversation with the generated artificial intelligence model.

[0095] Figure 2 is a block diagram of an electronic device (100) according to an embodiment.

[0096] The electronic device (100) of FIG. 2 may be an example of the electronic device (100) of FIG. 1.

[0097] In an embodiment, the electronic device (100) may be an electronic device capable of outputting images. For example, the electronic device (100) may be implemented as various types of electronic devices including a display. The electronic device (100) may be fixed or mobile, and may be a digital TV capable of receiving digital broadcasts, but is not limited thereto.

[0098] In an embodiment, the electronic device (100) may include at least one of a desktop, a smartphone, a tablet personal computer (PC), a mobile phone, a video phone, an e-book reader, a laptop personal computer (PC), a netbook computer, a digital camera, a personal digital assistant (PDA), a portable multimedia player (PMP), a camcorder, a navigation device, a wearable device, a smart watch, a home network system, a security system, and a medical device.

[0099] The electronic device (100) may be implemented as a flat display device, a curved display device having a screen with a curvature, or a flexible display device whose curvature can be adjusted. The output resolution of the electronic device (100) may have various resolutions, such as, for example, HD (High Definition), Full HD, Ultra HD, or a resolution clearer than Ultra HD.

[0100] In an embodiment, the electronic device (100) may be implemented as an electronic device that does not include a display. In this case, the electronic device (100) may directly receive various types of content provided by a content provider, provide the content to an external display or external speaker connected to the electronic device (100), and output the content through the external display or external speaker.

[0101] Referring to FIG. 2, the electronic device (100) may include a processor (110) and a memory (120).

[0102] The memory (120) according to the embodiment can store at least one instruction. The memory (120) can store at least one program executed by the processor (110). In addition, the memory (120) can store data input to or output from the electronic device (100).

[0103] The memory (120) may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), a RAM (Random Access Memory), a SRAM (Static Random Access Memory), a ROM (Read-Only Memory), an EEPROM (Electrically Erasable Programmable Read-Only Memory), a PROM (Programmable Read-Only Memory), a magnetic memory, a magnetic disk, and an optical disk.

[0104] In an embodiment, the memory (120) may store one or more instructions for obtaining at least one of context information and metadata corresponding to the content.

[0105] In an embodiment, the memory (120) may store one or more instructions for generating request information from at least one of contextual information and metadata corresponding to the content. In an embodiment, the request information may include text information in the form of an input dialogue.

[0106] In an embodiment, the memory (120) may store one or more instructions for transmitting request information to the server (200) and receiving a recommendation result based on the request information from the server (200).

[0107] In an embodiment, the memory (120) may store one or more instructions for transmitting request information to the server (200) based on receiving a control signal from the control device (50).

[0108] In an embodiment, the memory (120) may store one or more instructions for transmitting request information to the server (200) when a control signal corresponding to the execution of one of a search function, a channel change function, and a move to the main screen function is received.

[0109] In an embodiment, the memory (120) may store one or more instructions for outputting at least one of an audio signal and a video signal corresponding to a recommendation result received from the server (200).

[0110] In an embodiment, the memory (120) may store one or more instructions for outputting a video signal corresponding to a recommendation result received from the server (200) as text information in a conversational form.

[0111] In an embodiment, the memory (120) may store one or more instructions for outputting the recommendation results to a multi-view screen.

[0112] In an embodiment, the memory (120) may store one or more instructions for forming a multi-view screen. For example, the memory (120) may store multi-view screens configured differently depending on the number of partial screens.

[0113] In an embodiment, the memory (120) may store one or more instructions for generating a screen including a channel or content as an image or metadata according to a recommendation result received from the server (200) in the configuration of a pre-stored multi-view screen.

[0114] In an embodiment, the memory (120) may store one or more instructions for requesting screen information to be output when a first area among a plurality of partial screens included in a multi-view screen is selected from the server (200) before a user input for selecting the first area is received.

[0115] In an embodiment, the memory (120) may store one or more instructions for outputting a screen based on a recommendation result corresponding to screen information requested and received in advance when a user input for selecting a first area is received.

[0116] In an embodiment, at least one AI model (neural network model) may be stored in the memory (120).

[0117] In an embodiment, when a neural network model is stored in the memory (120), the neural network model can directly generate a multi-view screen including channels or content according to the recommendation results as images or metadata.

[0118] In an embodiment, when a neural network model is stored in the memory (120), the neural network model can perform an operation of generating screen information to be output when a first area among a plurality of partial screens included in a multi-view screen is selected in advance before a control signal for selecting the first area, i.e., a user input for selecting the first area, is received, and temporarily storing the same.

[0119] In an embodiment, user information may be stored in the memory (120). In an embodiment, the user information may be information stored to continue a user-customized conversation, and may include at least one of a short-term memory for storing short-term user information and a long-term memory for storing long-term user information.

[0120] In an embodiment, the short-term memory may store short-term history information regarding the user's use of the electronic device (100) from the time the user turns on the electronic device (100). For example, the short-term memory may store information regarding programs or channels viewed by the user, metadata regarding the information regarding the programs or channels viewed by the user, information regarding the source device or OTT service provider that provided the program or channel, etc. The short-term history information stored in the short-term memory stores information from the time the electronic device (100) is turned on, and may be deleted when the electronic device (100) is turned off.

[0121] In an embodiment, the long-term memory may store long-term history information regarding the user's use of the electronic device (100) from the time the user resets the electronic device (100). The history information stored in the long-term memory may be continuously stored regardless of whether the electronic device (100) is powered on or off, and may be deleted when the electronic device (100) is reset.

[0122] A processor (110) according to an embodiment controls the overall operation of an electronic device (100). The processor (110) can control the electronic device (100) to function by executing one or more instructions stored in a memory (120). In an embodiment, there may be one or more processors (110).

[0123] In an embodiment, one or more processors (110) may acquire context information by executing one or more instructions. In an embodiment, the context information may include at least one of context information, characteristic information, user information, and surrounding context information of content currently output by the electronic device (100).

[0124] In an embodiment, one or more processors (110) may acquire metadata corresponding to content by executing one or more instructions. In an embodiment, the metadata corresponding to content may include at least one of metadata corresponding to content currently output by the electronic device (100), metadata corresponding to content that the electronic device (100) can output, and programming information. In an embodiment, one or more processors (110) may acquire a programming table such as an EPG, metadata corresponding to content, etc. from a broadcasting station server, an OTT service provider server, etc.

[0125] In an embodiment, one or more processors (110) may generate request information including text information in the form of an input dialogue from at least one of contextual information and metadata corresponding to content by executing one or more instructions.

[0126] In an embodiment, one or more processors (110) may transmit request information to a server (200) and receive a recommendation result obtained based on the request information from the server (200) by executing one or more instructions.

[0127] In an embodiment, the recommendation result received from the server (200) may include at least one of information about an action that the electronic device (100) can perform and information about a channel or content that the electronic device (100) can output.

[0128] In an embodiment, one or more processors (110) may output at least one of an audio signal and a video signal corresponding to a recommendation result received from a server (200) by executing one or more instructions. In an embodiment, the video signal corresponding to the recommendation result may include text information in the form of a conversation. In an embodiment, the video signal corresponding to the recommendation result may be output through a multi-view screen. In an embodiment, one or more processors (110) may output a channel or content according to the recommendation result received from the server (200) through a partial screen included in the multi-view screen.

[0129] In an embodiment, when one or more processors (110) receive and output a multi-view screen as a recommendation result from the server (200), the electronic device (100) may request screen information of the next screen to be output when any one of a plurality of partial screens included in the multi-view screen is selected, from the server (200) in advance before a control signal for selecting the partial screen is received. For example, when the multi-view screen output by the electronic device (100) includes three areas, the user may select any one of the three areas included in the multi-view screen output by the electronic device (100).

[0130] In an embodiment, one or more processors (110) may request screen information about a plurality of partial screens included in a multi-view screen, i.e., a next screen to be output when a first area among the plurality of areas is selected, from the server (200) in advance, by executing one or more instructions, before a control signal for selecting the first area is received.

[0131] When the first area is selected, the next screen information to be output may be information including information to be output through a screen other than the first area when the first area is selected and moves to the center screen of the multi-view screen.

[0132] When the electronic device (100) requests the server (200) for next screen information after the user selects the first area, the server (200) must search for metadata about the content or program to be included in the next screen and transmit the search results back to the electronic device (100) in the form of thumbnails or metadata. Since a certain amount of time is required for the process in which the electronic device (100) and the server (200) transmit and receive information and for the server (200) to search for content or programs to be included in the next screen, that is, the new screen to be output through multiple areas after the first area moves to the center screen, there is a problem in that the user must wait for a certain amount of time or more from the time the first area is selected until the next screen is output.

[0133] In an embodiment, in order to prevent such time consumption, one or more processors (110) may request the server (200) in advance for next screen information to be output when each of a plurality of areas included in the multi-view screen is selected, before a control signal for selecting the area is received. In an embodiment, one or more processors (110) may receive in advance metadata about content or a program to be included in the next screen from the server (200) and temporarily store the same. In an embodiment, when a user selects one of a plurality of partial screens, one or more processors (110) may immediately output the next screen based on the recommendation result corresponding to the screen information based on the screen information received and stored in advance from the server (200), thereby increasing user convenience.

[0134] FIG. 3 illustrates a block diagram of a processor (110) according to an embodiment.

[0135] Referring to FIG. 3, the processor (110) of FIG. 3 may be an example of the processor (110) included in the electronic device (100) illustrated in FIG. 2.

[0136] In an embodiment, the processor (110) may include a situation information acquisition unit (111), a content metadata acquisition unit (113), and a request information generation unit (115).

[0137] In an embodiment, each component included in the processor (110) may be a module. In an embodiment, a module may refer to a functional and structural combination of hardware for implementing the technical concepts of the present disclosure and software for operating the hardware. For example, a module may refer to a logical unit of a given code and hardware resources for executing the given code, and is not necessarily limited to physically connected code or a single type of hardware.

[0138] In an embodiment, the situation information acquisition unit (111) can acquire situation information to generate request information.

[0139] In an embodiment, the context information may include at least one of information indicating the context of the electronic device (100), the context of the user, or the context around the electronic device (100).

[0140] In an embodiment, the context information may include context information of the electronic device (100). The context information of the electronic device (100) may include at least one of context information and characteristic information regarding content currently output by the electronic device (100).

[0141] In an embodiment, the context information acquisition unit (111) may acquire context information from content currently being output by the electronic device (100). In an embodiment, the context information may include various information that may be acquired about the content currently being output by the electronic device (100) through the screen. For example, the context information may include at least one of the title of the content or program, the channel name or number, subtitles, and logo or text information included in the screen, acquired from the screen being output.

[0142] In an embodiment, the situation information acquisition unit (111) may acquire characteristic information as situation information of the electronic device (100). In an embodiment, the characteristic information may include information related to unique characteristics of the electronic device (100), such as information on the resolution supported by the electronic device (100), information on whether the electronic device (100) can be rotated between a horizontal screen and a vertical screen, and information on whether the electronic device (100) can execute an ambient service.

[0143] In an embodiment, the situation information acquisition unit (111) can acquire the user's situation information. In an embodiment, the situation information acquisition unit (111) can acquire the user's situation information by reading it from a short-term memory or a long-term memory included in the memory (120).

[0144] In an embodiment, the user's context information may include user information. The user information may include various information about the user known to the electronic device (100), such as user profile information, user viewing history information, user setting information, and the user's OTT subscription information. In an embodiment, the user's viewing history information may include information about channels or content frequently viewed by the user, information about channels or content set as favorites, and the like.

[0145] In an embodiment, the context information acquisition unit (111) may acquire context information surrounding the electronic device (100). In an embodiment, the context information surrounding the electronic device (100) may be information that is unrelated to the electronic device (100) itself but represents a context necessary for generating request information. For example, the context information surrounding the electronic device (100) may include the current time or location information of the electronic device (100).

[0146] In an embodiment, the situation information acquisition unit (111) may acquire at least one of situation information of the electronic device (100), situation information of the user, and situation information around the electronic device (100), and transmit the same to the request information generation unit (115).

[0147] In an embodiment, the content metadata acquisition unit (113) may acquire metadata corresponding to the content. In an embodiment, the metadata corresponding to the content may be data that describes the properties or content of the content, or may be information indicating additional information about the content. The metadata corresponding to the content may be used to identify, analyze, and / or classify the content.

[0148] In an embodiment, the content metadata acquisition unit (113) can acquire at least one of metadata corresponding to content currently being output by the electronic device (100), metadata corresponding to content that the electronic device (100) is not currently outputting but can output, and program information.

[0149] In an embodiment, the content metadata acquisition unit (113) may acquire metadata corresponding to the content from a content provider, a broadcasting station server, an OTT service provider server, etc. In an embodiment, the content metadata acquisition unit (113) may also acquire the title or subtitle of the broadcast included in the broadcast stream as metadata from a digital broadcast stream.

[0150] In an embodiment, the content metadata acquisition unit (113) may acquire EPG (Electronic Program Guide) information including all currently viewable channel information and information that can be viewed shortly thereafter, or information indicating properties or categories of the content, such as the title, performers, series, production date, production company, cameos, and tags of the content. EPG may refer to a broadcast program guide service including broadcast time and content of a broadcast program, performer information, etc.

[0151] In an embodiment, the content metadata acquisition unit (113) can acquire various metadata for the content and transmit them to the request information generation unit (115).

[0152] In an embodiment, the request information generation unit (115) may be a module that creates an input dialogue for conversation with an artificial intelligence server. In an embodiment, the request information generation unit (115) may generate request information from at least one of the context information received from the context information acquisition unit (111) and the metadata received from the content metadata acquisition unit (113).

[0153] In an embodiment, the request information generation unit (115) can perform a natural language generation operation based on the collected information.

[0154] Natural language processing (NLP) refers to the process of enabling computers to process and understand human language. It can refer to the work of processing natural language so that it can be understood by computers through processes such as morphological analysis and semantic analysis, and generating the results into text, voice, or images that can be understood by humans. Natural language processing technology can be broadly divided into natural language understanding and natural language generation. Natural language understanding is a technology that enables machines to understand sentences in natural language, such as machine reading comprehension, sentiment analysis, and semantic similarity measurement, while natural language generation refers to the technology that generates the results of machine processing in natural language.

[0155] In an embodiment, the request information generation unit (115) can generate request information including text information in the form of an input dialogue using a natural language generation technology.

[0156] In an embodiment, the request information generation unit (115) can generate natural language in a conversational form by applying predetermined rules or statistics to the collected information. When the request information generation unit (115) generates request information using predetermined rules and programs, it can generate text information in the form of an input conversation at a high speed with a small amount of computation.

[0157] Alternatively, in an embodiment, the request information generation unit (115) may generate request information based on a model using a neural network. In an embodiment, the request information generation unit (115) may generate input dialogue using deep learning-based artificial intelligence technology that enables a computer to understand human language and speak human-like language. When the request information generation unit (115) performs an operation to generate natural language using a neural network, it requires more complexity and a greater amount of computation compared to an operation to generate natural language based on rules, but it can obtain more accurate results.

[0158] For example, the request information generation unit (115) can generate request information in the form of an input dialogue by using an artificial intelligence model included in the electronic device (100) using on-device AI technology, i.e., a neural network that performs natural language generation, to generate natural language. In this case, text information in the form of an input dialogue can be quickly obtained because the electronic device (100) collects, calculates, and processes information on its own without going through a cloud server. However, the present invention is not limited thereto, and the request information generation unit (115) can transmit at least one of contextual information and metadata corresponding to content to an external artificial intelligence model separate from the server (200), and receive request information including text information in the form of an input dialogue from the external artificial intelligence model.

[0159] In an embodiment, the request information generation unit (115) can transmit text information in the form of an input dialogue to the server (200).

[0160] In an embodiment, the request information generation unit (115) may generate additional information as request information in addition to text information in the form of an input dialogue. The request information generation unit (115) generates text information in the form of an input dialogue based on at least one of the information collected from the context information acquisition unit (111) and the content metadata acquisition unit (113), i.e., context information and metadata corresponding to the content, and may generate all or part of the collected information that is not used to generate the text information in the form of an input dialogue as request information in the form of additional information.

[0161] For example, the request information generation unit (115) can generate information such as characteristic information of the electronic device (100), the IP address of the electronic device (100), and information about the current time as separate additional information, rather than generating it through a conversation.

[0162] In an embodiment, the request information generation unit (115) may generate additional information in the form of metadata. In an embodiment, the request information generation unit (115) may transmit request information including text information in the form of input dialogue and additional information to the server (200).

[0163] In this way, according to an embodiment, the electronic device (100) can automatically generate request information based on metadata corresponding to various situations and contents, and transmit the same to the server (200).

[0164] Figure 4 is a block diagram of a situation information acquisition unit (111) according to an embodiment.

[0165] Referring to FIG. 4, the situation information acquisition unit (111) of FIG. 4 may be an example of the situation information acquisition unit (111) of FIG. 3.

[0166] In an embodiment, the context information acquisition unit (111) may include a context information acquisition unit (410), a characteristic information acquisition unit (420), a user information acquisition unit (430), and a surrounding context information acquisition unit (440).

[0167] In an embodiment, the context information acquisition unit (410) can recognize the context from the currently displayed screen. For example, the context information may be a channel or program name, or may even include the type of object or the movement of the object included in the captured screen. The context information acquisition unit (410) can extract various information that can be used to generate a dialogue, including the channel name or channel number, the title of the content, the characters appearing in the content, the movements of the characters, and script information, from the currently displayed screen.

[0168] In an embodiment, the context information acquisition unit (410) may utilize various technologies to identify content currently being output. For example, the context information acquisition unit (410) may capture a screen at regular time intervals and perform object detection on the captured screen. In an embodiment, the context information acquisition unit (410) may perform object detection by recognizing an object in the captured screen, classifying what the object is, and identifying the location (localization) of the object.

[0169] In an embodiment, the context information acquisition unit (410) may capture a screen at regular time intervals and perform object detection on the captured screen to acquire context information. In an embodiment, the context information acquisition unit (410) may delete context information when a predetermined time has elapsed. In an embodiment, the context information acquisition unit (410) may continue to acquire context information even before receiving a predetermined control signal from the control device (50). In an embodiment, when the context information acquisition unit (410) receives a predetermined control signal from the control device (50), the context information acquisition unit (410) may acquire context information collected for a predetermined time period before receiving the predetermined control signal, for example, from 30 seconds prior to the time of receiving the control signal, as valid context information, and transmit the same to the request information generation unit (115).

[0170] Alternatively, in an embodiment, the context information acquisition unit (410) may recognize content being output by the electronic device (100) using an Automatic Content Recognition (ACR) function and collect related information. In an embodiment, the context information acquisition unit (410) may acquire characteristics from an image, audio, or video using a watermarking technology or a fingerprinting technology and compare the characteristics with a sample to identify the content.

[0171] In an embodiment, the context information acquisition unit (410) may acquire context information about the content currently being output by acquiring metadata corresponding to the content currently being output from a digital stream provided by various OTT service providers or broadcasting station servers.

[0172] In an embodiment, the characteristic information acquisition unit (420) can acquire characteristic information.

[0173] In an embodiment, the characteristic information may include information indicating what resolution information the electronic device (100) supports, for example, whether it is HD (High Definition), Full HD, Ultra HD, or 8K.

[0174] In an embodiment, the characteristic information may include information indicating whether the electronic device (100) is a curved display device having a screen with a curvature, a flexible display device capable of adjusting the curvature, or a flat display device.

[0175] In an embodiment, the characteristic information may include information indicating whether the electronic device (100) provides an ambient service. An ambient service may refer to a service that outputs a meaningful image, such as a picture, photo, or clock, instead of a black screen on the display when the electronic device (100) including the display is in an off state.

[0176] The server (200) can use the unique characteristic information of the electronic device (100) to recommend content suited to the electronic device (100). For example, the server (200) can recommend content optimized for the resolution of the electronic device, or, if the electronic device (100) is a curved display device, can recommend content optimized for a curved screen. Alternatively, if the electronic device (100) provides ambient services, the server (200) can recommend images that can be output when the electronic device is off as recommendation results.

[0177] In an embodiment, the user information acquisition unit (430) can acquire user information. In an embodiment, the user information acquisition unit (430) can acquire user information by reading it from a short-term memory or a long-term memory included in the memory (120).

[0178] In an embodiment, user information may include user profile information. User profile information may include the user's account, age, gender, presence of a spouse or children, and the ages of the children. User profile information may be obtained only if the user consents to the use of such information. User profile information may be used to infer the user's preferences or to recommend content, such as content popular with people with similar profiles, to the user.

[0179] In an embodiment, user information may include information about the user's viewing history. The user's viewing history information may include information about the types or genres of programs or content the user enjoys watching, actors or directors who appear in content the user frequently watches, etc.

[0180] In embodiments, a user's viewing history information may be used to identify the user's tastes or preferences.

[0181] In an embodiment, user information may include user-defined settings. For example, user-defined settings may include user-defined preferred programs or preferred content directly entered by the user. Furthermore, user-defined settings may include information about channels or content set as favorites by the user. In some cases, the user may use user-defined settings to include specific content or programs in recommendation results, or to exclude content of a specific genre from recommendation results.

[0182] In embodiments, user information may further include the user's web or app account information, the user's OTT subscription information, etc.

[0183] In an embodiment, the surrounding context information acquisition unit (440) may acquire surrounding context information. In an embodiment, the surrounding context information may be information indicating the surrounding context that is unrelated to the electronic device (100) itself but can be used when generating request information. For example, the surrounding context information may include the current time or location information of the electronic device (100). When the electronic device (100) is located in Seoul, Korea or Washington, USA, the time of a broadcast program output by the electronic device (100) may vary depending on the location of the electronic device (100). In addition, depending on the location of the electronic device (100), a specific channel or program may or may not be output through the electronic device (100). Therefore, the surrounding context information acquisition unit (440) may acquire surrounding context information related to the context around the electronic device (100) and transmit it to the request information generation unit (115).

[0184] In some cases, the surrounding situation information may not be acquired by the surrounding situation information acquisition unit (440), but may be automatically acquired by the server (200) when the electronic device (100) transmits request information to the server (200). For example, when the electronic device (100) calls an API (Application Programming Interface) to transmit request information to the server (200), the server (200) may acquire the location of the electronic device (100) from this call attempt using IP information of the electronic device (100), and may also acquire the current time of the electronic device (100) based on the location of the electronic device (100).

[0185] FIG. 5 illustrates, according to an embodiment, an electronic device (100) outputting a recommendation result received from a server (200) as a video signal.

[0186] In an embodiment, the electronic device (100) may transmit request information to the server (200) and receive recommendation results obtained based on the request information from the server (200).

[0187] In an embodiment, the recommendation result transmitted by the server (200) to the electronic device (100) may include at least one of information corresponding to an executable action, an outputtable channel, and content. The information corresponding to a channel or content that the electronic device (100) can output may include at least one of a thumbnail image related to the channel or content, or metadata corresponding to the channel or content. Additionally, in an embodiment, the recommendation result transmitted by the server (200) to the electronic device (100) may include a reason for the recommendation.

[0188] In an embodiment, the electronic device (100) may receive a recommendation result from the server (200) and output at least one of an audio signal and a video signal corresponding to the recommendation result.

[0189] FIG. 5 illustrates an electronic device (100) outputting a video signal corresponding to a recommendation result received from a server (200).

[0190] In an embodiment, the electronic device (100) may be set to output the recommendation result as a video signal by default. Alternatively, in an embodiment, a user may operate the setting function of the electronic device (100) to set the recommendation result to be output as a video signal.

[0191] In an embodiment, the user can also directly set the degree of conciseness of the output dialogue, the type of language to be output, the font size and font of the text information to be output, etc.

[0192] In an embodiment, the display (510) may be integrally included in the electronic device (100), or may be connected to the electronic device (100) as a separate component via a wired cable or the like.

[0193] In an embodiment, the electronic device (100) may output a video signal corresponding to a recommendation result through the display (510). In an embodiment, when the electronic device (100) receives a recommendation result from the server (200) while outputting content on the display (510), the electronic device (100) may output a video signal corresponding to the recommendation result together with the content being output.

[0194] In an embodiment, the video signal corresponding to the recommendation result may include text information in the form of a conversation.

[0195] In an embodiment, the electronic device (100) can output a video signal in the form of text information through an interface screen (520).

[0196] In an embodiment, the electronic device (100) may display an interface screen (520) on a portion of the display (510). The size, output position, transparency, and / or shape of the interface screen (520) that outputs a video signal corresponding to the recommendation result may be variously modified.

[0197] In an embodiment, the electronic device (100) may output all text information at once, or may output the entire text information over a certain period of time by outputting one word at a time or one line at a time from the very beginning, as if subtitles are output in chronological order from the beginning.

[0198] The user can visually check the interface screen (520) that outputs a video signal corresponding to the recommendation result.

[0199] In an embodiment, the video signal corresponding to the recommendation result may include information corresponding to a channel or content that the electronic device (100) can output. For example, as illustrated in FIG. 5, the video signal corresponding to the recommendation result may include information recommending a specific broadcaster or program, such as "XXXX held a discussion broadcast."

[0200] In an embodiment, the video signal corresponding to the recommendation result may include an action that the electronic device (100) can perform. For example, as illustrated in FIG. 5, the video signal corresponding to the recommendation result may include a phrase guiding the user to perform a specific action, such as, "If you are interested, press the right key."

[0201] In an embodiment, the video signal corresponding to the recommendation result may include a reason for the recommendation. For example, as illustrated in FIG. 5, the video signal corresponding to the recommendation result may include a phrase explaining the reason for recommending XXXX's discussion broadcast, such as "The number of confirmed coronavirus cases has recently been rapidly increasing."

[0202] When viewing the reasons for recommendation, the user may think as if the user is having a conversation with the electronic device (100) or as if the user is having a conversation with the server (200).

[0203] FIG. 6 illustrates, according to an embodiment, an electronic device (100) outputting a recommendation result received from a server (200) as an audio signal.

[0204] In an embodiment, the electronic device (100) may output an audio signal corresponding to the recommendation result through a speaker. In an embodiment, the electronic device (100) may be set by default to output the recommendation result as an audio signal. Alternatively, in an embodiment, the user may manipulate the setting function of the electronic device (100) to set the recommendation result to be output as an audio signal.

[0205] In an embodiment, the user may also directly set the degree of conciseness of the output dialogue, the language to be output, the type of voice to be output, etc.

[0206] In an embodiment, the speaker may be integrally included in the electronic device (100), or may be a component separate from the electronic device (100) and connected to the electronic device (100) by wire or wirelessly.

[0207] In an embodiment, when the electronic device (100) receives a recommendation result from the server (200) while outputting content on the display (510), the electronic device (100) may output an audio signal (620) corresponding to the recommendation result through a speaker while continuing to output the content on the display (510).

[0208] In an embodiment, the electronic device (100) may output an audio signal (620) corresponding to a recommendation result through a speaker, while simultaneously adjusting the volume of the electronic device (100). For example, the electronic device (100) may adjust the volume of content that was previously being output to a lower level so that the user can hear the audio signal (620) corresponding to the recommendation result better, and may output that the volume of the existing content has been adjusted through a volume display interface screen (630).

[0209] In an embodiment, when the electronic device (100) outputs the recommendation result as an audio signal (620) through a speaker instead of as a video signal, the content being displayed on the screen is not obscured by the recommendation result, so the user can view the entire content through the screen.

[0210] In an embodiment, the electronic device (100) may output the recommendation results received from the server (200) using both video signals and audio signals. For example, the electronic device (100) may output an audio signal (620) corresponding to the recommendation results through a speaker, while simultaneously outputting a video signal corresponding to the recommendation results along with the currently output content through the display (510).

[0211] FIG. 7 illustrates, according to an embodiment, an electronic device (100) outputting a recommendation result received from a server (200) as a video signal.

[0212] In an embodiment, the electronic device (100) may receive a recommendation result from the server (200) and output at least one of an audio signal and a video signal corresponding to the recommendation result.

[0213] In an embodiment, the electronic device (100) may output a video signal corresponding to the recommendation result through a multi-view screen. In an embodiment, when the recommendation result is output through a multi-view screen, the electronic device (100) may output a screen based on the recommendation result through a plurality of areas included in the multi-view screen.

[0214] In an embodiment, the electronic device (100) can output images or metadata of channels or recommended content corresponding to the recommendation results received from the server (200) through a plurality of areas included in the multi-view screen.

[0215] For example, assume that a user is watching another channel, for example, channel 9, and a control signal corresponding to channel up is transmitted to the electronic device (100) through a user input using the control device (50).

[0216] In an embodiment, the electronic device (100) may transmit request information to the server (200) based on receiving a control signal corresponding to channel up, and receive a recommendation result based on the request information from the server (200).

[0217] As illustrated in FIG. 7, the electronic device (100) can output a video signal based on the recommendation result to a multi-view screen (700). Referring to FIG. 7, the multi-view screen (700) can include a center screen (710) and four partial screens (720, 730, 740, 750).

[0218] In an embodiment, the electronic device (100) can output channel 10, which is the channel following the existing channel 9, through the center screen (710).

[0219] FIG. 7 illustrates a case where an electronic device (100) outputs a multi-view screen (700) including a center screen (710) and four sub-screens (720, 730, 740, 750) arranged respectively to the left, right, top, and bottom of the center screen (710), for a total of five sub-screens. In an embodiment, the electronic device (100) may output a screen based on a recommendation result recommended by a server (200) through the sub-screen. The screen based on the recommendation result may include an image or metadata corresponding to a recommended channel or recommended content.

[0220] In an embodiment, the electronic device (100) may provide recommendation results through four partial screens that are matched to the four-way keys included in the control device (50). In this case, the user may conveniently select a partial screen at a position matching each button by selecting the left, right, up, and down buttons of the four-way keys included in the control device (50).

[0221] Alternatively, the user may select a sub-screen included in the multi-view screen using voice input. For example, the user may select one of the multiple sub-screens by uttering words such as "right" or "up" through a microphone provided in the control device (50) or electronic device (100).

[0222] Alternatively, the user may select one of the partial screens included in the multi-view screen by making a specific gesture toward a motion sensor or camera provided in the control device (50) or the electronic device (100). For example, the user may raise, lower, right, or left his / her hand toward the camera provided in the electronic device (100), thereby selecting a partial screen corresponding to the direction of movement of the user's hand.

[0223] In an embodiment, the electronic device (100) may determine which recommendation result to output through which sub-screen among the four sub-screens (720, 730, 740, 750) included in the multi-view screen (700). For example, the electronic device (100) may determine to output a popular channel or content to the public through the left sub-screen (730) and to output a user-customized recommended channel or content through the right sub-screen (720).

[0224] In an embodiment, the electronic device (100) may output channels or content popular with the public through a partial screen (730) located to the left of the center screen (710) among the four peripheral areas included in the multi-view screen (700), i.e., the four peripheral partial screens (720, 730, 740, 750), as illustrated in FIG. 7. For example, if 'Hero Part 2' provided by a specific OTT service provider, 'YHW', is a very popular program with the public, the server (200) may provide 'Hero Part 2' as a recommendation result.

[0225] In an embodiment, the electronic device (100) may output content popular among the public as recommended content through a partial screen (730) located on the left side of the central screen (710). In an embodiment, the electronic device (100) may output the title of the recommended content, i.e., 'Hero Part 2', and 'YHW', the OTT service provider providing the content, as metadata through the partial screen (730) located on the left side.

[0226] The user can conveniently select the partial screen (720) located on the left by selecting the left button among the four-way keys included in the control device (50). When the user selects the partial screen (730) located on the left by using the control device (50), the electronic device (100) can immediately play 'Hero Part 2' provided by 'YHW', or position the 'Hero Part 2' screen provided by 'YHW' on the center screen (710) among the multi-view screens, and output the next screen including other recommended channels or other recommended contents through four partial screens (720, 730, 740, 750) to the left, right, above, and below the 'Hero Part 2' screen provided by 'YHW' located in the center.

[0227] In an embodiment, the electronic device (100) can output recommended content as user-customized content through a partial screen (730) located on the right side of the central screen (710).

[0228] In an embodiment, the electronic device (100) can output channels or content searched based on user information as images or metadata, etc. through a partial screen (720) located on the right side of the central screen (710).

[0229] For example, the server (200) can determine the user history from the user information included in the request information, and if the user has recently enjoyed content or channels related to bank bankruptcy, the server can provide the latest content or channels related to bank bankruptcy, a topic of interest to the user, as search results.

[0230] For example, the electronic device (100) can output metadata including the title of the user-customized content, i.e., 'Bank Bankruptcy Documentary', and the name of the channel on which the content is output, 'TV Please', through the right partial screen (720).

[0231] In an embodiment, the electronic device (100) can output channels above and below the currently output channel on the upper and lower sides of the center screen (710). For example, if the currently output screen (710) is channel 10, the electronic device (100) can output the previous channel (channel 9) and the next channel (channel 11) through the partial screens (740, 750) above and below the currently output screen (channel 10), respectively.

[0232] In an embodiment, the electronic device (100) may randomly or according to a predetermined rule determine which recommended result to output through which of the four partial screens (720, 730, 740, 750) included in the multi-view screen (700). For example, as illustrated in FIG. 7, the electronic device (100) may determine to output a popular channel or content to the public through the partial screen (730) on the left, and to output a user-customized recommended channel or content through the partial screen (720) on the right.

[0233] However, this is an example, and the server (200), not the electronic device (100), may determine which recommendation result to output through which partial screen among the plurality of partial screens included in the multi-view screen (700) and transmit it to the electronic device (100).

[0234] In an embodiment, the electronic device (100) may output text information in the form of a dialogue corresponding to a recommendation result along with a multi-view screen. For example, as illustrated in FIG. 7, the electronic device (100) may output a video signal corresponding to a recommendation result in the form of text information through an interface screen (520) included in an area of ​​a central screen (710) among the multi-view screens (700).

[0235] In an embodiment, the video signal corresponding to the recommendation result may include information explaining the reason for recommending the corresponding channel or content for all or part of the plurality of partial screens (720, 730, 740, 750) included in the multi-view screen (700). For example, as illustrated in FIG. 7, the text information included in the interface screen (520) may include the reason for recommending 'Hero Part 2' provided by 'YHW', which is a specific content. In addition, the text information included in the interface screen (520) may include the reason for recommending 'Bank Bankruptcy Documentary' of 'TV Please', which is a specific channel.

[0236] In FIG. 7, a case where a channel name or channel number, etc. are output through four partial screens (720, 730, 740, 750) is illustrated, but this is only one embodiment, and the electronic device (100) may output a preview video or thumbnail image of a recommended channel or recommended content through at least one partial screen among the four partial screens (720, 730, 740, 750). For example, the electronic device (100) may output an advertising poster, preview video, thumbnail image, etc. of content called 'Hero Part 2' through the left partial screen (730) of the center screen (710).

[0237] In this way, according to an embodiment, the electronic device (100) can output a screen based on the recommendation result received from the server (200) through the multi-view screen (700). In addition, the electronic device (100) can output a video signal corresponding to the recommendation result in the form of text information through the interface screen (520).

[0238] The user can know why a particular content or channel was recommended, so it feels as if the electronic device (100) or server (200) knows the user's intention.

[0239] Additionally, the user can conveniently select and use recommended channels or content using the four-way keys on the control device (50).

[0240] FIG. 8 illustrates, according to an embodiment, an electronic device (100) outputting a recommendation result received from a server (200) as a video signal.

[0241] Referring to FIG. 8, as described in the description for FIG. 7, the electronic device (100) can output a video signal corresponding to the recommendation result to a multi-view screen (700).

[0242] For example, it is assumed that a user is watching channel 23 and then selects a channel change button on a control device (50) to transmit a control signal to the electronic device (100) to change the channel to the next channel.

[0243] In an embodiment, the electronic device (100) may transmit request information to the server (200) based on receiving a control signal requesting execution of a channel change function from the control device (50), and receive a recommendation result obtained based on the request information from the server (200).

[0244] In an embodiment, the server (200) may transmit four recommended channels to the electronic device (100). In an embodiment, the electronic device (100) may generate and output a multi-view screen that includes the channel the user was watching and the four recommended channels recommended by the server (200) as partial screens, respectively.

[0245] In an embodiment, the electronic device (100) may provide recommendation results through four partial screens that are matched to the four-way keys included in the control device (50). In this case, the user may conveniently select a partial screen at a position matching each button by selecting the center button and the left, right, up, and down buttons of the four-way keys included in the control device (50).

[0246] In an embodiment, the electronic device (100) can output channel 23, which is the channel the user was watching, through the upper part of the screen (740) of the center screen (710).

[0247] In an embodiment, the electronic device (100) can output channel 24, which is the channel following channel 23 that the user was watching, through the right screen (720) of the center screen (710).

[0248] In an embodiment, the server (200) can identify channels that the user does not normally watch and skips based on user information included in the request information. For example, it is assumed that the server (200) has identified that the user does not normally watch channels 24 to 29, which are golf channels, based on the user information. When the user requests a channel upgrade, the server (200) can skip channels 24 to 29, which are the next channels, based on channel 23, which is the channel the user was watching, and transmit channel 30 as a recommendation result to the electronic device (100).

[0249] In an embodiment, the electronic device (100) may output channel 30 based on the recommendation results through the center screen (710). For example, the electronic device (100) may output metadata for channel 30, images or preview videos related to channel 30, etc. through the center screen (710).

[0250] Additionally, in the embodiment, the server (200) can identify that the user does not normally watch shopping channels based on the user information included in the request information. Furthermore, the server (200) can identify that channel 31 is a shopping channel. The server (200) can skip channel 31 and obtain channel 32 as a recommendation result, following channel 30. The electronic device (100) can output channel 32, the next recommended channel, through the lower screen (750) of the central screen (710).

[0251] In an embodiment, the electronic device (100) can output skipped channels through the right screen (720). In addition, the electronic device (100) can output channel 1, which is a channel that the user usually enjoys watching at the same time as the current time, as a recommendation result through the left partial screen (730).

[0252] In an embodiment, the electronic device (100) can output text information in a conversational form corresponding to the recommendation result along with a multi-view screen.

[0253] In an embodiment, the electronic device (100) may output text information indicating the reason for recommending the corresponding channel or content through an interface screen (520) included in an area of ​​the central screen (710). For example, as illustrated in FIG. 8, the electronic device (100) may output the reason for skipping the corresponding channel, such as, "Channels 24 to 29 are golf channels, so you don't usually enjoy them." Additionally, the electronic device (100) may output the reason for skipping channel 32, a shopping channel, such as, "The next channel is a shopping channel, so I skipped it." The user can easily understand why the corresponding channel was skipped by viewing the recommendation reason.

[0254] In this way, according to an embodiment, when a user selects a channel change button using a control device (50), the server (200) can distinguish between channels that the user does not prefer and channels that the user prefers based on user information and transmit the recommendation results accordingly to the electronic device (100). The electronic device (100) can satisfy the user's needs by automatically skipping channels that the user does not normally watch and recommending channels that the user may watch.

[0255] Figure 9 is a block diagram of an electronic device (100) according to an embodiment.

[0256] The electronic device (100) of FIG. 9 may be an example of any one of the electronic devices (100) of FIGS. 1 to 4. Hereinafter, any description overlapping with that described in FIGS. 1 to 4 will be omitted.

[0257] Referring to FIG. 9, the electronic device (100) may include a processor (110) and a memory (120). The processor (110) and the memory (120) included in the electronic device (100) perform the same operations as the processor (110) and the memory (120) included in the electronic device (100) of FIG. 2, and thus the same drawing reference numerals are used.

[0258] In an embodiment, the electronic device (100) may further include, in addition to the processor (110) and the memory (120), a tuner unit (910), a communication unit (920), a detection unit (930), an input / output unit (940), a video processing unit (950), a display (510), an audio processing unit (970), an audio output unit (980), and a user input unit (990).

[0259] The tuner unit (910) can select and tune only the frequency of the channel to be received by the electronic device (100) from among many radio wave components through amplification, mixing, resonance, etc. of broadcast content received via wire or wireless. The content received through the tuner unit (910) is decoded and separated into audio, video, and / or additional information. The separated audio, video, and / or additional information can be stored in the memory (120) under the control of the processor (110).

[0260] The communication unit (920) may include at least one communication module capable of performing communication in accordance with the performance and structure of the electronic device (100). The communication unit (920) may include at least one of a wireless LAN module (921), a Bluetooth module (922), and a wired Ethernet (923).

[0261] The wireless LAN module (921) can transmit and receive Wi-Fi signals with peripheral devices according to the Wi-Fi communication standard.

[0262] The Bluetooth module (922) can receive Bluetooth signals transmitted from peripheral devices according to the Bluetooth communication standard. The Bluetooth module (922) can be a BLE (Bluetooth Low Energy) communication module and can receive BLE signals. The Bluetooth module (922) can continuously or temporarily scan BLE signals to detect whether a BLE signal is received.

[0263] In an embodiment, the communication unit (920) may connect the electronic device (100) to a peripheral device, an external device, a server (200), etc., under the control of the processor (110). In an embodiment, the communication unit (920) may transmit request information generated by the electronic device (100) to the server (200) using a wired or wireless communication network, and receive recommendation results obtained based on the request information from the server (200).

[0264] In an embodiment, the communication unit (920) may transmit and receive information with the server (200) using communication based on the Web Socket protocol or the HyperText Transfer Protocol (HTTP). In an embodiment, the electronic device (100) may configure a message in the form of a message template class object provided by a JSON (JavaScript Object Notation) object, but is not limited thereto. The communication unit (920) may be configured to transmit a payload in the form of a JSON or XML based text to the server (200) and receive a result payload from the server (200) in order to transmit and receive information including text information in the form of a conversation with the server (200).

[0265] The detection unit (930) detects the user's voice, the user's image, or the user's interaction, and may include a microphone (931), a camera unit (932), a light receiving unit (933), and a sensing unit (934).

[0266] When a user transmits a control signal to an electronic device (100) via a voice control device (50), the microphone (931) can receive an audio signal including the user's uttered voice or noise, and convert the received audio signal into an electrical signal and output it to the processor (110).

[0267] The camera unit (932) includes a sensor (not shown) and a lens (not shown), and can capture an image formed on the screen and transmit it to the processor (110).

[0268] The optical receiver (933) can receive an optical signal (including a control signal). The optical receiver (933) can receive an optical signal corresponding to a user input (e.g., touch, press, touch gesture, voice, or motion) from a control device (150) such as a remote control or a mobile phone.

[0269] The input / output unit (940) can receive video (e.g., dynamic image signal or still image signal), audio (e.g., voice signal or music signal), and additional information from an external device under the control of the processor (110).

[0270] The input / output unit (940) may include one of an HDMI port (High-Definition Multimedia Interface port, 941), a component jack (component jack, 1042), a PC port (PC port, 943), and a USB port (USB port, 944). The input / output unit (940) may include a combination of an HDMI port (941), a component jack (942), a PC port (943), and a USB port (944).

[0271] The video processing unit (950) processes image data to be displayed by the display (510) and can perform various image processing operations such as decoding, rendering, scaling, noise filtering, frame rate conversion, and resolution conversion for the image data.

[0272] The display (510) can display content received from a broadcasting station, an external server, an external storage medium, or other external devices, or provided by various apps, such as OTT service providers or metaverse content providers. The content can be a media signal, including video signals, images, text signals, etc.

[0273] In an embodiment, the display (510) may output a video signal corresponding to the recommendation result. The video signal corresponding to the recommendation result may include text information in the form of a conversation.

[0274] The audio processing unit (970) performs processing on audio data. The audio processing unit (970) can perform various processing such as decoding, amplification, and noise filtering on audio data.

[0275] The audio output unit (980) can output audio included in content received through the tuner unit (910) under the control of the processor (110), audio input through the communication unit (920) or input / output unit (940), and audio stored in the memory (120). The audio output unit (980) can include at least one of a speaker (981), a headphone (982), or an S / PDIF (Sony / Philips Digital Interface: output terminal) (983).

[0276] In an embodiment, the audio output unit (980) may output an audio signal corresponding to the recommendation result. In an embodiment, the audio output unit (980) may adjust the volume of an audio signal previously output as it outputs an audio signal corresponding to the recommendation result.

[0277] The user input unit (990) can receive user input for controlling the electronic device (100). The user input unit (990) can include various types of user input devices, including, but not limited to, a touch panel that detects a user's touch, a button that receives a user's push operation, a wheel that receives a user's rotation operation, a keyboard, and a dome switch, a microphone for voice recognition, and a motion detection sensor that senses motion.

[0278] In an embodiment, a user may transmit a control signal to a user input unit (990) by using a control device (50) through a key input provided in the control device (50), a voice input through a microphone provided in the control device (50), or a gesture input through a motion sensor provided in the control device (50). Alternatively, if the control device (50) is an electronic device that performs other operations as its main function in addition to controlling the electronic device (100), such as a smartphone or an AI speaker, the user may transmit a control signal to the user input unit (990) through a remote control app installed in the control device (50).

[0279] Fig. 10 is a block diagram of a server (200) according to an embodiment.

[0280] The server (200) of FIG. 10 may be an example of the server (200) of FIG. 1.

[0281] In an embodiment, the server (200) may include a processor (210) and a memory (220).

[0282] In an embodiment, the memory (220) may store at least one instruction. The memory (220) may store at least one program executed by the processor (210). In addition, the memory (220) may store data input to or output from the server (200).

[0283] The memory (220) may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), a RAM (Random Access Memory), a SRAM (Static Random Access Memory), a ROM (Read-Only Memory), an EEPROM (Electrically Erasable Programmable Read-Only Memory), a PROM (Programmable Read-Only Memory), a magnetic memory, a magnetic disk, and an optical disk.

[0284] A processor (210) according to an embodiment controls the overall operation of the server (200). The processor (210) can control the server (200) to function by executing one or more instructions stored in the memory (220). In an embodiment, there may be one or more processors (210).

[0285] In an embodiment, the processor (210) may store at least one neural network. In an embodiment, the processor (210) may generate output data from input data using one or more neural networks. Alternatively, the memory (220), rather than the processor (210), may store the neural network, i.e., the AI ​​model.

[0286] In an embodiment, one or more processors (210) may implement an artificial intelligence (AI) system or AI technology by using a neural network stored in the processor (210) and / or memory (220).

[0287] An artificial intelligence (AI) system is a computer system that implements human-level intelligence. It is a system in which the machine learns and makes judgments on its own, and its recognition rate improves with use.

[0288] Artificial intelligence technology consists of machine learning (deep learning) technology that uses algorithms that classify / learn the characteristics of input data on their own, and element technologies that use machine learning algorithms to simulate the cognitive and judgment functions of the human brain.

[0289] The element technologies may include, for example, at least one of: linguistic understanding technology that recognizes human language / characters; visual understanding technology that recognizes objects as if they were human vision; inference / prediction technology that judges information and logically infers and predicts; knowledge expression technology that processes human experience information into knowledge data; and motion control technology that controls autonomous driving of vehicles and movements of robots.

[0290] AI technology can be implemented using algorithms. Here, an algorithm or set of algorithms for implementing AI technology is called a neural network. A neural network can receive input data, perform operations for analysis and classification, and output result data. In order for a neural network to accurately output result data corresponding to the input data, the neural network must be trained. Here, "training" can mean inputting various data into the neural network and training the neural network so that the neural network can discover or learn methods for analyzing the input data, methods for classifying the input data, and / or methods for extracting features necessary for generating result data from the input data. Training a neural network means creating an artificial intelligence model with desired characteristics by applying a learning algorithm to a large amount of learning data. In an embodiment, this learning may be performed on the server (200) where the artificial intelligence is executed, or may be performed through a separate server / system.

[0291] Here, a learning algorithm is a method for training a target device (e.g., a robot) using a large amount of learning data, thereby enabling the target device to make decisions or predictions on its own. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning. Unless otherwise specified, the learning algorithms in the embodiments are not limited to the examples described above.

[0292] A set of algorithms that output output data corresponding to input data through a neural network, software that executes the set of algorithms, and / or hardware that executes the set of algorithms can be called an 'AI model' (or, 'artificial intelligence model' or neural network model, neural network).

[0293] In an embodiment, one or more processors (210) may process input data according to predefined operating rules or AI models. The predefined operating rules or AI models may be created using specific algorithms. Furthermore, the AI ​​models may be learned using specific algorithms. The processors (210) may generate output data corresponding to the input data through the AI ​​models.

[0294] In an embodiment, the neural network used by the processor (210) may be a generative artificial intelligence model.

[0295] Generative AI is a model that learns existing content, such as text, voice, and images, and then generates similar content. Generative AI is an AI technology that goes beyond classifying and recognizing input content patterns to generate new data that can communicate with people. Generative AI technology has rapidly developed using hyperscale AI technology.

[0296] A super-large AI model refers to a model trained based on large-scale data.

[0297] In an embodiment, the learning data may include various request information that can be received from the electronic device (100), various functions or operations that the electronic device (100) can perform, various contents, programs, channels that the electronic device (100) can output, metadata for these, etc.

[0298] In an embodiment, the server (200) may receive request information from the electronic device (100). In an embodiment, one or more processors (210) may obtain recommendation results corresponding to the request information using one or more neural networks.

[0299] In an embodiment, one or more neural networks may obtain recommendation results using inference prediction techniques. Inference prediction techniques are techniques for logically inferring and predicting based on information, and may include knowledge / probability-based inference, optimization prediction, preference-based planning, and recommendations.

[0300] In an embodiment, the recommendation result obtained by the server (200) using the artificial intelligence model may be information in the form of an output dialogue.

[0301] In an embodiment, one or more processors (210) may obtain a recommendation reason along with a recommendation result using an artificial intelligence model. In an embodiment, the server (200) may transmit the recommendation result including the recommendation reason to the electronic device (100).

[0302] In an embodiment, one or more processors (210) may obtain additional information. In an embodiment, the additional information may be information collected by the server (200) to provide more accurate recommendation results to the electronic device (100). For example, the additional information may include at least one of metadata corresponding to content, popular content, curated content, content selected based on a specific theme, and Key Performance Index (KPI) information.

[0303] In an embodiment, one or more processors (210) may obtain user information. The user information may include at least one of a user profile, configuration information, user or electronic device-specific preferred channels or content, user preferred genres, or past viewing history.

[0304] In an embodiment, one or more processors (210) may input at least one of additional information and user information together with request information to one or more neural networks, thereby obtaining a recommendation result based on the request information and at least one of the additional information and user information from the one or more neural networks.

[0305] In an embodiment, the server (200) may input candidate information that may serve as a reason for recommendation into one or more neural networks. The candidate information that may serve as a reason for recommendation is information likely to be related to the recommendation result, and may include user preferences, user viewing history, user settings, popular channels or content, etc.

[0306] In an embodiment, one or more neural networks utilized by one or more processors (210) may include multiple theme-specific neural networks. The theme-specific neural networks may refer to artificial intelligence models trained with learning data of different themes.

[0307] In an embodiment, the theme-specific neural network may include at least one of various themes, such as a channel switching theme, a personalized theme, a popular content theme, a curation theme, a fan theme, a KPI (Key Performance Index) theme, and other themes that a generative artificial intelligence model may suggest, such as themes learned from Wednesday dramas, Friday-Saturday dramas, variety shows, 90s movies, and classic movies.

[0308] In an embodiment, one or more processors (210) may input information corresponding to a plurality of theme-specific neural networks, obtained based on request information received from the electronic device (100), into a plurality of theme-specific neural networks. Alternatively, in an embodiment, one or more processors (210) may input information obtained based on request information received from the electronic device (100) into only one or a plurality of selected neural networks among the plurality of theme-specific neural networks.

[0309] In an embodiment, one or more processors (210) may obtain recommendation results from at least one of a plurality of theme-specific neural networks. In an embodiment, when recommendation results are obtained from a plurality of theme-specific neural networks, one or more processors (210) may synthesize the plurality of recommendation results to obtain a final recommendation result.

[0310] In an embodiment, one or more processors (210) may generate, as a recommendation result, at least one of information about an action that the electronic device (100) can perform, information about a channel or content that the electronic device (100) can output, and a recommendation reason.

[0311] In an embodiment, the recommendation result may further include information corresponding to additional functions that the electronic device (100) can perform. The information corresponding to additional functions that the electronic device (100) can perform may include, for example, information guiding additional operations.

[0312] In an embodiment, one or more processors (210) may transmit recommendation results to the electronic device (100).

[0313] In this way, according to an embodiment, the server (200) can generate recommendation results corresponding to a recommendation request transmitted by an electronic device (100) in a conversational form using one or more ultra-large artificial intelligence models.

[0314] Fig. 11 is a block diagram of a server (200) according to an embodiment.

[0315] In an embodiment, the server (200) may include a learning module (211), a reinforcement learning module (212), a communication module (213), a pre-trained model (214), a fine-tuned model (215), a recommendation result generation module (216), an additional information DB (217), and a user-customized DB (218).

[0316] In an embodiment, each module included in the server (200) may be a functional and structural combination of hardware for implementing the technical concepts of the present disclosure and software for operating the hardware. For example, a module may refer to a logical unit of a given code and hardware resources for executing the given code, and is not necessarily limited to physically connected code or a single type of hardware.

[0317] In an embodiment, the server (200) may collect additional information to provide more accurate recommendation results. The additional information DB (217) according to the embodiment may be a database that stores additional information.

[0318] In an embodiment, the additional information may include at least one of metadata corresponding to the content, popular content, curated content, content selected based on a specific theme, and Key Performance Index (KPI) information.

[0319] For example, if the electronic device (100) is a television, the additional information may include metadata corresponding to all programs or content that can be broadcast through the television, i.e., additional information about the programs or content. For example, the additional information may include EPG information for all programs or content, program schedule information, program program guide, title, content, cast, director, award-winning status, rating, etc. of the program or content, preview videos or thumbnail images for the program or content, etc.

[0320] In an embodiment, the server (200) may directly obtain and collect additional information from an external content provider, OTT service provider, broadcasting station, etc.

[0321] In an embodiment, the additional information acquired by the server (200) may overlap with the information included in the request information received from the electronic device (100). The overlapping information may be information acquired from the same source or from different sources. For example, the request information generated by the electronic device (100) may include metadata corresponding to specific content, for example, metadata corresponding to the first content. In this case, the electronic device (100) may acquire the metadata corresponding to the first content from metadata included in a broadcast stream received from the broadcasting station. Alternatively, the server (200) may acquire metadata corresponding to the first content from an OTT service provider other than the broadcasting station and use this as additional information.

[0322] According to an embodiment, a user-customized DB (218) may store user information. The user-customized DB (218) may store a history of past selections made by a user or an electronic device (100) for each user, for example, for each user account, or for each electronic device (100), for example, for each IP address of the electronic device (100) or each account unique to the electronic device (100). In an embodiment, the server (200) may reflect user information in the recommendation results based on the past history stored in the user-customized DB (218). The past history data stored in the user-customized DB (218) may be updated when user feedback is received from the electronic device (100).

[0323] The pre-trained model (214) according to the embodiment may be a generative AI capable of recognizing and generating conversations. In the embodiment, the pre-trained model (214) may be a super-large AI such as the already known ChatGPT or GPT4.0, or a super-large AI model to be developed in the future.

[0324] In an embodiment, the pre-trained model (214) may be a large-scale learning model trained on large-scale training data, capable of processing and analyzing large-scale data with high accuracy. For example, the pre-trained model (214) may be a model with millions to hundreds of billions of parameters.

[0325] According to an embodiment, the learning module (211) may be a module that trains an artificial intelligence model to obtain a desired result through a neural network. The learning module (211) may train a pre-trained model (214) using large-scale learning data to obtain a recommendation result. The learning module (211) may train the pre-trained model (214) to set the values ​​of a plurality of weights applied to each of a plurality of nodes forming a neural network. Here, the weights may mean the connection strength between each node of the neural network. The weight values ​​may be optimized through iterative learning and may be iteratively modified until the accuracy of the result satisfies a predetermined reliability level. The learning module (211) may train the pre-trained model (214) to generate a fine-tuned model (215) having the final set weight values.

[0326] In an embodiment, the learning module (211) may enable the pre-trained model (214) to learn input conversations. In an embodiment, the learning module (211) may train the pre-trained model (214) to learn various recommendation request conversations that can be received from the electronic device (100). In an embodiment, the learning module (211) may input various situational request information that can be received from the electronic device (100) into the pre-trained model (214) and train the pre-trained model (214) to obtain recommendation results corresponding to the request information from the pre-trained model (214).

[0327] In an embodiment, the learning module (211) can train the model by adjusting the learning cycle of the pre-trained model (214) according to the learning speed. For example, if learning takes one week, the learning module (211) can train the pre-trained model (214) in one-week units.

[0328] The fine-tuned model (215) according to the embodiment may be a model generated by continuously training the pre-trained model (214). The fine-tuned model (215) may be continuously trained with additional training data. In the embodiment, the fine-tuned model (215) may be specialized as a model suitable for generating recommendation results according to the present disclosure by additionally training data utilized in the domain of the electronic device (100) and user-customized data.

[0329] In an embodiment, the fine-tuned model (215) can be tuned into a model that can generate recommendation results that better match the user's intention by continuously learning based on additional information included in the additional information DB (217), user history data included in the user-customized DB (218), user feedback results, etc.

[0330] The reinforcement learning module (212) according to the embodiment may be a module that learns user feedback. In the embodiment, the reinforcement learning module (212) may be a module that additionally learns feedback provided by the user with respect to the recommendation results obtained by the fine-tuned model (215). In the embodiment, the reinforcement learning module (212) may continuously train the fine-tuned model (215) based on additional information included in the additional information DB (217), user history data included in the user-customized DB (218), user feedback results, etc.

[0331] In an embodiment, the reinforcement learning module (212) can also train the model by setting a cycle according to the learning speed, similar to the learning module (211). In an embodiment, the reinforcement learning module (212) can train the model by applying additional information, user input, user feedback data, etc. to the model according to the cycle at which the additional information DB (217) or the user-customized DB (218) is updated, the cycle at which user feedback is received, etc.

[0332] The communication module (213) according to the embodiment may be a module capable of performing communication with the communication unit (920) of the electronic device (100). In the embodiment, the communication module (213) may provide an API (Application Programming Interface) that can exchange conversations with a client through network communication.

[0333] The communication module (213) may include at least one of a wireless LAN module, a Bluetooth module, and a wired Ethernet. In an embodiment, the communication module (213) may receive request information from an electronic device (100) using a wired or wireless communication network, and transmit a recommendation result based on the request information generated by the server (200) to the electronic device (100).

[0334] In an embodiment, the communication module (213) may transmit and receive information with the electronic device (100) using communication based on the Web Socket protocol or the HyperText Transfer Protocol (HTTP). The communication module (213) may transmit and receive a payload in the form of JSON or XML based on a test to transmit and receive information including text information in the form of a conversation with the electronic device (100).

[0335] The recommendation result generation module (216) according to the embodiment may be a module that generates a recommendation result using the latest learned fine-tuned model (215).

[0336] In an embodiment, the recommendation result generation module (216) can input request information into the fine-tuned model (215) and obtain a recommendation result corresponding to the request information from the fine-tuned model (215).

[0337] In an embodiment, the fine-tuned model (215) may include multiple theme-specific neural networks. In an embodiment, when the fine-tuned model (215) includes multiple theme-specific neural networks, the recommendation result generation module (216) may obtain multiple recommendation results using the multiple theme-specific neural networks. In an embodiment, the recommendation result generation module (216) may obtain a final recommendation result by synthesizing multiple recommendation results obtained from the multiple theme-specific neural networks.

[0338] In an embodiment, the recommendation result generation module (216) may obtain additional information from an additional information DB (217) and user information from a user-customized DB (218). In an embodiment, the recommendation result generation module (216) may input additional information and user information together with request information into a fine-tuned model (215) and obtain a recommendation result from the fine-tuned model (215).

[0339] In an embodiment, the recommendation result may include information about various contents, programs, channels, etc. that can be output by the electronic device (100). In an embodiment, the recommendation result may include a URL address indicating a specific location on an external server or an external OTT service provider server where a broadcasting station channel or contents, etc., or contents are stored, where recommended channels, recommended contents, or recommended programs are available, metadata indicating additional information about the recommended channels, contents, or programs, and advertisement videos, preview videos, thumbnails, etc. for the recommended channels, contents, or programs.

[0340] In an embodiment, the recommendation result generation module (216) may generate an action that can be performed on the electronic device (100) as a recommendation result. For example, the recommendation result generation module (216) may obtain guidance information on an action for a Multi Device Experience (MDE) that can be performed on the electronic device (100) as a recommendation result.

[0341] In an embodiment, the recommendation result generation module (216) may obtain a recommendation reason along with the recommendation result. In an embodiment, the recommendation reason may be information explaining the reason for recommending an action that the electronic device (100) can perform, or the reason for recommending a specific channel, content, or channel.

[0342] In an embodiment, the recommendation result generation module (216) may generate recommendation results in a conversational format. In an embodiment, the recommendation result generation module (216) may output the recommendation results including the reason for the recommendation. For example, if the recommendation result generation module (216) outputs the recommendation results as metadata for four channels or content, the recommendation result generation module (216) may generate the reason for recommending at least one of the four channels or content along with the recommendation results.

[0343] FIG. 12 illustrates that, according to an embodiment, when a server (200) obtains a recommendation result, it also obtains a recommendation reason.

[0344] Referring to the left drawing of FIG. 12, in an embodiment, the recommendation result generation module (216) may input request information into the fine-tuned model (215). In an embodiment, the recommendation result generation module (216) may input an input dialogue (1212) as request information into the fine-tuned model (215).

[0345] In an embodiment, the recommendation result generation module (216) may obtain candidate information (1211). The candidate information (1211) indicates candidates that may be related to the recommendation result and thus may be reasons for the recommendation, and may include various information such as user preferred or dispreferred genres that can be identified from EPGs, program schedules, user information already stored in a user-customized DB, channels or programs that the user prefers or disprefers, user viewing tendencies, programs or popular content that are popular among other people with similar profiles when considering the user's age, the user's region of residence, the user's family relationships, etc., or programs or content that are popular among the general public without considering the user profile.

[0346] In an embodiment, the recommendation result generation module (216) may input candidate information (1211) together with an input dialogue (1212) into the fine-tuned model (215).

[0347] In an embodiment, the fine-tuned model (215) may be a neural network trained to receive candidate information (1211) that may be a reason for recommendation along with an input conversation (1212), and obtain a recommendation result by considering the input conversation (1212) and the candidate information (1211) together.

[0348] In an embodiment, the fine-tuned model (215) can obtain a recommendation result including a result conversation (1213). For example, if the fine-tuned model (215) recommends a specific content, the fine-tuned model (215) can obtain a recommendation reason, such as whether the content is recommended because the genre of the content is a genre preferred by the user, because the content is a sequel to content that the user has previously watched, because the content is currently available based on EPG information, because the content is popular among people of a similar age and gender to the user, or because the content is related to a topic that is very popular among the general public these days.

[0349] In an embodiment, the fine-tuned model (215) can identify which category of candidate information among the various categories of candidate information included in the candidate information (1211) is related to the recommendation result when obtaining the recommendation result. In an embodiment, the fine-tuned model (215) can score the degree of relevance or matching between the candidate information for each of the multiple categories included in the candidate information (1211) and the recommendation result. For example, it is assumed that the fine-tuned model (215) obtains a relevance score of 90 points between the recommended content and content that the user has viewed, a relevance score of 70 points between the recommended content and the user's preferred genre, and a relevance score of 30 points between the recommended content and public popularity.

[0350] In an embodiment, the fine-tuned model (215) may obtain candidate information with a high priority among candidate information (1211) as a reason for recommendation. The candidate information with a high priority may be candidate information with a high relevance score between the recommended content and the candidate information.

[0351] In an embodiment, the recommendation result generation module (216) may obtain one or more candidate information as reasons for recommendation in order of priority. For example, the recommendation result generation module (216) may obtain that the recommended content is highly related to content that the user has previously viewed and that the recommended content is in the user's preferred genre as reasons for recommendation. The recommendation result generation module (216) may generate a result dialogue (1213) that includes the reason for recommending the content, such as, "The recommended content is a sequel to content that the user has previously viewed. Also, I recommended it because it is in the adventure genre, which is the user's favorite genre."

[0352] In another embodiment, referring to the right drawing of FIG. 12, the recommendation result generation module (216) may obtain recommendation results by inputting only an input dialogue (1231) without inputting candidate information using the fine-tuned model (215). In the embodiment, the recommendation result generation module (216) may use another neural network to select categories that are likely to be candidate information related to the reason for recommendation among multiple categories. Another neural network may select categories that are likely to be candidate information and obtain the result as an AI suggestion (1232).

[0353] In an embodiment, the recommendation result generation module (216) is a different model from the fine-tuned model (215), and is a fine-tuned model (1250) that outputs a reason, which can input an AI suggestion (1232). The fine-tuned model (1250) that outputs a reason can input an AI suggestion (1232) and, among the input AI suggestions (1232), can obtain candidate information that is highly relevant to the recommendation result as a recommendation reason.

[0354] In an embodiment, the recommendation result generation module (216) may obtain a recommendation result from a fine-tuned model (215) and obtain a recommendation reason from another model, a fine-tuned model (1250) that outputs a reason. The recommendation result generation module (216) may merge the recommendation result and the recommendation reason to generate a recommendation result that includes the recommendation reason.

[0355] In this way, according to an embodiment, the server (200) inputs candidate information that can be a reason for recommendation along with request information into the same or separate generation AI model, thereby obtaining a recommendation result and a reason for recommendation together or separately from the generation AI model. The server (200) obtains a recommendation result including a reason for recommendation and transmits it to the electronic device (100), thereby explaining to the user the reason for recommending a specific content, program, or action.

[0356] FIG. 13 is a diagram showing, according to an embodiment, a server (200) generating recommendation results in a consistent manner.

[0357] In an embodiment, the server (200) may obtain a recommendation result corresponding to the request information from the electronic device (100). In an embodiment, the server (200) may obtain the recommendation result in a consistent manner and transmit the recommendation result to the electronic device (100).

[0358] The electronic device (100) can configure a multi-view screen based on the recommendation result received from the server (200), and output images or metadata of channels or contents corresponding to the recommendation result received from the server (200) through a plurality of partial screens included in the multi-view screen.

[0359] In an embodiment, the electronic device (100) may determine which recommendation result to output through which sub-screen of the multi-view screen. Alternatively, the server (200) may determine which recommendation result to output through which sub-screen of the multi-view screen and notify the electronic device (100) of this.

[0360] Referring to FIG. 13, the server (200) or the electronic device (100) can output content according to channel switching through the upper and lower partial screens of the multi-view screen, and output recommended content through the left and right partial screens.

[0361] The user can select any one of the multi-view screens using the four-way keys provided on the control device (50).

[0362] For example, if a user selects a partial screen above or below the multi-view screen, the server (200) may determine that the user is not interested in content recommendations but rather in channel browsing. In response to the user selecting a partial screen above or below the multi-view screen, the server (200) may continuously and consistently transmit the next channel or previous channel centered on the current channel as a recommendation result to the electronic device (100). The user may view the recommended results in a consistent manner through the multi-view screen, and may easily select and use the next channel or the previous channel by selecting a partial screen above or below the multi-view screen using the four-way keys provided on the control device (50).

[0363] Alternatively, if the user selects the left or right partial screen of the multi-view screen, the server (200) may determine that the user is not interested in channel browsing but is interested in recommended content. In response to the user selecting the left or right partial screen of the multi-view screen, the server (200) may continue to search for recommended content centered on the current content output through the center partial screen and transmit the recommendation result to the electronic device (100). For example, if the user selects the left partial screen of the multi-view screen, the server (200) may determine that the user is browsing content popular among the public, and may continue to obtain content popular among the public as a recommendation result and output the recommendation result through the left partial screen of the multi-view screen.

[0364] Likewise, for example, if a user selects the right part of the multi-view screen, the server (200) may determine that the user is searching for user-customized content rather than content popular with the public, and may continue to obtain user-customized content as a recommendation result and output the recommendation result through the right part of the multi-view screen.

[0365] In an embodiment, whether a user selected a partial screen in a multi-view screen using a control device (50) can be stored as a user history in a user-customized DB, and can then be used to identify the user's preferences, etc. when the server (200) recommends content.

[0366] The user can view the recommended results in a consistent manner through the multi-view screen and select and use a specific partial screen using the four-way keys provided on the control device (50).

[0367] FIG. 14 is a diagram showing, according to an embodiment, a server (200) generating recommendation results in a consistent manner.

[0368] For example, assume that while the electronic device (100) is outputting channel 20, the user selects the channel up button using the control device (50). In response to receiving the channel up button from the control device (50), the electronic device (100) can generate request information and transmit it to the server (200).

[0369] In an embodiment, the server (200) can obtain recommendation results in a consistent manner and transmit them to the electronic device (100). For example, the server (200) can output only channels that the user is interested in through the upper and lower partial screens of the multi-view screen, and output skipped channels through the left and right partial screens.

[0370] The user can select any one of the multi-view screens using the four-way key provided on the control device (50). For example, if the user selects the upper or lower partial screen of the multi-view screen, the server (200) can determine that the user wants to search only for channels of interest while skipping channels of no interest. In response to the user selecting the upper or lower partial screen of the multi-view screen, the server (200) can continuously and consistently skip channels of no interest with the current channel as the center. The server (200) can recommend only channels of interest to the user through the upper or lower partial screen of the multi-view screen.

[0371] Alternatively, if the user selects the left or right partial screen of the multi-view screen, the server (200) may determine that the user is not interested in quick channel navigation and wants to browse channels one by one without skipping channels. In response to the user selecting the left or lower partial screen of the multi-view screen, the server (200) may continuously output the next channel or previous channel as recommended results through the left and right partial screens, centered on the current channel output through the center partial screen.

[0372] In an embodiment, whether a user selected a partial screen in a multi-view screen using a control device (50) can be stored as a user history in a user-customized DB, and can then be used when the server (200) recommends a channel.

[0373] In this way, according to the embodiment, the server (200) can generate recommended results in a consistent manner. The user can view the recommended results in a consistent manner through the multi-view screen and easily access the desired results using the four-way keys provided on the control device (50).

[0374] FIG. 15 is a diagram showing, according to an embodiment, a server (200) obtaining recommendation results using multiple theme-specific neural networks.

[0375] Referring to FIG. 15, in an embodiment, the server (200) can obtain a recommendation result using an AI model (1520).

[0376] In an embodiment, the AI ​​model (1520) used by the server (200) may include multiple theme-specific neural networks. In an embodiment, the multiple theme-specific neural networks may be fine-tuned models, each trained with training data of different themes.

[0377] In an embodiment, the theme-specific neural network may be a neural network trained with training data of various themes. In an embodiment, the theme-specific neural network may be in various forms, for example, a channel switching theme model trained with information about channels frequently viewed by a user, a personalized theme model trained with a user's viewing history, preferences, user setting information, etc., a popular content theme model trained with popular programs or contents among the public, a fan theme model trained with content including a specific celebrity or performer, a KPI theme model trained with KPI (Key Performance Index) information collected by the server (200) from content information within the electronic device (100), such as popular apps, popular contents, and preferred menus of the electronic device (100), and other various themes that a generative artificial intelligence model may suggest, for example, an advertisement meta model trained with only advertisements, a theme model trained with Wednesday dramas, Friday-Saturday dramas, entertainment programs, 90s movies, classic movies, etc. Additionally, thematic neural networks can include curated themes, which administrators provide by mapping channel numbers to specific channels. A curated theme can be a theme model trained solely on specific topics, such as programs related to environmental protection, as determined by the server administrator or others.

[0378] In an embodiment, the recommendation result generation module (216) may analyze the input conversation included in the request information. In an embodiment, the recommendation result generation module (216) may use natural language understanding (NLU) technology to decipher the meaning of the text included in the request information and derive its intent. In an embodiment, if the request information is in voice form rather than text, the recommendation result generation module (216) may utilize automatic speech recognition (ASR) and NLU together to analyze the data.

[0379] In an embodiment, the recommendation result generation module (216) may obtain only available conversations among the information included in the request information as input conversations. In an embodiment, the recommendation result generation module (216) may input the input conversations into each of multiple theme-specific neural networks.

[0380] Alternatively, in an embodiment, the recommendation result generation module (216) may determine which of the multiple theme-specific neural networks to input the input conversation to. In an embodiment, the recommendation result generation module (216) may select all or only some of the multiple theme-specific neural networks based on their relevance to at least one of the request information, additional information, and user information. The recommendation result generation module (216) may input the input conversation only to the neural network of the selected theme.

[0381] In an embodiment, the recommendation result generation module (216) may input the same input conversation to all of the multiple theme-specific neural networks or to the neural networks of some selected themes.

[0382] Alternatively, in an embodiment, the recommendation result generation module (216) may modify the input dialogue to fit the neural network of the selected theme. For example, rather than inputting the same input dialogue to multiple theme-specific neural networks, the recommendation result generation module (216) may modify the input dialogue to fit each theme-specific neural network. The recommendation result generation module (216) may modify the input dialogue to obtain information corresponding to the theme-specific neural network, and input the information corresponding to the theme-specific neural network to the theme-specific neural network.

[0383] Generative AI models can produce more accurate and relevant results when asking questions using keywords that are as clear and concise as possible. Therefore, rather than inputting all input conversations into the neural networks for each theme, the recommendation result generation module (216) extracts only the input conversations relevant to each theme, thereby obtaining accurate and concise input conversations tailored to each theme's neural network.

[0384] In an embodiment, the recommendation result generation module (216) may classify text included in the request information using natural language understanding technology. For example, if there is request information related to the user's preferences among the request information received from the electronic device (100), the recommendation result generation module (216) may analyze the text included in the request information, identify only the text related to the user's preferences, rather than the entire text included in the request information, and obtain this as input data to be input into a neural network trained with a personalized theme.

[0385] For example, if the request information includes a request to recommend a popular action movie among the public, the recommendation result generation module (216) may obtain only the text requesting a recommendation of a popular action movie among the public as input data, rather than the entire request information, and input this into the popular content theme.

[0386] In an embodiment, the recommendation result generation module (216) may obtain multiple recommendation results from multiple theme-specific neural networks. The recommendation result generation module (216) may obtain one or more recommended channels or recommended contents in order of high scores among the multiple recommendation results as the final recommendation result. The channels or contents with high scores among the multiple recommendation results may be, for example, channels or contents obtained as recommendation results in duplicate from multiple theme-specific neural networks. Alternatively, the recommendation result generation module (216) may obtain the final recommendation result by assigning a high weight to a channel or contents obtained from a theme-specific neural network that is highly related to at least one of the request information, additional information, and user information.

[0387] In this way, according to the embodiment, the server (200) can obtain more accurate recommendation results by using multiple theme-specific neural networks.

[0388] FIG. 16 illustrates a recommendation result output by an electronic device (100) according to an embodiment.

[0389] Referring to FIG. 16, the electronic device (100) can output a recommendation result received from the server (200). In an embodiment, the electronic device (100) can output at least one of an audio signal and a video signal corresponding to the recommendation result.

[0390] In an embodiment, the electronic device (100) may receive, as a recommendation result, at least one piece of information about a channel or content that the electronic device (100) can output. The information about the channel or content may include metadata, a preview video, an advertisement video, a thumbnail, etc. corresponding to the channel or content.

[0391] The electronic device (100) can identify a recommended channel or recommended content based on metadata corresponding to the recommended channel or recommended content transmitted by the server (200), and output images, videos, and related information of the identified recommended channel or recommended content through a partial screen included in the multi-view screen. Alternatively, the electronic device (100) can output a thumbnail for the recommended channel or recommended content transmitted by the server (200) through a partial screen.

[0392] In an embodiment, the electronic device (100) may receive, as a recommendation result, an action that the electronic device (100) can perform. The action that the electronic device (100) can perform may include information that guides the electronic device (100) to perform an action for a Multi Device Experience (MDE).

[0393] In an embodiment, the operation that the electronic device (100) can perform may include, for example, information guiding the download of a specific app when the recommended result recommended by the server (200) is content that can only be obtained by executing a specific app and the specific app is not currently downloaded on the electronic device (100). In addition, for example, when the recommended result recommended by the server (200) is content provided by a specific OTT service provider and it is unclear whether the user is a member who can use the specific OTT service, the recommended result may include information guiding the user to log in to use the specific OTT service.

[0394] Furthermore, if a specific app is a paid app or the user is not a member who can use a specific OTT service, the server (200) may include in the recommendation results information guiding the user to perform additional actions such as payment to download the paid app, or information guiding the user to perform additional actions such as membership registration or subscription to use the OTT service.

[0395] The server (200) may include in the recommendation results information that guides actions related to additional functions that the electronic device (100) can perform, such as a link to an Internet shopping mall address where a user can purchase a specific product that the user may be interested in, or information that guides information related to payment or subscription for a paid service to provide paid content or paid OTT service that the electronic device (100) can output.

[0396] Referring to FIG. 16, the electronic device (100) can output the recommendation result in the form of a video signal through an interface screen (520). In an embodiment, the electronic device (100) can display the interface screen (520) on a portion of the display area.

[0397] In an embodiment, the video signal corresponding to the recommendation result may include information guiding actions related to additional functions that the electronic device (100) can perform. For example, as illustrated in FIG. 16, the video signal corresponding to the recommendation result may include a phrase guiding a subscription to an OTT service called YHW.

[0398] Additionally, in an embodiment, a video signal corresponding to a recommendation result may include shortcut information (1620) that guides the user to immediately use a specific service. The shortcut information (1620) may include information such as a QR code, an internet address on an app / website, etc.

[0399] A user can view information guiding actions related to additional functions included in the recommended results output by the electronic device (100) and perform actions according to the guidance. For example, a user can easily perform additional functions, i.e., subscribe to an OTT service, by utilizing shortcut information (1620) included in the recommended results.

[0400] FIG. 17 is a flowchart illustrating an operation method of an electronic device (100) and a server (200) according to an embodiment.

[0401] In an embodiment, the electronic device (100) can obtain context information including at least one of context information, user information, characteristic information, and surrounding context information of the currently output content.

[0402] In an embodiment, the electronic device (100) may obtain metadata corresponding to content, including at least one of metadata corresponding to content currently being output, metadata corresponding to content that the electronic device (100) is not currently outputting but can output, and programming information.

[0403] In an embodiment, the electronic device (100) may generate request information from at least one of contextual information and metadata corresponding to the content (step 1710).

[0404] In an embodiment, the request information may include text information in the form of an input dialogue. In an embodiment, the request information may further include additional information along with the text information.

[0405] In an embodiment, the electronic device (100) may transmit request information to the server (200) (step 1720).

[0406] In an embodiment, the server (200) may receive request information from the electronic device (100).

[0407] In an embodiment, the server (200) can obtain a recommendation result based on the request information (step 1730).

[0408] In an embodiment, the server (200) may obtain a recommendation result based on request information based on a super-large-scale generative AI model. In an embodiment, the server (200) may obtain a recommendation result that includes at least one of information about actions that the electronic device (100) can perform and information about channels or content that the electronic device (100) can output. In an embodiment, the server (200) may obtain a recommendation result that includes a reason for the recommendation.

[0409] In an embodiment, the server (200) may transmit the recommendation result to the electronic device (100) (step 1740).

[0410] FIG. 18 is a flowchart illustrating a method in which an electronic device (100) and a server (200) operate according to an embodiment.

[0411] Referring to FIG. 18, in an embodiment, the server (200) may transmit the recommendation result to the electronic device (100) (step 1810).

[0412] In an embodiment, the electronic device (100) can receive a recommendation result from the server (200).

[0413] In an embodiment, the electronic device (100) may output the recommendation results through a multi-view screen (step 1820). In an embodiment, the multi-view screen may include multiple partial screens.

[0414] In an embodiment, the electronic device (100) may request the server (200) in advance the next screen information to be output when a first area among a plurality of partial screens is selected.

[0415] In an embodiment, the server (200) may obtain a recommendation result corresponding to the following screen information in response to a request from the electronic device (100) (step 1840).

[0416] The next screen may include, for example, a screen in which, when a user input for selecting the first area is received while the first content is being output in the first area other than the center area among multiple areas included in the multi-view screen, the first content that was being output in the first area is output in the center area other than the first area, and information about new content or new channels is output through areas above, below, left, and right of the center area. Recommendation results corresponding to the next screen information may include metadata, thumbnail images, etc. of the content or channel to be output through the upper, lower, left, and right portions of the center portion screen.

[0417] In an embodiment, the server (200) may transmit recommendation results corresponding to the following screen information to the electronic device (100) (step 1850).

[0418] In an embodiment, the electronic device (100) may receive a recommendation result corresponding to the following screen information from the server (200) and temporarily store it in a temporary memory or the like.

[0419] In an embodiment, the electronic device (100) may receive a control signal for selecting a first area (step 1860). That is, the electronic device (100) may receive a control signal for selecting a first area of ​​a multi-view screen or first content, which is screen information currently output in the first area, from the control device (50) (step 1860).

[0420] In an embodiment, the electronic device (100) may, in response to receiving a first region selection control signal for selecting screen information currently output in the first region, reconfigure the multi-view screen so that the first content output in the first region is output in the center screen. In an embodiment, in response to receiving the first region selection control signal, the electronic device (100) may read out a recommendation result corresponding to the next screen information received by requesting the server (200) in advance, and may, based on the same, immediately output the next screen based on the recommendation result corresponding to the next screen information (step 1870).

[0421] For example, the electronic device (100) can read metadata, thumbnail images, or information about the content or channels to be output through the upper, lower, left, and right partial screens of the multi-view screen from a temporary memory in the multi-view screen to be output as the next screen. The electronic device (100) can obtain content to be output through the partial screen based on the thumbnail image or metadata, and output the content through the partial screen of the multi-view screen.

[0422] However, this is just one example, and the server (200), rather than the electronic device (100), may generate the next screen as a multi-view screen and transmit it to the electronic device (100). In this case, the electronic device (100) may store the next screen received from the server (200) in a temporary memory, and then, when receiving the first area selection control signal, may immediately read out and output the next screen.

[0423] In an embodiment, the electronic device (100) may request in advance from the server (200) information about the next screen to be output when the second to fourth regions are selected, in preparation for the case where each of a plurality of other partial screens included in the multi-view screen currently being output, such as the second region, the third region, and the fourth region, are selected in addition to the first region.

[0424] The server (200) can obtain the next screen information to be output when the second to fourth areas are selected based on the generated AI model and transmit it to the electronic device (100). The electronic device (100) can temporarily store the next screen information in a temporary memory.

[0425] Thereafter, when the user selects any one of the second to fourth areas, the electronic device (100) can immediately read out the recommendation result corresponding to the selected area from the temporary memory and output a screen based on the recommendation result.

[0426] The method and device for operating an electronic device according to some embodiments may also be implemented in the form of a recording medium containing computer-executable instructions, such as program modules executed by a computer. Computer-readable media may be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Furthermore, computer-readable media may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information, such as computer-readable instructions, data structures, program modules, or other data. Communication media typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave, or other transport mechanism, and includes any information delivery media.

[0427] In addition, the electronic device and the operating method thereof according to the above-described embodiment of the present disclosure may be implemented as a computer program product including a computer-readable recording medium / storage medium having recorded thereon a program for implementing the operating method of the electronic device, the program including the step of transmitting request information including text information in the form of an input dialogue obtained from at least one of contextual information and metadata corresponding to content to a server, and the step of receiving a recommendation result based on the request information from the server.

[0428] A device-readable storage medium may be provided in the form of a non-transitory storage medium. Here, the term "non-transitory storage medium" simply means a tangible device that does not contain signals (e.g., electromagnetic waves). This term does not distinguish between cases where data is permanently stored in the storage medium and cases where data is temporarily stored. For example, a "non-transitory storage medium" may include a buffer in which data is temporarily stored.

[0429] According to one embodiment, the method according to various embodiments disclosed in the present document may be provided as included in a computer program product. The computer program product may be traded as a product between a seller and a buyer. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or may be distributed online (e.g., downloaded or uploaded) through an application store or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product (e.g., a downloadable app) may be temporarily stored or temporarily generated in a machine-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or an intermediary server.

Claims

1. In an electronic device (100), A memory (120) storing one or more instructions; and comprising one or more processors (110) for executing one or more instructions stored in the memory; One or more of the above processors Transmitting request information including text information in the form of input dialogue obtained from at least one of context information and metadata corresponding to the content to the server (200), An electronic device that receives recommendation results based on the request information from the server.

2. In the first paragraph, the one or more processors An electronic device that transmits the request information to the server when a control signal corresponding to the execution of one of the functions of a search function, a channel change function, and a movement to the main screen function is received.

3. An electronic device according to claim 1 or 2, wherein the situation information includes at least one of context information of currently output content, user information, characteristic information, and surrounding situation information.

4. An electronic device according to any one of claims 1 to 3, wherein the metadata corresponding to the content includes at least one of metadata corresponding to content currently being output, metadata corresponding to content that the electronic device can output, and programming information.

5. In any one of paragraphs 1 to 4, the recommendation result received from the server includes at least one of information corresponding to an executable action, an outputtable channel, or content, The one or more processors control at least one of an audio signal and a video signal corresponding to the recommendation result received from the server to be output, An electronic device wherein the above video signal contains text information in the form of a conversation.

6. In the fifth paragraph, if the recommendation result is output as a multi-view screen, An electronic device in which the one or more processors receive next screen information to be output in the first area included in the multi-view screen from the server, and, when a user input for selecting screen information currently output in the first area is received, control the device to output a screen based on a recommendation result corresponding to the next screen information received from the server.

7. In the server (200), A memory (220) storing one or more instructions; and comprising one or more processors (210) for executing one or more instructions stored in the memory; One or more of the above processors, When request information including text information in the form of input dialogue is received from an electronic device (100), A server that transmits a recommendation result including a recommendation reason corresponding to the request information obtained based on one or more neural networks to the electronic device.

8. In paragraph 7, The one or more processors obtain at least one of additional information and user information, A server that inputs at least one of the additional information and the user information together with the request information into the one or more neural networks, thereby obtaining the recommendation result from the one or more neural networks.

9. In paragraph 8, the additional information includes at least one of metadata corresponding to the content, popular content, curated content, content selected according to a specific theme, and KPI (Key Performance Index) information. A server, wherein the user information includes at least one of setting information, user profile, user viewing history information, and channel switching history information.

10. In any one of clauses 7 to 9, the one or more processors A server that obtains candidate information with a high priority among candidate information related to a recommendation reason from one or more neural networks as a recommendation reason.

11. In any one of clauses 7 to 10, the one or more neural networks include a plurality of theme-specific neural networks, A server in which the one or more processors input information corresponding to the plurality of theme-specific neural networks obtained based on the request information into the plurality of theme-specific neural networks, and obtain the recommendation result from at least one of the plurality of theme-specific neural networks.

12. A server according to any one of clauses 7 to 11, wherein the recommendation result further includes information corresponding to additional functions that the electronic device can perform.

13. In the method of operating an electronic device, A step of transmitting request information including text information in the form of an input dialogue obtained from at least one of context information and metadata corresponding to the content to the server; and A method of operating an electronic device, comprising the step of receiving a recommendation result based on the request information from the server.

14. In the 13th paragraph, the step of transmitting the request information to the server An operating method of an electronic device, comprising the step of transmitting the request information to the server when a control signal corresponding to the execution of one of a search function, a channel change function, and a function of moving to a main screen is received.

15. In the 13th or 14th clause, further comprising a step of outputting at least one of an audio signal and a video signal corresponding to the recommendation result, A method of operating an electronic device, wherein the video signal includes text information in the form of a conversation.

Images