Electronic apparatus of providing message about health care and operating method thereof, and recording medium

Abstract

An electronic apparatus according to one embodiment may comprise a processor, and a memory for storing instructions. When executed by the processor, the instructions can instruct the electronic apparatus to: receive sensor data about a user of an external electronic device from the external electronic device; search for whether an insight information generation condition is satisfied; acquire, on the basis of a determination that the insight information generation condition is satisfied, a unique ID corresponding to the insight information generation condition; acquire a common prompt corresponding to the unique ID; acquire, on the basis of the sensor data, a data prompt corresponding to the unique ID so as to generate an input prompt; and input the input prompt into an artificial neural network model so as to generate insight information.

Description

Electronic device for providing health care-related messages, method of operation thereof, and recording medium

[0001] Embodiments of the present invention relate to an electronic device providing a message regarding health care, a method of operating the same, and a recording medium.

[0002] Electronic devices equipped with health information capabilities are becoming increasingly widespread. These devices can provide health information to users through generative language models. Generative language models are artificial intelligence models trained on large-scale text data. Generative language models can be used for text generation and question-answering. For example, a generative language model may include a large language model (LLM). Electronic devices can generate responses (e.g., health management messages) based on user queries.

[0003] The above information may be provided as background art to aid in understanding the present disclosure. No claim or determination is made as to whether any of the above-described matters constitute prior art related to the present disclosure.

[0004] An electronic device according to one embodiment may include at least one processor including processing circuitry and a memory storing instructions. When the instructions are individually or collectively executed by the at least one processor, the electronic device may receive sensor data about a user of the external electronic device from an external electronic device. When the instructions are individually or collectively executed by the at least one processor, the electronic device may determine whether an insight information generation condition is satisfied. When the instructions are individually or collectively executed by the at least one processor, the electronic device may obtain a unique ID corresponding to the insight information generation condition based on a determination that the insight information generation condition is satisfied. When the instructions are individually or collectively executed by the at least one processor, the electronic device may obtain a common prompt corresponding to the unique ID. When the above commands are individually or collectively executed by the at least one processor, the electronic device may be caused to, when executed by the processor, obtain a data prompt corresponding to the unique ID based on the sensor data and generate an input prompt. When the above commands are individually or collectively executed by the at least one processor, the electronic device may be caused to input the input prompt into an artificial neural network model to generate insight information.

[0005] An operating method of an electronic device according to one embodiment may include an operation of receiving sensor data about a user of an external electronic device from an external electronic device. The operating method may include an operation of determining whether an insight information generation condition is satisfied. The operating method may include an operation of obtaining a unique ID corresponding to the insight information generation condition based on a determination that the insight information generation condition is satisfied. The operating method may include an operation of obtaining a common prompt corresponding to the unique ID. The operating method may include an operation of obtaining a data prompt corresponding to the unique ID based on the sensor data and generating an input prompt. The operating method may include an operation of inputting the input prompt to an artificial neural network model to generate insight information.

[0006] FIG. 1 is a block diagram of an electronic device within a network environment according to one embodiment.

[0007] FIG. 2 is an example of a block diagram of an electronic device according to one embodiment.

[0008] Figure 3 is a block diagram of a health app service provision module (220) according to one embodiment.

[0009] FIG. 4 is a drawing for explaining the structure of an input prompt according to one embodiment.

[0010] FIG. 5 is a diagram illustrating a method for verifying insight information according to one embodiment.

[0011] Figure 6 is a flowchart illustrating a method for generating insight information according to one embodiment.

[0012] FIG. 7 is a flowchart illustrating an example of an operation for generating an insight message according to one embodiment.

[0013] FIG. 8 is a flowchart illustrating an example of an operation for generating an input prompt according to one embodiment.

[0014] FIG. 9 is a diagram illustrating an example of a screen on which insight information is displayed on an electronic device (200) according to one embodiment.

[0015] FIG. 10 is a diagram illustrating an example of a screen provided to obtain information of interest to a user according to one embodiment.

[0016] FIG. 11 is a diagram illustrating a method for generating insight information based on a multi-modal artificial neural network according to one embodiment.

[0017] FIG. 12 is an example of a screen displaying insight information generated using a multi-modal artificial neural network according to one embodiment.

[0018] Fig. 13 illustrates an example of an electronic device according to one embodiment.

[0019] Hereinafter, embodiments will be described in detail with reference to the attached drawings. In the description with reference to the attached drawings, identical components are assigned the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted.

[0020] FIG. 1 is a block diagram of an electronic device (101) within a network environment (100) according to an embodiment. Referring to FIG. 1, in the network environment (100), the electronic device (101) may communicate with the electronic device (102) via a first network (198) (e.g., a short-range wireless communication network), or may communicate with at least one of the electronic device (104) or the server (108) via a second network (199) (e.g., a long-range wireless communication network). According to an embodiment, the electronic device (101) may communicate with the electronic device (104) via the server (108). According to one embodiment, the electronic device (101) may include a processor (120), a memory (130), an input module (150), an audio output module (155), a display module (160), an audio module (170), a sensor module (176), an interface (177), a connection terminal (178), a haptic module (179), a camera module (180), a power management module (188), a battery (189), a communication module (190), a subscriber identification module (196), or an antenna module (197). In some embodiments, the electronic device (101) may omit at least one of these components (e.g., the connection terminal (178)), or may have one or more other components added. In some embodiments, some of these components (e.g., the sensor module (176), the camera module (180), or the antenna module (197)) may be integrated into one component (e.g., the display module (160)).

[0021] The processor (120) may control at least one other component (e.g., a hardware or software component) of the electronic device (101) connected to the processor (120) by executing, for example, software (e.g., a program (140)), and may perform various data processing or calculations. According to one embodiment, as at least a part of the data processing or calculation, the processor (120) may store a command or data received from another component (e.g., a sensor module (176) or a communication module (190)) in a volatile memory (132), process the command or data stored in the volatile memory (132), and store the resulting data in a non-volatile memory (134). According to one embodiment, the processor (120) may include a main processor (121) (e.g., a central processing unit or an application processor) or a secondary processor (123) (e.g., a graphics processing unit, a neural processing unit (NPU), an image signal processor, a sensor hub processor, or a communication processor) that can operate independently or together therewith. For example, if the electronic device (101) includes a main processor (121) and a secondary processor (123), the secondary processor (123) may be configured to use less power than the main processor (121) or to be specialized for a specified function. The secondary processor (123) may be implemented separately from the main processor (121) or as a part thereof.

[0022] The auxiliary processor (123) may control at least a portion of functions or states associated with at least one component (e.g., a display module (160), a sensor module (176), or a communication module (190)) of the electronic device (101), for example, on behalf of the main processor (121) while the main processor (121) is in an inactive (e.g., sleep) state, or together with the main processor (121) while the main processor (121) is in an active (e.g., application execution) state. In one embodiment, the auxiliary processor (123) (e.g., an image signal processor or a communication processor) may be implemented as a part of another functionally related component (e.g., a camera module (180) or a communication module (190)). In one embodiment, the auxiliary processor (123) (e.g., a neural network processing unit) may include a hardware structure specialized for processing artificial intelligence models. The artificial intelligence models may be generated through machine learning. This learning can be performed, for example, on the electronic device (101) itself where the artificial intelligence model is executed, or can be performed through a separate server (e.g., server (108)). The learning algorithm can include, for example, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but is not limited to the examples described above. The artificial intelligence model can include multiple artificial neural network layers.The artificial neural network may be one of a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a restricted Oltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), a deep Q-network, or a combination of two or more of the above, but is not limited to the examples described above. In addition to, or alternatively to, a hardware structure, an artificial intelligence model may include a software structure.

[0023] The memory (130) can store various data used by at least one component (e.g., processor (120) or sensor module (176)) of the electronic device (101). The data can include, for example, software (e.g., program (140)) and input data or output data for commands related thereto. The memory (130) can include volatile memory (132) or non-volatile memory (134).

[0024] The program (140) may be stored as software in the memory (130) and may include, for example, an operating system (142), middleware (144), or an application (146).

[0025] The input module (150) can receive commands or data to be used in a component of the electronic device (101) (e.g., a processor (120)) from an external source (e.g., a user) of the electronic device (101). The input module (150) can include, for example, a microphone, a mouse, a keyboard, a key (e.g., a button), or a digital pen (e.g., a stylus pen).

[0026] The audio output module (155) can output audio signals to the outside of the electronic device (101). The audio output module (155) can include, for example, a speaker or a receiver. The speaker can be used for general purposes, such as multimedia playback or recording playback. The receiver can be used to receive incoming calls. According to one embodiment, the receiver can be implemented separately from the speaker or as part of the speaker.

[0027] The display module (160) can visually provide information to an external device (e.g., a user) of the electronic device (101). The display module (160) may include, for example, a display, a holographic device, or a projector, and a control circuit for controlling the device. According to one embodiment, the display module (160) may include a touch sensor configured to detect a touch, or a pressure sensor configured to measure the intensity of a force generated by the touch. The display module (160) may be implemented with an illustrative foldable structure and / or a rollable structure. For example, the size of the display screen of the display module (160) may be reduced when folded, and may be expanded when unfolded.

[0028] The audio module (170) can convert sound into an electrical signal, or vice versa, convert an electrical signal into sound. According to one embodiment, the audio module (170) can acquire sound through the input module (150), output sound through the sound output module (155), or an external electronic device (e.g., electronic device (102)) (e.g., speaker or headphone) directly or wirelessly connected to the electronic device (101).

[0029] The sensor module (176) can detect the operating status (e.g., power or temperature) of the electronic device (101) or the external environmental status (e.g., user status) and generate an electrical signal or data value corresponding to the detected status. According to one embodiment, the sensor module (176) can include, for example, a gesture sensor, a gyro sensor, a barometric pressure sensor, a magnetic sensor, an acceleration sensor, a grip sensor, a proximity sensor, a color sensor, an IR (infrared) sensor, a biometric sensor, a temperature sensor, a humidity sensor, or an illuminance sensor.

[0030] The interface (177) may support one or more designated protocols that may be used to directly or wirelessly connect the electronic device (101) with an external electronic device (e.g., the electronic device (102)). In one embodiment, the interface (177) may include, for example, a high definition multimedia interface (HDMI), a universal serial bus (USB) interface, an SD card interface, or an audio interface.

[0031] The connection terminal (178) may include a connector through which the electronic device (101) may be physically connected to an external electronic device (e.g., electronic device (102)). According to one embodiment, the connection terminal (178) may include, for example, an HDMI connector, a USB connector, an SD card connector, or an audio connector (e.g., a headphone connector).

[0032] A haptic module (179) can convert electrical signals into mechanical stimuli (e.g., vibration or movement) or electrical stimuli that a user can perceive through tactile or kinesthetic sensations. According to one embodiment, the haptic module (179) can include, for example, a motor, a piezoelectric element, or an electrical stimulation device.

[0033] The camera module (180) can capture still images and videos. According to one embodiment, the camera module (180) may include one or more lenses, image sensors, image signal processors, or flashes.

[0034] The power management module (188) can manage power supplied to the electronic device (101). According to one embodiment, the power management module (188) can be implemented as, for example, at least a part of a power management integrated circuit (PMIC).

[0035] A battery (189) may power at least one component of the electronic device (101). In one embodiment, the battery (189) may include, for example, a non-rechargeable primary battery, a rechargeable secondary battery, or a fuel cell.

[0036] The communication module (190) may support the establishment of a direct (e.g., wired) communication channel or a wireless communication channel between the electronic device (101) and an external electronic device (e.g., electronic device (102), electronic device (104), or server (108)), and the performance of communication through the established communication channel. The communication module (190) may operate independently from the processor (120) (e.g., application processor) and may include one or more communication processors that support direct (e.g., wired) communication or wireless communication. According to one embodiment, the communication module (190) may include a wireless communication module (192) (e.g., a cellular communication module, a short-range wireless communication module, or a global navigation satellite system (GNSS) communication module) or a wired communication module (194) (e.g., a local area network (LAN) communication module, or a power line communication module). Among these communication modules, the corresponding communication module can communicate with an external electronic device (104) via a first network (198) (e.g., a short-range communication network such as Bluetooth, wireless fidelity (WiFi) direct, or infrared data association (IrDA)) or a second network (199) (e.g., a long-range communication network such as a legacy cellular network, a 5G network, a next-generation communication network, the Internet, or a computer network (e.g., a LAN or WAN)). These various types of communication modules can be integrated into a single component (e.g., a single chip) or implemented as multiple separate components (e.g., multiple chips). The wireless communication module (192) can verify or authenticate the electronic device (101) within a communication network such as the first network (198) or the second network (199) by using subscriber information (e.g., an international mobile subscriber identity (IMSI)) stored in the subscriber identification module (196).

[0037] The wireless communication module (192) can support 5G networks and next-generation communication technologies following the 4G network, such as NR access technology (new radio access technology). The NR access technology can support high-speed transmission of high-capacity data (eMBB (enhanced mobile broadband)), minimization of terminal power and connection of multiple terminals (mMTC (massive machine type communications)), or high reliability and low latency (URLLC (ultra-reliable and low-latency communications)). The wireless communication module (192) can support, for example, a high-frequency band (e.g., mmWave band) to achieve a high data transmission rate. The wireless communication module (192) can support various technologies for securing performance in a high-frequency band, such as beamforming, massive multiple-input and multiple-output (MIMO), full dimensional MIMO (FD-MIMO), array antenna, analog beam-forming, or large scale antenna. The wireless communication module (192) can support various requirements specified in the electronic device (101), an external electronic device (e.g., the electronic device (104)), or a network system (e.g., the second network (199)). According to one embodiment, the wireless communication module (192) can support a peak data rate (e.g., 20 Gbps or more) for eMBB realization, a loss coverage (e.g., 164 dB or less) for mMTC realization, or a U-plane latency (e.g., 0.5 ms or less for downlink (DL) and uplink (UL), or 1 ms or less for round trip) for URLLC realization.

[0038] The antenna module (197) can transmit or receive signals or power to or from an external device (e.g., an external electronic device). According to one embodiment, the antenna module (197) may include an antenna including a radiator formed of a conductor or a conductive pattern formed on a substrate (e.g., a PCB). According to one embodiment, the antenna module (197) may include a plurality of antennas (e.g., an array antenna). In this case, at least one antenna suitable for a communication method used in a communication network, such as the first network (198) or the second network (199), may be selected from the plurality of antennas, for example, by the communication module (190). A signal or power may be transmitted or received between the communication module (190) and an external electronic device via the selected at least one antenna. According to some embodiments, in addition to the radiator, another component (e.g., a radio frequency integrated circuit (RFIC)) may be additionally formed as a part of the antenna module (197).

[0039] In one embodiment, the antenna module (197) may form a mmWave antenna module. In one embodiment, the mmWave antenna module may include a printed circuit board, an RFIC disposed on or adjacent to a first side (e.g., a bottom side) of the printed circuit board and capable of supporting a designated high-frequency band (e.g., a mmWave band), and a plurality of antennas (e.g., an array antenna) disposed on or adjacent to a second side (e.g., a top side or a side side) of the printed circuit board and capable of transmitting or receiving signals in the designated high-frequency band.

[0040] At least some of the above components can be interconnected and exchange signals (e.g., commands or data) with each other via a communication method between peripheral devices (e.g., a bus, GPIO (general purpose input and output), SPI (serial peripheral interface), or MIPI (mobile industry processor interface)).

[0041] According to one embodiment, commands or data may be transmitted or received between the electronic device (101) and an external electronic device (104) via a server (108) connected to a second network (199). Each of the external electronic devices (102 or 104) may be the same or a different type of device as the electronic device (101). According to one embodiment, all or part of the operations executed in the electronic device (101) may be executed in one or more of the external electronic devices (102, 104, or 108). For example, when the electronic device (101) is to perform a certain function or service automatically or in response to a request from a user or another device, the electronic device (101) may, instead of or in addition to executing the function or service by itself, request one or more external electronic devices to perform the function or at least a part of the service. One or more external electronic devices that receive the request may execute at least a portion of the requested function or service, or an additional function or service related to the request, and transmit the result of the execution to the electronic device (101). The electronic device (101) may process the result as is or additionally and provide it as at least a portion of a response to the request. For this purpose, cloud computing, distributed computing, mobile edge computing (MEC), or client-server computing technology may be used, for example. The electronic device (101) may provide an ultra-low latency service by using distributed computing or mobile edge computing, for example. In another embodiment, the external electronic device (104) may include an Internet of Things (IoT) device. The server (108) may be an intelligent server utilizing machine learning and / or a neural network. According to one embodiment, the external electronic device (104) or the server (108) may be included in the second network (199).The electronic device (101) can be applied to intelligent services (e.g., smart home, smart city, smart car, or healthcare) based on 5G communication technology and IoT-related technology.

[0042] FIG. 2 is an example of a block diagram of an electronic device according to one embodiment.

[0043] Referring to FIG. 2, according to one embodiment, an electronic device (200) (e.g., the electronic device (101) of FIG. 1) may include a processor (210) (e.g., the processor (120) of FIG. 1) and a memory (240) (e.g., the memory (130) of FIG. 1). The health app service provision module (220) and the artificial neural network model selection module (230) may be configured as one or more of a program code including instructions that can be executed by the processor (210), an application, an algorithm, a routine, a set of instructions, or an artificial intelligence learning model that can be stored in the memory (240). One or more of the health app service provision module (220) and the artificial neural network model selection module (230) may be implemented as hardware and / or a combination of hardware and software. The memory (240) can store the on-device artificial neural network model (250), and according to an embodiment, a memory (not shown) other than the memory (240) included in the electronic device (200) can store the on-device artificial neural network model (250). Hereinafter, the term "module" may mean, for example, a unit including one or a combination of two or more of hardware, software, or firmware. The term "module" may be used interchangeably with terms such as unit, logic, logical block, component, or circuit. The "module" may be the smallest unit of an integrally configured component or a part thereof. The "module" may also be the smallest unit performing one or more functions or a part thereof. The "module" may be implemented mechanically or electronically.For example, a "module" may include at least one of an application-specific integrated circuit (ASIC) chip, field-programmable gate arrays (FPGAs), or programmable-logic device, known or to be developed in the future, that performs certain operations.

[0044] According to one embodiment, the memory (240) may include a status data DB (not shown) that stores user status data. The user status data may include information related to the user's biometrics and / or lifestyle. For example, the user status data may include health information data (e.g., the user's current health status, symptoms, and discomforts), information data related to lifestyle patterns (e.g., eating habits, meal times, sleep patterns, and daily activities), exercise record data (e.g., daily activity amount, type of exercise, exercise time, and amount of calories burned during exercise), medical record data (e.g., electronic medical record (EMR) and electronic health record (EHR)), and biometric information data (e.g., information obtained from a sensor (e.g., a biometric sensor)). The biometric information data may also be referred to as sensor data.

[0045] According to one embodiment, the electronic device (200) may be implemented as at least one of a smartphone, a tablet personal computer, a mobile phone, a speaker (e.g., an AI speaker), a video phone, an e-book reader, a desktop personal computer, a laptop personal computer, a netbook computer, a workstation, a server, a personal digital assistant (PDA), a portable multimedia player (PMP), an MP3 player, a mobile medical device, or a camera.

[0046] According to one embodiment, an external electronic device (260) (e.g., electronic device (102) and / or electronic device (104) of FIG. 1) may include a processor (261), a memory (262), a communication module (263), and sensor(s) (264). The external electronic device (260) may include a wearable device. The form of the external electronic device (260) may include a method worn on the wrist (e.g., a smart watch, a smart band), a method worn on the finger like a ring (e.g., a smart ring, a smart glove), a method worn on the head including the ear (e.g., a head mounted display (HMD), an extended reality (XR) device, smart glasses, a smart headset, a smart hat), a method inserted into the ear (e.g., a hearable device, TWS-true wireless stereo, OWS-open wireless stereo), a method attached to the body (e.g., a patch, a smart tattoo), a method worn on clothes (e.g., a smart garment, a smart pin), a method worn around the neck, a method worn on the foot (e.g., a smart shoe, a smart pad), and is not limited to the examples described above.

[0047] According to one embodiment, the external electronic device (260) may be connected to the electronic device (200) via a communication module (263). For example, the external electronic device (260) may be connected to the electronic device (200) wirelessly. For example, the electronic device (200) may be connected to the external electronic device (260) via short-range wireless communication (e.g., Bluetooth, Wi-Fi). The method by which the electronic device (200) and the external electronic device (260) are connected is not limited thereto, and the external electronic device (260) may be connected to the electronic device (200) via a wire. There may be a plurality of external electronic devices (260), and the plurality of external electronic devices may be connected to each other between the electronic device (200) and other external electronic devices.

[0048] According to one embodiment, an external electronic device (260) can obtain sensor data about a user through sensor(s) (264) and transmit the obtained sensor data to the electronic device (200). For example, the external electronic device (260) can be a smart watch, and the smart watch can obtain biometric information (e.g., heart rate information) of the user and transmit the obtained biometric information to the electronic device (200). The electronic device (200) can store the sensor data in the memory (240). As described above, the sensor data can be stored in a status data DB.

[0049] According to one embodiment, the electronic device (200) may be connected to a message management server (270) using short-range wireless communication (e.g., Wi-Fi) or mobile communication (e.g., 4G, 5G). The electronic device (200) may receive a unique ID (ID) corresponding to a condition for generating a message from the message management server (270) and provide information so that insight information matching the condition is generated. The insight information is information for inducing an understanding of health data and changes in user behavior and awareness, and may be, for example, a health management message. However, the insight information is not limited to the form of a message. For example, the insight information may be provided in the form of a graph, animation, or icon. The message management server (270) may include a prompt updater (271), and the operation of the prompt updater (271) is described in detail below with reference to FIGS. 3 and 7 .

[0050] According to one embodiment, the electronic device (200) may use sensor data received from an external electronic device (260) to construct an input prompt that meets the corresponding user conditions provided by the message management server (270), and input the input prompt into a pre-trained artificial neural network model (e.g., a large language model (LLM)) to generate insight information. The input prompt may refer to an initial sentence or question input into the artificial neural network model. For example, if the artificial neural network model is a large language model (LLM), the input prompt may be composed of natural language text and may determine the direction of the text to be generated by the large language model, and may help the large language model provide appropriate information and respond fluently. The large language model may be referred to as a language model composed of an artificial neural network that has been pre-trained with a massive amount of text data. The large language model may include more than 10 times as many parameters (e.g., more than 100 billion parameters) as a conventional general language model. Large-scale language models can utilize a transformer artificial neural network architecture based on an attention mechanism. An attention mechanism is a technique that helps an AI model focus on important parts of input data. The attention mechanism can be used to predict output data by predicting the degree to which at least a portion of time-series input data (e.g., input data such as voice or video, or input data from a neural network layer) contributes to the intermediate or final output of the neural network.The recurrent neural network (RNN) structure, which sequentially processes each element of a sequence, has poor prediction performance when there is information dependency between long time series distances, but the attention mechanism can consider information dependency between long time series distances by controlling the degree of weight concentration within the overall (or partial) context of the input data.

[0051] For example, a large-scale language model may include a transformer with an encoder-decoder structure. The encoder processes input data and outputs compressed information (e.g., an attention mechanism), and the decoder processes the compressed information and outputs token-based output data. Each encoder and decoder may include an independent attention network, and a cross-attention network connecting the encoder and decoder may be included.

[0052] For example, large-scale language models can be trained in two stages: pre-training and fine-tuning. Pre-training involves training a large-scale language model to process massive amounts of text data and acquire general linguistic knowledge. For example, this could involve self-supervised learning, such as predicting the next word using a sequence of previous words in a text sequence. Fine-tuning involves training a large-scale language model to fit a specific domain (e.g., chatbot, translation, summarization, Q&A) or task. This can be further supervised (or adaptive) based on a pre-trained model using a dataset tailored to the domain's purpose. Large-scale language models can perform tasks with text inputs containing natural language, called prompts. For example, large-scale language models can include BERT (bidirectional encoder representations from transformer) and GPT (generative pre-trained transformer). The term "LLM" can refer to the neural network model itself, but can also refer to the model of an LLM-based application (e.g., chatbot, translation, summarization, text classification, sentence generation). For example, an LLM-based chatbot such as chatGPT can also be referred to as an LLM. "LLM" can also include an inference engine that utilizes the LLM neural network model. For example, "entering an input prompt into an LLM" can be referred to as "entering an input prompt into an LLM-based inference engine."

[0053] A method for generating insight information according to one embodiment defines the structure of an input prompt and a method for obtaining the input prompt for generating accurate and personalized insight information. The structure of the input prompt and the method for obtaining the input prompt are described in detail below with reference to FIG. 4.

[0054] According to one embodiment, the electronic device (200) can collect and record data transmitted from an external electronic device (260). The electronic device (200) may have an application program installed thereon, or a corresponding module performing the same function may be integrated into the platform of the electronic device (200). The application program may include a health app service provision module (220) that provides practical services to the user, and an artificial neural network model selection module (230) that can select which model to use to generate insight information.

[0055] According to one embodiment, the health app service providing module (220) can provide a practical service related to the user's health using sensor data transmitted from an external electronic device (260). The health app service providing module (220) can calculate specific health indicators and provide the corresponding information to the user in various ways, generate insight information, query and provide data defined within an input prompt required for generating the insight information, and provide an operation for displaying the generated insight information to the user. The specific operation of the health app service providing module (220) is described in detail below with reference to FIG. 3.

[0056] According to one embodiment, the artificial neural network model selection module (230) can determine one of the on-device artificial neural network model (250) and the server-based artificial neural network model (281) as input to the input prompt.

[0057] According to one embodiment, the on-device artificial neural network model (250) and the server-based artificial neural network model (281) may be artificial neural network models trained to generate insight information. For example, the on-device artificial neural network model (250) and the server-based artificial neural network model (281) may be language models trained to receive an input prompt and generate an insight message. The on-device artificial neural network model (250) is a model that can process information on the electronic device (200) itself without a separate server, and can be used at any time even when the electronic device (200) is not connected to a network. The on-device artificial neural network model (250) can be operated as an application program within the electronic device (200), and can generate insight information without linking to other external modules or artificial neural network models by using an input prompt for a health app service transmitted through the artificial neural network model selection module (230). An application including an on-device artificial neural network model (250) may include metadata, which may be used to provide support information, such as supported languages ​​and supported functions, to the party requesting message generation. The on-device artificial neural network model may be fine-tuned to suit more detailed domains or services.

[0058] According to one embodiment, the electronic device (200) may be connected to an artificial neural network model server (280) using short-range wireless communication (e.g., Wi-Fi) or mobile communication (e.g., 4G, 5G). The artificial neural network model server (280) may include a server-based artificial neural network model (281). The server-based artificial neural network model (281) may be referred to as a cloud-based artificial neural network model and may generate insight information using a language model having a larger number of parameters than the on-device artificial neural network model (250). For example, the on-device artificial neural network model (250) may be a small language model (SML), and the server-based artificial neural network model (281) may be a large language model. However, the above example means that the on-device artificial neural network model (250) can have a smaller number of parameters than the server-based artificial neural network model (281), and thus the on-device artificial neural network model (250) can also be configured as a large-scale language model.

[0059] According to one embodiment, the artificial neural network model selection module (230) operates as a single relay module and allows a server-based artificial neural network model (281) or an on-device artificial neural network model (250) to be selected when the health app service provision module (220) generates insight information. The artificial neural network model selection module (230) can include a selection function by defining various message generation functions, thereby allowing one app service to generate insight information of different formats or functions through the artificial neural network model.

[0060] According to one embodiment, an external electronic device (260) can measure various health-related sensor data through sensor(s) (264). The sensor data includes the number of steps, heart rate, body temperature, exercise, and oxygen saturation, and these can be measured through the sensor(s) (264) suitable for the purpose. The external electronic device (260) can continuously obtain the user's biometric information and activity information, and generate various health indicators based on the biometric information. The health indicator is an indicator that can indicate the user's health status, and the sensor data itself can be the health indicator, or it can be obtained by processing the sensor data. The order of the health indicator can be determined based on the number of times the sensor data is processed. For example, if the sensor data itself is the health indicator, the health indicator can be defined as the primary indicator, and if the sensor data is processed once to obtain the health indicator, the health indicator can be defined as the secondary indicator.

[0061] According to one embodiment, the sensor(s) (264) may include a biosensor, an electrode sensor, an inertial sensor, or a position sensor. However, the sensor(s) (264) is not limited to the examples described above.

[0062] In one embodiment, a biosensor is a sensor that detects a user's physiological signals. Typically, an optical sensor can be used that irradiates light onto a living body and receives the absorbed, scattered, and reflected light. At this time, by irradiating light of a designated band onto the living body, various substances such as skin tissue, cells, blood vessels, proteins, deoxyribonucleic acid (DNA), and metabolites can be detected. For example, a biosensor may include a photoplethysmogram (PPG) sensor that irradiates light onto blood vessels in a living body and receives and measures the absorbed, scattered, and reflected light. In addition, a biosensor may receive light for various biomarker information. Sensor data that can be acquired through a biosensor may include heart rate (HR), heart rate variability (HRV), stress, oxygen saturation (SpO2), respiration, and blood pressure. However, the sensor data that can be acquired through a biosensor is not limited to the examples described above. The sensor data acquired through biometric sensors can be used to view long-term trends or to obtain health indicators using values ​​in specific situations. For example, based on the sensor data from biometric sensors, health indicators such as resting heart rate (HR), sleep HR, active HR, sleep HRV, and stress resilience can be obtained. Furthermore, by synthesizing multiple indicators such as heart rate and HRV, secondary and tertiary indicators such as sleep score (an indicator that determines whether the user has had a good night's sleep), energy score (an indicator that determines how active the user can be during the day), and activity score (an indicator that determines whether the user has engaged in appropriate activity) can be obtained.Furthermore, sleep-related indicators, such as oxygen saturation and respiration, can be utilized not only for sleep scores but also for indicators requiring high-precision clinical evaluation, such as sleep apnea. Furthermore, machine learning can be used to define correlations between various sensor data and various phenomena, generating a variety of indicators.

[0063] In one embodiment, although various indicators can be measured using biometric sensors, even for the same indicator (HR, HRV, SpO2, saturation), the values ​​that occur while awake and while asleep may have different meanings. For example, heart rate while awake may be closely related to exercise, while heart rate during sleep may be related to whether or not one ate before bed. Heart rate variability while awake may be related to stress, while heart rate variability during sleep may be related to recovery. Therefore, biometric data during sleep can be utilized in various ways to understand the user. Various health indicators used in relation to sleep may include time of falling asleep, time of waking up, sleep duration, sleep stages (light sleep, deep sleep, REM (rapid eye movement)), sleep cycle (REM sleep repetition), sleep heart rate / heart rate variability / body temperature (skin temperature) / blood pressure, oxygen saturation, sleep efficiency (how much one tosses and turns during sleep), breathing rate, snoring, regularity (whether one falls asleep at a regular time), sleep continuity (whether one wakes up during sleep), and whether one naps during the day. However, health indicators related to sleep are not limited to the examples mentioned above.

[0064] In one embodiment, the electrode sensor can detect electrical signals or measure impedance by making contact with a living body through multiple electrodes. The electrode sensor can detect electrical signals generated in the body. For example, the electrode sensor can be attached to the heart to detect electrocardiogram (ECG) signals, attached to the brain to detect electroencephalography (EEG) signals, attached to a muscle to detect electromyography (EMG) signals, and attached to the eye to detect electrooculography (EOG) signals. Furthermore, the electrode sensor can detect body composition through body impedance analysis (BIA).

[0065] According to one embodiment, a temperature sensor is a sensor that measures the temperature of a living body or a component, and may be a contact or non-contact sensor. The measured temperature value may be stored in memory (262) or transmitted to a processor (261) to be used to estimate body temperature. Constant monitoring of the temperature sensor can detect sudden increases in skin temperature, thereby detecting signs of various diseases, such as colds. Furthermore, detecting the temperature sensor during sleep can be used as an indicator for estimating the menstrual cycle in women.

[0066] According to one embodiment, the inertial sensor is a sensor that detects inertia, and may include, for example, an acceleration sensor and a gyroscope. The inertial sensor may consist of only a three-axis acceleration sensor, or may consist of a three-axis acceleration sensor and a three-axis gyroscope, for a total of six axes. The inertial sensor can sense the motion, gesture, impact, posture, and activity (sedentary, moving, sports) of the device. Using the inertial sensor, the user's activity and exercise can be detected by an external electronic device (260). A representative health indicator that can be obtained through the inertial sensor is the step count. Walking and running can be distinguished through impact amount or step frequency, and a healthy step can be determined based on the step duration. This can be utilized together with a biometric sensor to calculate an activity score. Inertial sensors can automatically detect movements such as walking and running, as well as repetitive movements like ellipticals. Furthermore, repetitive movements like squats can be counted using inertial sensors. Combining inertial and biometric sensors can generate a variety of indicators. For example, resting HR is the heart rate during a state of stillness. Movement is determined by inertial sensors, and heart rate is measured using biometric sensors. The heart rate under these conditions can be stored and then designated as the representative resting HR at a specific time of day or at the lowest value. Furthermore, calorie burning during activity or exercise can be estimated using these combinations, and high HRs can be filtered out based on their association with exercise, as determined by inertial sensors.

[0067] In one embodiment, a position sensor can be used to calculate a user's movement position. During exercise, the position sensor can be used to measure speed while moving. Combined with inertial and biometric sensors, the sensor can determine left-right balance, undulation, ground contact time, and flight time. Furthermore, heart rate information during running can be combined to measure the user's maximum oxygen uptake (VO2 max), a measure of their exercise capacity.

[0068] Figure 3 is a block diagram of a health app service providing module according to one embodiment.

[0069] Referring to FIG. 3, according to one embodiment, the health app service providing module (220) may include an insight display module (310), a health tracking service providing module (320), an insight information generation module (330), and a service data providing module (340). The health tracking service providing module (320) may use sensor data transmitted from an external electronic device (e.g., the external electronic device (260) of FIG. 2) to calculate specific health indicators and provide the corresponding information to a user in various ways. The insight information generation module (330) may generate insight information, and the insight information generation module (330) may also be referred to as an insight information generation framework. The service data providing module (340) may query and provide data defined in an input prompt required for generating insight information. The insight display module (310) may perform an operation to display the generated insight message to the user.

[0070] According to one embodiment, the insight information generation module (330) includes components necessary for generating insight information through a condition-compliant artificial neural network model (e.g., the on-device artificial neural network model (250) or the server-based artificial neural network model (281) of FIG. 2). More specifically, the insight information generation module (330) may include a prompt manager (331), a prompt generation module (332), and a response verification module (333).

[0071] In one embodiment, the prompt manager (331) combines input prompt components according to the structure of a predefined prompt. The input prompt may include common prompts and data prompts. The common prompt may include knowledge information related to a specific health indicator, instructions for message generation regarding the format and length of the message to be generated, and precautions for message generation. The data prompt may be composed of user data and additional information for message generation. More specifically, the data prompt is composed of data parameter names and related actual data, and the data parameters may be sensors of an external electronic device (e.g., sensor(s) (264) of FIG. 2) and health-related elements supported within a health app service. The actual data may be mapped to numeric values, rating information based on specific criteria, and various other information. Each parameter information may be accompanied by words or sentences corresponding to multiple languages ​​for multilingual support, or may be added as supplementary explanatory material. The common prompt may be adjusted in terms of the amount and length of the included content depending on the service and implementation form of the foundation artificial neural network (e.g., a server-based artificial neural network (e.g., a server-based artificial neural network (281) of FIG. 2 or an on-device artificial neural network (250) of FIG. 2), and may be applied in different versions depending on the basic performance of the final service and the artificial neural network model.

[0072] According to one embodiment, the prompt generation module (332) may acquire user health data stored within the electronic device (e.g., memory (240) of FIG. 2) or in an external environment (e.g., an external device that can interact with the electronic device) to map user status data included in the prompt, and may map the acquired health data to a prompt in which no data has yet been entered, thereby integrating knowledge information and detailed instructions for generating insight information and actual user health data to generate a complete input prompt. The prompt generation module (332) may be referred to as a parameter getter.

[0073] According to one embodiment, the response verification module (333) may provide a function to parse insight information and block or filter response content as needed. Furthermore, a verifier based on an artificial neural network model may be included in the response verification module (333). The generated insight information may be checked for abnormalities in content and format through an artificial neural network model, and when an abnormality occurs, new insight information may be regenerated or blocked / filtered through an artificial neural network model (e.g., a server-based artificial neural network (281) of FIG. 2 or an on-device artificial neural network (e.g., an on-device artificial neural network (250) of FIG. 2). A method for verifying insight information is described in detail below with reference to FIG. 5.

[0074] According to one embodiment, the message management server (270) may transmit a unique ID corresponding to a condition for generating insight information to the insight information generation module (330) to provide information so that insight information matching the condition is generated. As will be described in detail below, the electronic device may detect whether the insight information generation condition is satisfied and generate insight information based on a determination that the insight information generation condition is satisfied. At this time, an input prompt may exist for each insight information generation condition, and the message management server (270) may assign a unique ID to the input prompt according to the condition for generating insight information. For example, an input prompt corresponding to a condition for generating the first insight information may be defined as a first input prompt, and in a similar manner, an input prompt corresponding to a condition for generating the n-th (n is a natural number) insight information may be defined as an n-th input prompt.

[0075] According to one embodiment, the message management server (270) can generate information that can update the input prompt and the configuration information of the input prompt embedded in the insight information generation module (330) through the prompt updater (271). For example, the prompt updater (271) can review insight information generated in an electronic device and generate information for removing sensor data or health indicators that are determined to be unhelpful for generating insight information from the input prompt, and transmit the information to the insight information generation module (330). Alternatively, the prompt updater (271) can review insight information generated in an electronic device to detect the cause of hallucination, and generate information for removing the cause, and transmit the information to the insight information generation module (330). The insight information generation module (330) can receive the information and update the prompt. The operation of updating the prompt is not limited to the above example, and the prompt can be updated in various situations. Since the electronic device updates the input prompt via the prompt updater (271) of the message management server (270), message generation-related components can be changed and added without updating or reissuing the health app service itself. By actively and dynamically updating the system's default prompt based on the situation, service type, and user-owned device type using the prompt updater (271), various insight information can be provided to the user, either integrated or individually, depending on the type of service, user preferences, and user-owned device.

[0076] FIG. 4 is a drawing for explaining the structure of an input prompt according to one embodiment.

[0077] Referring to FIG. 4, a common prompt (410) and a data prompt (420) according to one embodiment may be integrated in a structured format to create a single input prompt (400). When an insight information creation condition is satisfied in the insight information creation module (330), a single input prompt (400) is created by the prompt manager (331) and the prompt creation module (332), and the input prompt (400) may be transmitted to the artificial neural network model selection module (230). Depending on the selection of the artificial neural network model selection module (230), the input prompt (400) may be input to an on-device artificial neural network model (250) or a server-based artificial neural network model (281), so that insight information corresponding to the input prompt (400) may be created.

[0078] According to one embodiment, the common prompt (410) may include various types of knowledge information (411) to be included in a service to be provided to the user. The knowledge information (411) is information related to health indicators and may be referred to as knowledge-based information or health knowledge information. For example, the knowledge information (411) may include definitions of health indicators, such as "[Step count] Step count is considered a measure of daily light intensity physical㪋", "[Rest]: Also known as "recovery". Defined as building up energy㪋", "[Stress Level]: Refer to the level of mental or physical㪋". However, the knowledge information (411) is not limited to the above examples and may include various types of information to be included in a service to be provided to the user.

[0079] According to one embodiment, the common prompt (410) may include role information (412) defined about which role an artificial neural network model (e.g., a server-based artificial neural network (281) of FIG. 2 or an on-device artificial neural network (e.g., an on-device artificial neural network (250) of FIG. 2) will use to generate insight information. For example, the role information (412) may include information about a role defined when an artificial neural network model (e.g., a server-based artificial neural network (281) of FIG. 2 or an on-device artificial neural network (e.g., an on-device artificial neural network (250) of FIG. 2) generates insight information, such as "[ROLE DETAILS]:You are a native English speaker, so쪋", [ROLE DETAILS] You are a professional health and wellness coach쪋". However, the role information (412) is not limited to the above example, and may include information about various roles that may be defined when generating insight information. May include.

[0080] According to one embodiment, the common prompt (410) may include message generation method information (413) defined for methods to reduce the accuracy of insight information and the level of hallucination. The message generation method information (413) may be referred to as approach information. For example, the message generation method information (413) may include definitions for message generation methods, such as "[INPUT DESCRIPTION]: The "user status" section offers insights about", [TASK OVERVIEW]: Tou will write a paragraph with a brief health coaching message." However, the message generation method information (413) is not limited to the above examples, and may include information on various methods to reduce the accuracy of insight information and the level of hallucination.

[0081] According to one embodiment, the common prompt (410) may include model output format information (414) defining the form and structure of the insight information. The model output format information (414) may be referred to as an output template or an output example.

[0082] According to one embodiment, the common prompt (410) may include model output rule information (415) defining the generation rules of insight information. The model output rule information (415) may include model output rules (Output rules) that specifically define the format, structure, and number of characters of the output message. For example, the model output rule information (415) may include generation rules of insight information, such as "[Output Rule]: Write a message in English language" and "[Output Rule]: Do not refer to information". However, the model output rule information (415) is not limited to the above examples and may include information on various model output rules.

[0083] According to one embodiment, the data prompt (420) may include user status information (421) defined for the user status. The user status information (421) may include a sentence representing the user status that matches the current insight information generation condition, thereby enabling an artificial neural network model (e.g., the server-based artificial neural network (281) of FIG. 2 or an on-device artificial neural network (e.g., the on-device artificial neural network (250) of FIG. 2) to use it as base information when generating insight information, and may be used as information for suppressing hallucination. For example, the user status information (421) may include text-based information defined for the user status, such as "You're doing a good job maintaining your health on." However, the user status information (421) is not limited to the above example, and may include definitions for various user statuses.

[0084] According to one embodiment, the data prompt (420) may include user data information (422) composed of user status data including sensor data about the user. The user data information (422) may be utilized as specific data that can support the user status information (421) when a message is generated by including status data most closely related to the user status information (421). For example, the user data information (422) may include the user's biometric data acquired through sensor data, such as "Sleep time: 9 hours 30 minutes." However, the user data information (422) is not limited to the above example and may include various user status data.

[0085] According to one embodiment, the data prompt (420) may include user additional data information (423), which is not included in the user data information (422), but is comprised of data about the user that can be considered in the current state. The user additional data information (423) may provide additional user health information to the user along with the current user state information (421) and the user data information (422), and may be utilized as data for generating personalized messages. For example, the user additional data information (423) may include additional user status data that can be considered in the current state, such as "[Additional data]: Average step count: 5500 steps." However, the user additional data information (423) is not limited to the above example, and may include various additional user status data that can be considered in the current state.

[0086] FIG. 5 is a diagram illustrating a method for verifying insight information according to one embodiment.

[0087] Referring to FIG. 5, a response verification module (333) according to one embodiment can verify content included in generated insight information through methods such as parsing, and perform blocking and filtering on problematic response content. The response verification module (333) may include a verification prompt generation module (510).

[0088] According to one embodiment, the prompt generation module (332) can generate an input prompt (501) (e.g., the input prompt (400) of FIG. 4). The input prompt (501) can be input into an artificial neural network model (520) (e.g., the server-based artificial neural network (281) of FIG. 2 or an on-device artificial neural network (e.g., the on-device artificial neural network (250) of FIG. 2), and the artificial neural network model (520) can generate insight information (502) corresponding to the input prompt (501). The insight information (502) is insight information that has not yet been verified and may be referred to as primary insight information to distinguish it from final insight information (506).

[0089] According to one embodiment, insight information (502) may be input into a verification prompt generation module (510) of a response verification module (333). The verification prompt generation module (510) may be a module that generates a verification input prompt (503) for verification together with the insight information (502). The verification prompt generation module (510) may be an artificial neural network model trained based on machine learning. The verification input prompt (503) is a prompt for instructing the verification artificial neural network model (530) on whether to verify the input prompt (501), and may include insight information (502), which is a verification target, and a verification instruction generated by the verification prompt generation module (510). For example, the verification input prompt (503) may be composed of a verification target and a verification instruction, such as "Verify the input prompt (501) using method A." The verification input prompt (503) is not limited to the above example and may include various verification instructions.

[0090] According to one embodiment, the input prompt (503) for verification can be input to a verification artificial neural network model (530). The verification artificial neural network model (530) can be a large-scale language model. The verification artificial neural network model (530) can be the same artificial neural network model as the artificial neural network model (520) used to generate the insight information (502), or can be a different artificial neural network model from the artificial neural network model (520). For example, the verification artificial neural network model (530) can have a larger parameter scale than the used artificial neural network model (520) or can be a heterogeneous large-scale language model. The verification artificial neural network model (530) can check the content included in the insight information (502) and perform verification follow-up tasks such as blocking and filtering if necessary, and output verification result information (504) including the content. For example, the verification artificial neural network model (530) can verify the insight information (502) for a plurality of items, and a verification score can be output for each item. Whether or not each item passes the verification can be determined based on the verification score for each item as a threshold. The verification artificial neural network model (530) can additionally generate analysis information to solve the problem for items that fail the verification. For example, if the insight information (502) does not pass the hallucination verification item, the verification artificial neural network model (530) can generate hallucination analysis information. The verification result information (504) can include one or more of the verification score, whether or not the verification passed, and the analysis information. However, the verification result information (504) is not limited to the above-described examples, and can include various configurations.

[0091] According to one embodiment, a final input prompt (505) may be generated based on the verification result information (504) and the input prompt (501). The final input prompt (505) may be input into an artificial neural network model (520), and the artificial neural network model (520) may generate final insight information (506) corresponding to the final input prompt (505). Since the final input prompt (505) includes the verification result information (504), the artificial neural network model (520) may generate final insight information (506) in which problematic response contents in the insight information (502) are blocked and filtered. The final insight information (506) may be referred to as secondary insight information to distinguish it from the insight information (502).

[0092] Figure 6 is a flowchart illustrating a method for generating insight information according to one embodiment.

[0093] Referring to FIG. 6, according to one embodiment, operations 610 and 660 may be performed sequentially, but are not necessarily performed sequentially. For example, the order of each operation (610 and 660) may be changed, and at least two operations may be performed in parallel. For convenience of explanation, operations (610 to 660) are described as being performed using the electronic device (200) illustrated in FIG. 2. However, these operations (610 to 660) may be used through any other suitable electronic device and within any suitable system. According to one embodiment, operations (610 to 660) may be understood as being performed in a processor (e.g., processor 210 of FIG. 2) of an electronic device (e.g., electronic device (200) of FIG. 2).

[0094] In operation 610, according to one embodiment, the electronic device (200) may receive sensor data about a user of the external electronic device from an external electronic device (e.g., the external electronic device (260) of FIG. 2).

[0095] In operation 620, according to one embodiment, the electronic device (200) may determine whether a condition for generating insight information is satisfied. The electronic device (200) may continuously perform a process for checking a specific insight information generation condition.

[0096] In operation 630, according to one embodiment, the electronic device (200) may obtain a unique ID corresponding to the insight information generation condition based on a determination that the insight information generation condition has been satisfied. The prompt manager (e.g., the prompt manager (331) of FIG. 3) may obtain a unique ID corresponding to the corresponding generation condition when a specific insight information generation condition is triggered.

[0097] In operation 640, according to one embodiment, the electronic device (200) may obtain a common prompt (e.g., common prompt (410) of FIG. 4) corresponding to the unique ID. The prompt manager may generate the common prompt, which is a structured prompt for generating appropriate insight information.

[0098] In operation 650, according to one embodiment, the electronic device (200) may obtain a data prompt (e.g., data prompt (420) of FIG. 4) corresponding to a unique ID based on sensor data to generate an input prompt (e.g., input prompt (400) of FIG. 4). A prompt generation module (e.g., prompt generation module (332) of FIG. 3) may obtain data necessary for configuring the input prompt and generate the input prompt.

[0099] According to one embodiment, the input prompt (400) composed of a common prompt (410) and a data prompt (420) may have different contents and configurations depending on which artificial neural network model (e.g., the on-device artificial neural network model (250) of FIG. 2 or the server-based artificial neural network model (281)) it is transmitted to. For example, the contents and configurations of the input prompt may be applied and differentiated depending on the input token availability range and foundation model performance of the artificial neural network model (e.g., the on-device artificial neural network model (250) of FIG. 2 or the server-based artificial neural network model (281)). For example, the input prompt (400) for the on-device artificial neural network model (250) may be composed of only very simple output message requirements and a data prompt.

[0100] According to one embodiment, the artificial neural network model selection module (230) can select one of the on-device artificial neural network model (250) and the server-based artificial neural network model (281). The electronic device (200) can process the input prompt (400) based on the selected artificial neural network model. If the artificial neural network model selection module (230) selects the on-device artificial neural network model (250), the electronic device (200) can reduce the number of parameters configuring the input prompt (400) by considering the input token availability range of the on-device artificial neural network model (250). For example, the knowledge information (411) may occupy the largest proportion in the input prompt (400). In this case, the electronic device (200) can reduce the size of the input prompt (400) by deleting the knowledge information (411) from the input prompt (400). When the artificial neural network model selection module (230) selects a server-based artificial neural network model (281), the electronic device (200) can input the input prompt (400) as is into the server-based artificial neural network model (281). However, the method of processing the input prompt (400) is not limited to the above example.

[0101] According to one embodiment, the artificial neural network model selection module (230) may select one of the on-device artificial neural network model (250) and the server-based artificial neural network model (281) as input to the input prompt (400) based on the user's selection. For example, the electronic device (200) may receive a user's selection of which model to use between the on-device artificial neural network model (250) and the server-based artificial neural network model (281) at the initial setup stage of the health app. The electronic device (200) may provide the user with a guidance text informing him / her that the on-device artificial neural network model (250) may have lower accuracy but does not require sending personal information to the server, and the server-based artificial neural network model (281) may provide him / her with a guidance text informing him / her that personal information must be sent to the server but more accurate insight information can be received.

[0102] According to another embodiment, the artificial neural network model selection module (230) may select one of the on-device artificial neural network model (250) and the server-based artificial neural network model (281) as an input of the input prompt (400) depending on the importance of the insight information generation condition. For example, if the importance of the insight information generation condition is greater than a threshold value (e.g., the heart rate during sleep falls below the threshold value), the artificial neural network model selection module (230) may select the server-based artificial neural network model (281) as an input of the input prompt (400). Alternatively, if the importance of the insight information generation condition is below a threshold value (e.g., the time for providing regular health indicators has arrived), the artificial neural network model selection module (230) may select the on-device artificial neural network model (250) as an input of the input prompt (400).

[0103] In operation 660, according to one embodiment, the electronic device (200) may input an input prompt to an artificial neural network model to generate insight information. The insight display module (310) may perform an operation to display the insight information to the user.

[0104] FIG. 7 is a flowchart illustrating an example of an operation for generating an insight message according to one embodiment.

[0105] Referring to FIG. 7, operations 710 and 790 may be performed sequentially, but are not necessarily performed sequentially. For example, the order of each operation (710 and 790) may be changed, and at least two operations may be performed in parallel. For convenience of explanation, operations (710 to 790) are described as being performed using the electronic device (200) illustrated in FIG. 2. However, these operations (710 to 790) may be used through any other suitable electronic device and within any suitable system. For example, these operations (710 to 790) may also be performed by the external electronic device (260) illustrated in FIG. 2. According to one embodiment, operations (710 to 790) may be understood as being performed by a processor (e.g., the processor (210) illustrated in FIG. 2) of an electronic device (e.g., the electronic device (200) illustrated in FIG. 2).

[0106] In operation 710, according to one embodiment, the electronic device (200) checks for a prompt update, in operation 720, the electronic device (200) updates the input prompt (400) in the insight information generation module (330), and in operation 730, the electronic device can initialize the insight information generation module (330). When the health app service starts, the electronic device (200) can check whether a new prompt component is ready through the prompt updater (271) of the message management server (270). If there is an updated prompt component, the electronic device (200) can update existing prompt components in the insight information generation module (330) and then initialize the insight information generation module (330). If there is no updated component, the electronic device (200) can initialize the insight information generation module (330) immediately. The electronic device (200) can complete the configuration of a predefined input prompt (400) by initializing the prompt for each insight information generation unique ID and the data items required for each individual prompt during the initialization process.

[0107] In operation 740, according to one embodiment, the electronic device (200) continuously performs a process of checking a condition for occurrence of specific insight information, and in operation 750, the electronic device (200) secures data necessary for configuring an input prompt (400) corresponding to a corresponding unique ID when a condition for occurrence of specific insight information is triggered, and completes generation of the input prompt (400).

[0108] In operation 760, according to one embodiment, a target artificial neural network model (e.g., the on-device artificial neural network model (250) or the server-based artificial neural network model (281) of FIG. 2) is called through the artificial neural network model selection module (230), and an input prompt (400) may be transmitted to the selected artificial neural network model. In operation 770, the electronic device (200) may generate insight information using the transmitted input prompt, and in operation 780, the generated insight information may be transmitted to the insight information generation module (330).

[0109] According to one embodiment, the insight information transmitted to the insight information generation module (330) can be inspected for content included in the insight information through the response verification module (333) and, if necessary, perform follow-up verification tasks such as blocking and filtering. In operation 790, the insight information that passes the evaluation criteria of the response verification module (333) is finally displayed in the insight information display window of the health app and delivered to the user, and the insight information generation service can be terminated.

[0110] FIG. 8 is a flowchart illustrating an example of an operation for generating an input prompt according to one embodiment.

[0111] Referring to FIG. 8, operations 810 and 880 may be performed sequentially, but are not necessarily performed sequentially. For example, the order of each operation (810 and 880) may be changed, and at least two operations may be performed in parallel. For convenience of explanation, operations (810 to 880) are described as being performed using the electronic device (200) illustrated in FIG. 2. However, these operations (810 to 880) may be used through any other suitable electronic device and within any suitable system. For example, these operations (810 to 880) may also be performed by the external electronic device (260) illustrated in FIG. 2. According to one embodiment, operations (8120 to 880) may be understood as being performed by a processor (e.g., the processor (210) illustrated in FIG. 2) of an electronic device (e.g., the electronic device (200) illustrated in FIG. 2).

[0112] In operation 810, according to one embodiment, the electronic device (200) may initialize the insight information generation module (330). In operation 820, the electronic device (200) may configure a structure of an input prompt (400) for each unique ID. The input prompt (400) for each unique ID is composed of a common prompt (410) and a data prompt (420). In operation 830, the electronic device (200) checks an insight information generation condition, and in operation 840, if the insight information generation condition is met, the electronic device (200) may acquire and configure the status data required to configure the input prompt (400) at that point in time. However, there may be cases where the data is not available due to various reasons such as the user's selection or lifestyle pattern among the status data. At this time, in operation 850, the electronic device (200) checks data availability, and if data is not available, in operation 860, the electronic device (200) can update the configuration of the data prompt (420) by removing the user data information (422) items in the data prompt (420) by unique ID. Alternatively, in some cases, the electronic device (200) can update the data prompt by including the result display regarding data availability while maintaining the configuration of the data prompt (420) items. If data is available, in operation 870, the electronic device (200) can confirm the data prompt (420) related to the unique ID, and in operation 880, the electronic device (200) can generate the input prompt (400) by unique ID.

[0113] FIG. 9 is a diagram illustrating an example of a screen on which insight information is displayed on an electronic device (200) according to one embodiment.

[0114] Referring to FIG. 9, according to one embodiment, when an insight information generation condition is met, insight information (first insight information (911), second insight information (912), and third insight information (913)) may be displayed on a screen 910 of the electronic device (200). On the screen 910, the first insight information (911) provides summary information on energy points, the second insight information (912) provides summary information on today's calorie consumption, and the third insight information (913) provides summary information on today's step count.

[0115] According to one embodiment, when receiving an input from a user to select first insight information (911), a screen 920 may be displayed on the electronic device (200). On the screen 920, the fourth insight information (921) may provide a graph regarding daily energy score trends, the fifth insight information (922) may provide a graph regarding today's energy score, and the sixth insight information (923) may provide an insight message related to the energy score.

[0116] According to one embodiment, the energy score is a health indicator that indicates the physical and mental energy status of a user, indicating the extent to which the user is prepared to live a day, and can be calculated by utilizing heart rate information, heart rate variability information, exercise and activity information, and sleep information measured by an external electronic device (e.g., the external electronic device (260) of FIG. 2). The electronic device (200) can determine the energy score by scoring a composite user health status by utilizing various user status data (e.g., health data and sensor data) to measure the user's physical and mental energy status, rather than a score for a specific indicator data or sensor data. Therefore, even if the score is the same, the meaning and status may differ for each user, and thus, based on the score that is updated each time and the data that constitutes the score, the energy score value, score change, score trend, and the value itself, change, and trend of specific data that constitutes or influences the score, personalized insight information (e.g., insight messages) that can help the user's health and lifestyle can be provided at the same time. Furthermore, the electronic device (200) can transmit the current user's status data to an artificial neural network model (e.g., the on-device artificial neural network model (250) or the server-based artificial neural network model (281) of FIG. 2) so that new insight information can be automatically generated and provided to the user each time by actively utilizing individual conditions and data, rather than a uniform and fragmentary message based on rules.

[0117] FIG. 10 is a diagram illustrating an example of a screen provided to obtain information of interest to a user according to one embodiment.

[0118] Referring to FIG. 10, according to one embodiment, the electronic device (200) can obtain health interest information (first interest information (1011), second interest information (1012), third interest information (1013), and fourth interest information (1014)) from the user. On screen 1010, the user can select health interest information (first interest information (1011) related to overall health, second interest information (1012) related to sleep, third interest information (1013) related to exercise, and fourth interest information (1014) related to healthy weight). However, the health interest information is not limited to the above example.

[0119] According to one embodiment, the electronic device (200) can generate insight information generation conditions based on at least one of health interest information and health indicators. Insight information generation can be triggered by various events and provided to the user. The events can be provided based on the user's health interest information, and based on such interest information, status data of an external electronic device (260) or a health indicator algorithm embedded within the electronic device (200) can be modified or personalized to apply various criteria for generating insight information events.

[0120] According to one embodiment, an event may include a case where a user's sensor data measured by an external electronic device (260) and a health indicator created using the sensor data reach, exceed, or fall below a specific threshold. For example, a message may be generated when the user's sleep time is shorter than average or when the user's exercise amount is higher than average. Additionally, an event may also be generated when a function implemented with a specific algorithm within the electronic device (200) continuously checks the user's status data and detects a point in time when a specific condition defined in the algorithm is satisfied. Furthermore, the time at which insight information is provided may be designated depending on the nature of the health indicator. In a situation where it is necessary to provide the user with health-related lifestyle guidance today using data from yesterday or the past several days, a specific time, such as between 06:00 and 07:00, in the morning or within 30 minutes after waking up, may be defined as a message generation event.

[0121] FIG. 11 is a diagram illustrating a method for generating insight information based on a multi-modal artificial neural network according to one embodiment.

[0122] Referring to FIG. 11, according to one embodiment, an artificial neural network model (e.g., an on-device artificial neural network model (250) or a server-based artificial neural network model (281) of FIG. 2) may be a multi-modal artificial neural network model (1140). More specifically, the electronic device (200) may be a multi-modal artificial neural network model (1140) capable of providing various types of content, and may receive state data (1110), an input prompt (1120), and external data (1130), and may generate and provide to a user, in addition to text-based insight information (1180), image-based insight information (1150), audio-based insight information (1160), video-based insight information (1170), and text-based insight information (1180).

[0123] According to one embodiment, a multi-modal artificial neural network model (1140) can receive various inputs as status data. The multi-modal artificial neural network model (1140) can receive not only the user's health-related status data, but also the user's current ambient temperature, user environment parameters through recording and recognition of surrounding environment images / videos, and various recognition parameters through surrounding environment sounds as input, and can utilize these to provide the most appropriate insight information for the user's current time in various formats.

[0124] FIG. 12 is an example of a screen displaying insight information generated using a multi-modal artificial neural network according to one embodiment.

[0125] Referring to FIG. 12, according to one embodiment, the electronic device (200) can generate energy scores, which are insight information, by using a multi-modal artificial neural network (e.g., the multi-modal artificial neural network (1140) of FIG. 11). When using insight information generated based on the multi-modal artificial neural network, the level of user experience can be improved by adding diverse insight information (e.g., first insight information (1210), second insight information (1220), third insight information (1230), and fourth insight information (1240)) to the 1200 screen provided and explained to the user. The 1200 screen can provide the user with first insight information (1210), which is an icon that visualizes the factor that most influenced the current score among the health data items that constitute the energy score, second insight information (1220) that animates the icon, third insight information (1230), which is information about the health indicator that most influenced the energy score, and fourth insight information (1240), which is an insight message about the energy score.

[0126] According to one embodiment, the electronic device (200) may provide additional animations and icons in addition to the insight message to further enhance the user's understanding, thereby increasing the user's willingness to improve the factor or reinforcing the feeling of positive user encouragement.

[0127] In one embodiment, the form of the generated insight information may vary depending on the type of device on which the insight information is displayed. For example, when generating insight information on a tablet PC with a larger screen than a smartphone, an insight message containing a greater amount of information may be generated. Alternatively, when generating insight messages on a device with a smaller display, such as a wearable device, an insight image may be generated. To achieve this, the form and length of sentences can be adjusted by configuring the input prompt. For example, a wearable device may configure the input prompt to generate short, core phrases, while a large-screen device, such as a tablet PC, may configure the input prompt to provide a more in-depth insight message that includes various health data-based insight information and recommendations related to exercise and lifestyle.

[0128] In one embodiment, the insight information generation method described herein can be utilized even in a screenless voice AI service. Using user status data and input prompts, insight messages can be generated for the user, and provided to the voice AI service, enabling natural speech to deliver the message to the user through voice.

[0129] According to one embodiment, the method for generating insight information according to one embodiment can provide insight information generated through an artificial neural network model to a user terminal by bridging the information even on a TV or set-top box with built-in smart functions, and a home appliance with built-in screen and HUB functions.

[0130] In one embodiment, the same insight information generation pipeline can be used to generate and provide insight information for various health indicators offered by a health app service. By establishing knowledge-based content describing each health indicator and standards for the content and format required to provide insight information for each individual health indicator, services can be made possible using the same approach.

[0131] According to one embodiment, for health insight messages already provided for existing health indicators, insight information can be newly generated and provided to users using existing rule-based insight messages and related input prompts so that a wider variety of phrases or health recommendation items can be generated while maintaining the insight content and context.

[0132] According to one embodiment, when a health-related wearable device or external device is connected to a health app, shares actual data, or starts / stops calculating an indicator by an algorithm, text or related health information / knowledge that can provide relevant guidance to the user can be automatically provided using user data, device connection information, related message generation prompts, and an artificial neural network model.

[0133] In one embodiment, if essential configuration data required for message generation is missing, a measurement guide or explanation of the need for that data can be configured to be automatically provided at the necessary time by utilizing user data, device information, relevant message generation prompts, and artificial neural network models.

[0134] In one embodiment, a service can be provided by transmitting a base message generated using the same input prompt and an artificial neural network model to devices connected to the main terminal in various modified forms through the artificial neural network model. For example, a summary of the base message can be generated and provided to a wearable device, and if the main terminal has limitations in the display area of ​​the insight message, the sentence structure or length can be modified to accommodate this limitation.

[0135] Fig. 13 illustrates an example of an electronic device according to one embodiment.

[0136] The electronic device (1300) may include a memory (1310) and a processor (1330). The electronic device (1300) may include the electronic device (101) of FIG. 1 and / or the electronic device (200) of FIG. 2. The memory (1310) may include the memory (240) of FIG. 2. The processor (1330) may include the processor (210) of FIG. 2.

[0137] The memory (1310) may store instructions (e.g., programs) executable by the processor (1330). For example, the instructions may include instructions for executing operations of the processor (1330) and / or operations of each component of the processor (1330).

[0138] The memory (1310) may be implemented as a volatile memory device or a non-volatile memory device.

[0139] Volatile memory devices can be implemented as dynamic random access memory (DRAM), static random access memory (SRAM), thyristor RAM (T-RAM), zero capacitor RAM (Z-RAM), or twin transistor RAM (TTRAM).

[0140] The nonvolatile memory device may be implemented as an Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory, Magnetic RAM (MRAM), Spin-Transfer Torque (STT)-MRAM, Conductive Bridging RAM (CBRAM), Ferroelectric RAM (FeRAM), Phase change RAM (PRAM), Resistive RAM (RRAM), Nanotube RRAM, Polymer RAM (PoRAM), Nano Floating Gate Memory (NFGM), holographic memory, Molecular Electronic Memory Device, or Insulator Resistance Change Memory.

[0141] The processor (1330) can process data stored in the memory (1310). The processor (1330) can execute computer-readable code (e.g., software) stored in the memory (1310) and instructions generated by the processor (1330).

[0142] The processor (1330) may be a data processing device implemented as hardware having a circuit with a physical structure for executing desired operations. For example, the desired operations may include code or instructions included in a program.

[0143] For example, a data processing device implemented in hardware may include a microprocessor, a central processing unit, a processor core, a multi-core processor, a multiprocessor, an application-specific integrated circuit (ASIC), or a field programmable gate array (FPGA).

[0144] The electronic device (101) of FIG. 1 and / or the electronic device (200) of FIG. 2 may be stored in a memory (1310) and executed by a processor (1330) or embedded in the processor (1330). The processor (1330) may perform substantially the same operations as the electronic device (101) and / or the electronic device (200) with reference to FIGS. 1 to 12. Therefore, a detailed description thereof will be omitted.

[0145] The embodiments described through FIGS. 1 to 12 can be applied to the embodiment of FIG. 13.

[0146] According to one embodiment, an electronic device (e.g., electronic device (101) of FIG. 1, electronic device (200) of FIG. 2, and electronic device (1300) of FIG. 13) may include at least one processor (e.g., processor (201) of FIG. 2 and processor (1330) of FIG. 13) including processing circuitry, and a memory (e.g., memory (203) of FIG. 2 and memory (1310) of FIG. 13) that stores instructions. When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, and 1300) may: receive sensor data about a user of the external electronic device (102, 260) from the external electronic device (102, 260), determine whether an insight information generation condition is satisfied, and, based on a determination that the insight information generation condition is satisfied, obtain a unique ID corresponding to the insight information generation condition, obtain a common prompt (410) corresponding to the unique ID, obtain a data prompt (420) corresponding to the unique ID based on the sensor data to generate an input prompt (400), and input the input prompt (400) into an artificial neural network model to generate insight information.

[0147] The common prompt (410) may include one or more of knowledge information (411) related to health indicators, role information (412) defined as to which role the artificial neural network model will use to generate the insight information, message generation method information (413) defined as to how to reduce the accuracy and hallucination level of the insight information, model output format information (414) defined as to the form and composition of the insight information, and model output rule information (415) defined as to the generation rule of the insight information.

[0148] The above data prompt (420) may include one or more of user state information (421) defined for the state of the user, user data information (422) consisting of sensor data for the user, and user additional data information (423) consisting of sensor data for the user that is not included in the user data information (422) but can be considered in the current state.

[0149] The electronic devices (101, 200, and 1300) according to one embodiment may further include a display that outputs the insight information.

[0150] The artificial neural network model may include at least one of an on-device artificial neural network model (250) and a server-based artificial neural network model (281), and when the instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) may: select at least one of the on-device artificial neural network model (250) and the server-based artificial neural network model (281), process the input prompt (400) based on the selected artificial neural network model, and input the processed input prompt (400) to the selected artificial neural network model to generate insight information.

[0151] When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) may: obtain health interest information from a user, generate a health indicator based on the sensor data, and generate an insight information generation condition based on at least one of the health interest information and the health indicator.

[0152] When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) may be caused to: obtain primary insight information (502), generate a verification input prompt (503) based on the primary insight information (502), input the verification input prompt (503) into a verification artificial neural network model (530), generate verification result information (504), and input the verification result information (504) and the input prompt (501) used when obtaining the primary insight information (502) into the artificial neural network model, thereby obtaining final insight information (506).

[0153] When the above commands are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) may: receive update information capable of updating the input prompt (400) and the configuration information of the input prompt (400) from the message management server (270), and update the input prompt (400) and the configuration information of the input prompt (400) based on the update information.

[0154] The above insight information may have at least one form of text, image, voice, and video.

[0155] When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) may be caused to: process the insight information based on the external electronic device (102, 260), and transmit the processed insight information to the external electronic device (102, 260).

[0156] A method performed by an electronic device (101, 200, and 1300) according to one embodiment may include: receiving sensor data about a user of an external electronic device (102, 260) from an external electronic device (102, 260); determining whether an insight information generation condition is satisfied; obtaining a unique ID corresponding to the insight information generation condition based on a determination that the insight information generation condition is satisfied; obtaining a common prompt (410) corresponding to the unique ID; obtaining a data prompt (420) corresponding to the unique ID based on the sensor data to generate an input prompt (400); and inputting the input prompt (400) into an artificial neural network model to generate insight information.

[0157] The common prompt (410) may include one or more of knowledge information (411) related to health indicators, role information (412) defined as to which role the artificial neural network model will use to generate the insight information, message generation method information (413) defined as to how to reduce the accuracy and hallucination level of the insight information, and model output rule information (415) defined as to the form and composition of the insight information.

[0158] The above data prompt (420) may include one or more of user state information (421) defined for the state of the user, user data information (422) consisting of sensor data for the user, and user additional data information (423) consisting of sensor data for the user that is not included in the user data information (422) but can be considered in the current state.

[0159] The method performed by the electronic device (101, 200, and 1300) according to one embodiment may further include an operation of outputting the insight information.

[0160] The artificial neural network model may include at least one of an on-device artificial neural network model (250) and a server-based artificial neural network model (281), and the operation of generating the insight information may further include an operation of selecting at least one of the on-device artificial neural network model (250) and the server-based artificial neural network model (281); an operation of processing the input prompt (400) based on the selected artificial neural network model; and an operation of inputting the processed input prompt (400) into the selected artificial neural network model to generate the insight information.

[0161] A method performed by an electronic device (101, 200, and 1300) according to one embodiment may further include: obtaining health interest information from a user; generating a health indicator based on the sensor data; and generating an insight information generation condition based on at least one of the health interest information and the health indicator.

[0162] A method performed by an electronic device (101, 200, and 1300) according to one embodiment may further include: obtaining primary insight information (502); generating a verification input prompt (503) based on the primary insight information (502); inputting the verification input prompt (503) into a verification artificial neural network model (530) to generate verification result information (504); and inputting the verification result information (504) and the input prompts (400, 501) used when obtaining the primary insight information (502) into the artificial neural network model to obtain final insight information (506).

[0163] A method performed by an electronic device (101, 200, and 1300) according to one embodiment may further include: receiving update information capable of updating the input prompt (400) and configuration information of the input prompt (400) from a message management server (270); and updating the input prompt (400) and configuration information of the input prompt (400) based on the update information.

[0164] A method performed by an electronic device (101, 200, and 1300) according to one embodiment may further include an operation of processing the insight information based on the external electronic device (102, 260); and an operation of transmitting the processed insight information to the external electronic device (102, 260).

[0165] The effects that can be obtained from the disclosure are not limited to the effects mentioned above, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the art to which the disclosure pertains from the description of the present disclosure.

[0166] Electronic devices according to the various embodiments disclosed in this document may take various forms. Electronic devices may include, for example, portable communication devices (e.g., smartphones), computer devices, portable multimedia devices, portable medical devices, cameras, wearable devices, or home appliances. Electronic devices according to the embodiments of this document are not limited to the aforementioned devices.

[0167] The various embodiments of this document and the terminology used therein are not intended to limit the technical features described in this document to specific embodiments, but should be understood to include various modifications, equivalents, or substitutes of the embodiments. In connection with the description of the drawings, similar reference numerals may be used for similar or related components. The singular form of a noun corresponding to an item may include one or more of the items, unless the context clearly indicates otherwise. In this document, each of the phrases "A or B", "at least one of A and B", "at least one of A or B", "A, B, or C", "at least one of A, B, and C", and "at least one of A, B, or C" can include any one of the items listed together in the corresponding phrase among those phrases, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used merely to distinguish one component from another, and do not limit the components in any other respect (e.g., importance or order). When a component (e.g., a first component) is referred to as "coupled" or "connected" to another (e.g., a second component), with or without the terms "functionally" or "communicatively," it means that the component can be connected to the other component directly (e.g., wired), wirelessly, or through a third component.

[0168] The term "module" used in various embodiments of this document may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be an integral component, or a minimum unit or part of such a component that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0169] Various embodiments of the present document may be implemented as software (e.g., a program (1740)) including one or more instructions stored in a storage medium (e.g., an internal memory (1736) or an external memory (1738)) readable by a machine (e.g., an electronic device (1701)). For example, a processor (e.g., a processor (1720)) of the machine (e.g., an electronic device (1701)) may call at least one instruction among the one or more instructions stored from the storage medium and execute it. This enables the machine to operate to perform at least one function according to the at least one called instruction. The one or more instructions may include code generated by a compiler or code executable by an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-transitory' simply means that the storage medium is a tangible device and does not contain signals (e.g., electromagnetic waves), and the term does not distinguish between cases where data is stored semi-permanently or temporarily on the storage medium.

[0170] According to one embodiment, the method according to various embodiments disclosed in this 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 (e.g., Play Store™) or directly between two user devices (e.g., smart phones). In the case of online distribution, at least a portion of the computer program product 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.

[0171] According to various embodiments, each component (e.g., a module or a program) of the above-described components may include one or more entities, and some of the entities may be separated and arranged in other components. According to various embodiments, one or more components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In such a case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to the integration. According to various embodiments, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

Claims

1. In electronic devices (101, 200, and 1300), At least one processor (210, 1330) comprising processing circuitry; and Includes a memory (240, 1310) for storing instructions, When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, and 1300) causes: Receive sensor data about a user of the external electronic device (102, 260) from the external electronic device (102, 260), Explore whether the conditions for generating insight information are met, Based on the judgment that the above insight information generation condition has been met, a unique ID corresponding to the above insight information generation condition is obtained, Obtain a common prompt (410) corresponding to the above unique ID, Based on the above sensor data, a data prompt (420) corresponding to the unique ID is obtained to generate an input prompt (400), Generate insight information by inputting the above input prompt (400) into the artificial neural network model. To do, Electronic devices (101, 200, and 1300).

2. In paragraph 1, The above common prompt (410) is Including at least one of knowledge information (411) related to health indicators, role information (412) defined on the basis of which role the artificial neural network model will generate the insight information, message generation method information (413) defined on a method for reducing the accuracy and hallucination level of the insight information, model output format information (414) defined on the form and composition of the insight information, and model output rule information (415) defined on the generation rule of the insight information. Electronic devices (101, 200, and 1300).

3. In either of paragraphs 1 and 2, The above data prompt (420) Including at least one of user status information (421) defined for the status of the user, user data information (422) consisting of sensor data for the user, and user additional data information (423) consisting of sensor data for the user that is not included in the user data information (422) but can be considered in the current status. Electronic devices (101, 200, and 1300).

4. In any one of paragraphs 1 to 3, Outputting the above insight information display; including more, Electronic devices (101, 200, and 1300).

5. In any one of paragraphs 1 to 4, The above artificial neural network model Contains at least one of an on-device artificial neural network model (250) and a server-based artificial neural network model (281), When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) causes: Select at least one of the above on-device artificial neural network model (250) and server-based artificial neural network model (281), Based on the above-mentioned selected artificial neural network model, the input prompt (400) is processed, Generate insight information by inputting the processed input prompt (400) into the selected artificial neural network model. To do, Electronic devices (101, 200, and 1300).

6. In any one of paragraphs 1 to 5, When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) causes: Obtain health interest information from users, Based on the above sensor data, health indicators are generated, Generating the insight information generation condition based on at least one of the above health interest information and the above health indicator. To do, Electronic devices (101, 200, and 1300).

7. In any one of paragraphs 1 to 6, When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) causes: Obtain primary insight information (502), Generate an input prompt (503) for verification based on the above primary insight information (502), By inputting the above verification input prompt (503) into the verification artificial neural network model (530), verification result information (504) is generated, The verification result information (504) and the input prompt (501) used to obtain the first insight information (502) are input into the artificial neural network model to obtain the final insight information (506). To do, Electronic devices (101, 200, and 1300).

8. In any one of paragraphs 1 to 7, When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) causes: Receive update information that can update the input prompt (400) and configuration information of the input prompt (400) from the message management server (270), Based on the above update information, the input prompt (400) and the configuration information of the input prompt (400) are updated. To do, Electronic devices (101, 200, and 1300).

9. In any one of paragraphs 1 to 8, The above insight information is Having at least one form of text, image, voice and video, Electronic devices (101, 200, and 1300).

10. In any one of paragraphs 1 to 9, When the above instructions are individually or collectively executed by the at least one processor (210, 1330), the electronic device (101, 200, 1300) causes: Based on the external electronic device (102, 260), the insight information is processed, Transmitting the processed insight information to the external electronic device (102, 260) To do, Electronic devices (101, 200, and 1300).

11. In a method performed by an electronic device (101, 200, and 1300), An operation of receiving sensor data about a user of an external electronic device (102, 260) from an external electronic device (102, 260); An action to explore whether the conditions for generating insight information are met; An operation of obtaining a unique ID corresponding to the insight information generation condition based on a judgment that the above insight information generation condition has been satisfied; An action to obtain a common prompt (410) corresponding to the above unique ID; An operation of generating an input prompt (400) by obtaining a data prompt (420) corresponding to the unique ID based on the sensor data; and An action to generate insight information by inputting the above input prompt (400) into an artificial neural network model. including, method.

12. In paragraph 11, The above common prompt (410) is Including at least one of knowledge information (411) related to health indicators, role information (412) defined as to which role the artificial neural network model will use to generate the insight information, message generation method information (413) defined as to how to reduce the accuracy and hallucination level of the insight information, and model output rule information (415) defined as to the form and composition of the insight information. method.

13. In any one of paragraphs 11 and 12, The above data prompt (420) Including at least one of user status information (421) defined for the status of the user, user data information (422) consisting of sensor data for the user, and user additional data information (423) consisting of sensor data for the user that is not included in the user data information (422) but can be considered in the current status. method.

14. In any one of paragraphs 11 to 13, Action to output the above insight information including more, method.

15. In any one of paragraphs 11 to 14, The above artificial neural network model Contains at least one of an on-device artificial neural network model (250) and a server-based artificial neural network model (281), The action of generating the above insight information is An operation of selecting at least one of the on-device artificial neural network model (250) and the server-based artificial neural network model (281); An operation of processing the input prompt (400) based on the selected artificial neural network model; and An operation of generating insight information by inputting the processed input prompt (400) into the selected artificial neural network model. including more, method.

Images