Method for providing recommendation message related to input message by using context information, and electronic device supporting same

Abstract

A method for generating a message by an electronic device according to various embodiments of the present disclosure may comprise the operations of: executing an application; obtaining an input message for the application; selecting first context information related to the input message from context information that has been obtained by the electronic device prior to obtaining the input message and stored in a memory of the electronic device; determining, on the basis of the input message and the first context information, whether the input message conflicts with the first context information; when the input message conflicts with the first context information, generating a recommendation message on the basis of the first context information and the input message; and displaying the recommendation message on an execution screen of the application.

Description

A method for providing recommendation messages related to input messages using context information and an electronic device supporting the same

[0001] The present disclosure relates to a method for providing a recommendation message related to an input message using context information and an electronic device supporting the same.

[0002] With the recent rapid development of mobile messengers and social network services, the demand for real-time communication among users has increased significantly. Consequently, issues such as typos, contextual inappropriateness, and insufficient transmission of intent have emerged during message transmission and reception, creating a situation where existing simple text input methods alone are insufficient to effectively resolve these problems. Previously, solutions were limited to users manually editing messages or providing auto-completion features based on limited algorithms; however, these approaches often fail to adequately reflect the flow or context of complex conversations or may even provide incorrect information.

[0003] Meanwhile, along with the recent rapid advancements in the field of natural language processing, the emergence of Large Language Models (LLMs) is enabling the implementation of more sophisticated and context-aware text generation technologies. LLMs learn language usage patterns across various contexts based on vast amounts of training data, thereby acquiring the ability to more accurately understand large volumes of content and generate content suitable for them. Recently, such artificial intelligence is being utilized in providing a wide variety of content.

[0004] The information described above may be provided as related art for the purpose of aiding understanding of the present disclosure. No claim or determination is made as to whether any of the foregoing may be applied as prior art in relation to the present disclosure.

[0005] A method for generating a message using an electronic device according to various embodiments of the present disclosure may include: an operation of executing an application; an operation of obtaining an input message entered into the application in response to a first input to a transmission icon that transmits an input message to the application; an operation of selecting first context information related to the input message from context information obtained by the electronic device and stored in the memory of the electronic device prior to the input message being obtained; an operation of determining whether the input message conflicts with the first context information using the input message and the first context information; an operation of generating a recommendation message based on the first context information and the input message when the input message conflicts with the first context information; and an operation of displaying the recommendation message on an execution screen of the application.

[0006] An electronic device according to various embodiments of the present disclosure includes a memory storing context information and instructions, and at least one processor, wherein the instructions are executed by the at least one processor, and the electronic device: executes an application, obtains an input message for the application, selects a first context information related to the input message from the context information obtained by the electronic device prior to the input message and stored in the memory of the electronic device, determines whether the input message corresponds to the first context information using the input message and the first context information, and if the input message does not correspond to the first context information, generates a recommendation message based on the first context information and the input message, and displays the recommendation message on the execution screen of the application.

[0007] A computer-readable recording medium according to one embodiment may store a computer program that enables an electronic device to execute and perform the method described above.

[0008] FIG. 1 is a diagram illustrating an overview of an electronic device providing a recommendation message based on an input message using context information, according to one embodiment of the present disclosure.

[0009] FIG. 2 is a flowchart illustrating the operation of an electronic device generating a recommendation message using context information based on an input message, according to one embodiment of the present disclosure.

[0010] FIG. 3 is a diagram illustrating an example of an operation in which an electronic device acquires an input message for a messenger application or a memo application according to one embodiment of the present disclosure.

[0011] FIG. 4 is a diagram illustrating an example of an operation in which an electronic device acquires an input message containing image data according to one embodiment of the present disclosure.

[0012] FIG. 5 is a flowchart illustrating the operation of an electronic device generating a prompt according to one embodiment of the present disclosure.

[0013] FIG. 6 is a drawing for illustrating an example of a prompt input to an artificial intelligence model of an electronic device according to one embodiment of the present disclosure.

[0014] FIG. 7 is a drawing for illustrating an example of the configuration of an electronic device including an artificial intelligence model according to one embodiment of the present disclosure.

[0015] FIG. 8 is a diagram illustrating the types of context information stored in the memory of an electronic device according to one embodiment of the present disclosure.

[0016] FIG. 9 is a diagram illustrating an example of context information classified and stored according to the storage period of context information according to one embodiment of the present disclosure.

[0017] FIG. 10 is a diagram illustrating an example of context information classified and stored according to the time of creation of context information, according to one embodiment of the present disclosure.

[0018] FIG. 11 is a drawing for illustrating an example in which an electronic device generates a recommendation message based on search results through a server connected to the electronic device, according to one embodiment of the present disclosure.

[0019] FIG. 12 is a diagram illustrating an example of generating a recommendation message using context information based on a message transmitted by an electronic device to another electronic device according to one embodiment of the present disclosure.

[0020] FIG. 13 is a flowchart illustrating an operation to generate a recommendation message using context information based on an unacknowledged message among messages transmitted by an electronic device to another electronic device according to one embodiment of the present disclosure.

[0021] FIG. 14 is a diagram illustrating an example of an operation to generate a recommendation message using context information based on an unacknowledged message among messages transmitted by an electronic device to another electronic device according to one embodiment of the present disclosure.

[0022] FIG. 15 is a diagram illustrating the operation of an electronic device generating a recommendation message using a context selector according to one embodiment of the present disclosure.

[0023] FIG. 16 is a diagram showing a system including a generative artificial intelligence model according to one embodiment of the present disclosure.

[0024] FIG. 17 is a block diagram of an electronic device in a network environment according to various embodiments.

[0025] In relation to the description of the drawings, the same or similar reference numerals may be used for identical or similar components.

[0026] The technical problems to be solved in this document are not limited to those mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art to which this invention belongs from the description below.

[0027] Hereinafter, embodiments are described in detail with reference to the attached drawings so that those skilled in the art can easily implement the present invention. However, the disclosed embodiments may be implemented in various different forms and are not limited to the embodiments described herein. Furthermore, in order to clearly explain the disclosure in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification have been given similar reference numerals.

[0028] The terms used in this disclosure are described in their current, general form considering the functions mentioned herein; however, they may refer to various other terms depending on the intent of those skilled in the art, case law, the emergence of new technologies, etc. Accordingly, the terms used in this disclosure should not be interpreted solely by their names, but should be interpreted based on the meaning of the terms and the overall content of this disclosure.

[0029] Additionally, terms such as "first," "second," etc., may be used to describe various components, but the components are not limited by these terms. These terms are used for the purpose of distinguishing one component from another.

[0030] Throughout the specification, when a part is described as being "connected" to another part, this includes not only cases where they are "directly connected," but also cases where they are "electrically connected" or "operationally connected" with other elements interposed between them. Furthermore, when a part is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0031] Phrases such as "in one embodiment" appearing in various places in this disclosure do not necessarily refer to the same embodiment.

[0032] One embodiment of the present disclosure may be represented by functional block configurations and various processing steps. Some or all of these functional blocks may be implemented by various numbers of hardware and / or software configurations that execute specific functions. For example, the functional blocks of the present disclosure may be implemented by one or more microprocessors or by circuit configurations for a specific function. Additionally, for example, the functional blocks of the present disclosure may be implemented in various programming or scripting languages. The functional blocks may be implemented as algorithms executed on one or more processors. Furthermore, the present disclosure may employ prior art for electronic configuration, signal processing, and / or data processing, etc. Terms such as "mechanism," "element," "means," and "configuration" may be used broadly and are not limited to mechanical and physical configurations.

[0033] Furthermore, the connecting lines or connecting members between the components depicted in the drawings are merely illustrative of functional connections and / or physical or circuit connections. In the actual device, connections between components may be represented by various alternative or added functional connections, physical connections, or circuit connections.

[0034] In the present disclosure, context information may be information collected in relation to an electronic device and / or a user of the electronic device as information indicating a situation related to an electronic device.

[0035] Context information may include at least one of information obtained from an application of the electronic device (1801), surrounding environment information of the electronic device (1801), state information of the electronic device (1801), state information of the user, usage history information of the user's electronic device (1801), and schedule information of the user, but is not limited thereto. Information obtained from an application of the electronic device (1801) is information obtained from an application installed on the electronic device (1801), and may include conversation history information, user information, transmitted and received image information, transmitted and received file information, transmitted and received URL information, etc. obtained from a messenger application installed on the electronic device (1801), but is not limited thereto. The surrounding environment information of the electronic device (1801) refers to environmental information within a predetermined radius from the electronic device (1801), and may include, for example, weather information, temperature information, humidity information, illuminance information, noise information, sound information, etc., but is not limited thereto. The state information of the electronic device (1801) may include, but is not limited to, mode information of the electronic device (1801) (e.g., sound mode, vibration mode, silent mode, power saving mode, blocking mode, multi-window mode, automatic rotation mode, etc.), location information of the electronic device (1801), time information, activation information of a communication module (e.g., Wi-Fi ON / Bluetooth OFF / GPS ON / NFC ON, etc.), network connection status information of the electronic device (1801), and application information executed on the electronic device (1801) (e.g., application identification information, application type, application usage time, application usage cycle). The user's state information is information regarding the user's movement, lifestyle patterns, etc., and may include, but is not limited to, information regarding the user's walking state, exercise state, driving state, sleep state, user's mood state, etc.The user's electronic device (1801) usage history information is information regarding the history of the user's use of the electronic device (1801), and may include, but is not limited to, the execution history of an application, the history of functions executed in the application, the user's call history, and the user's text message history.

[0036] Context information is information that an electronic device can refer to when generating recommendation messages, and may include all information constituting the situation, such as message information, user profile information, time information, location information, and schedule information. However, it is evident that context information in this document is not limited to the examples described above.

[0037] In the present disclosure, the input message may be a message input by a user into an electronic device (1801). The input message may be, for example, a message input through a message application and transmitted to and from an external electronic device, but is not limited thereto. The input message in the present disclosure is not limited to a message intended for transmission to and from an external electronic device, and the input message may include a message utilized in various applications for various purposes such as memos, schedules, email transmission, and business collaboration.

[0038] Functions related to artificial intelligence according to the present disclosure are operated through a processor and memory. The processor may be composed of one or more processors. In this case, the one or more processors may be general-purpose processors such as CPUs, APs, and DSPs (Digital Signal Processors), graphics-dedicated processors such as GPUs and VPUs (Vision Processing Units), or artificial intelligence-dedicated processors such as NPUs. The one or more processors control the processing of input data according to predefined operation rules or artificial intelligence models stored in memory. If the one or more processors are artificial intelligence-dedicated processors, the artificial intelligence-dedicated processors may be designed with a hardware structure specialized for processing a specific artificial intelligence model. The predefined operation rules or artificial intelligence models are characterized by being created through learning. Here, being created through learning means that a predefined operation rules or artificial intelligence models configured to perform desired characteristics (or objectives) are created by a basic artificial intelligence model (or deep learning model) being trained using multiple training data by a learning algorithm. Such learning may be performed on the device itself where the artificial intelligence according to the present disclosure is executed, or it may be performed through a separate server and / or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the examples mentioned above.

[0039] In the present disclosure, an artificial intelligence model (or deep learning model) may be composed of a plurality of neural network layers. Each of the plurality of neural network layers has a plurality of weight values ​​and performs neural network operations through operations between the results of operations of a previous layer and the plurality of weights. The plurality of weights possessed by the plurality of neural network layers may be optimized by the learning results of the artificial intelligence model. For example, the plurality of weights may be updated so that the loss value or cost value obtained from the artificial intelligence model during the learning process is reduced or minimized. The artificial neural network may include a Deep Neural Network (DNN), and examples include, but are not limited to, CNN (Convolutional Neural Network), DNN (Deep Neural Network), RNN (Recurrent Neural Network), RBM (Restricted Boltzmann Machine), DBN (Deep Belief Network), BRDNN (Bidirectional Recurrent Deep Neural Network), or Deep Q-Networks.

[0040] The present disclosure will be described in detail below with reference to the attached drawings.

[0041] FIG. 1 is a diagram illustrating an overview of an electronic device (1801) providing a recommendation message based on an input message using context information, according to one embodiment of the present disclosure.

[0042] According to one embodiment, an electronic device (1801) may acquire an input message for an application and generate a recommendation message by modifying the input message using context information related to the input message. For example, in 110 of FIG. 1, the electronic device (1801) may acquire an input message and generate a recommendation message according to 160 of FIG. 1 by modifying the input message using context information stored in memory according to 140 of FIG. 1. The recommendation message generated by the electronic device (1801) may be a message generated by modifying the input message to correspond to the context information.

[0043] According to one embodiment, the electronic device (1801) checks whether an input message for an application conflicts (e.g., corresponds) with context information stored in the memory of the electronic device (1801), and if the input message conflicts with the context information (e.g., does not correspond), it can generate a recommendation message that modifies the input message so as not to conflict with the context information (e.g., corresponds). The electronic device (1801) can determine whether the information included in the input message is compatible with or conflicts with the context information stored in the memory of the electronic device (1801). For example, the operation of determining compatibility or conflict may include the operation of identifying that the context information contains information representing different actions (or schedules) that are incompatible at a certain time. If there is an action called a PT class at a specific time (e.g., 9 o'clock), the electronic device (1801) may determine that it is incompatible because a conflict occurs with an action that meets at the same specific time (e.g., 9 o'clock). The electronic device (1801) can determine whether there is a conflict by determining whether the input message and the first context information are incompatible based on the time and / or attributes of the actions included in the input message and the first context information. The electronic device (1801) can determine whether there is a conflict by considering the temporal elements of the schedule included in the input message and the first context information. For example, the information included in the input message "Let's see each other at 9 o'clock today." is incompatible with the context information "9 o'clock today: PT class," and the electronic device (1801) can determine that the input message does not correspond to the context information. The electronic device (1801) can generate a recommendation message by using an artificial intelligence model to modify the input message so that it does not conflict with the context information.For example, the recommendation message generated by the electronic device (1801) using an artificial intelligence model may be a message that modifies the input message "See you at 9 o'clock today, then." so as not to be compatible with or conflict with the context information "Today at 9 o'clock: PT class."

[0044] According to one embodiment, the electronic device (1801) can determine whether an input message conflicts with the first context information. Cases where the input message conflicts with the first context information may include cases where the input message contradicts the first context information. For example, cases where the content of the input message is mutually exclusive with the content of the first context information may include cases where the content of the input message and the content of the first context information are incompatible. For example, cases where the content of the input message and the content of the first context information are logically contradictory, or where the content of the input message and the content of the first context information cannot be established simultaneously. For example, cases where the input message is stated to satisfy a specific condition, but the corresponding condition is not satisfied in the first context information, may be cases where the input message conflicts with the first context information. Alternatively, cases where the input message conflicts with the first context information may include cases where the input message is inconsistent with the first context information. For example, this may include cases where the content of the input message does not match the content of the first context information, or where the numerical value and / or value of the input message differs from the numerical value and / or value of the first context information. For example, if the input message states "Reservation completed," but there is no reservation history in the first context information, this may be a case where the input message conflicts with the first context information. Alternatively, cases where the input message conflicts with the first context information may include cases where the content of the input message cannot be interpreted. For example, this may include cases where the content of the input message is unclear and therefore cannot be interpreted. For example, this may include cases where the content of the input message cannot be interpreted because it is technically impossible to establish.For example, this may include cases where the sentence structure of the input message is not grammatically correct or the terms used in the input message are combined illogically.

[0045] According to one embodiment, the electronic device (1801) can obtain an input message for an application. For example, referring to 110 in FIG. 1, the electronic device (1801) can obtain an input message "See you at 9 o'clock today, then." entered for the application. For example, although not shown in 110 in FIG. 1, the electronic device (1801) can obtain an input message entered for the application by receiving input for a message send button. In FIG. 1, the application is shown as a messenger application, but is not limited to a messenger application and may include other types of applications such as a memo application, a schedule application, etc.

[0046] According to one embodiment, the memory of the electronic device (1801) may store context information. The memory of the electronic device (1801) may store context information obtained by the electronic device (1801). The context information may include information necessary for the electronic device (1801) to generate a recommendation message. For example, referring to 140 in FIG. 1, the context information may include information regarding a schedule (calendar), information regarding traffic conditions (traffic info), information obtainable through chat (other chat session), information regarding weather (weather info), etc., which are stored in the memory of the electronic device (1801). However, it is not limited thereto.

[0047] According to one embodiment, an electronic device (1801) can construct a context database (145) using context information. The electronic device (1801) can construct a context database (145) using information stored in memory. For example, referring to 140 in FIG. 1, the electronic device (1801) can construct a context database (145) using information related to a schedule application, information related to a messenger application, traffic information, and / or weather information. The context database (145) may include context information. For example, the electronic device (1801) can construct a context database (145) including a query for using an artificial intelligence model and information stored in each application using information stored in memory.

[0048] In one embodiment, the context DB (145) may be created and / or updated dependently by the electronic device (1801) according to the user. For example, as the user moves and the location of the electronic device (1801) changes, the context DB (145) containing current location information may be created and / or updated. In one embodiment, the context DB (145) may be created and / or updated dependently by the electronic device (1801) according to time. For example, as the weather changes over time, the context DB (145) containing weather information may be updated dependently over time. According to one embodiment, the electronic device (1801) may select first context information related to an input message from the context information. The electronic device (1801) may select first context information related to an input message from the context information included in the context DB (145). The electronic device (1801) may select first context information related to an input message from the context information using an artificial intelligence model. The electronic device (1801) can select a first context information that is highly related to the input message among the context information by inputting the embedded input message and the embedded and stored context information into an artificial intelligence model and calculating the similarity. The electronic device (1801) may also input the embedded input message and the embedded and stored context information into an artificial intelligence model to obtain the first context information selected by the artificial intelligence model among the context information. However, the operation of the electronic device (1801) selecting the first context information from the context information is not limited to an operation using an artificial intelligence model. In one embodiment, the electronic device (1801) may select the first context information from the context information through vector search. In one embodiment, the electronic device (1801) may select the first context information through keyword search.The detailed operation of selecting the first context information from the context information will be described later.

[0049] According to one embodiment, an electronic device (1801) can determine whether an input message for an application corresponds to first context information. The electronic device (1801) can determine whether an input message for an application conflicts with the first context information. The electronic device (1801) can determine whether an input message corresponds to the first context information by using the first context information and the input message. The electronic device (1801) can identify the similarity between the input message and the first context information, and if the similarity is below a threshold, determine that the input message does not correspond to the first context information. The electronic device (1801) can determine that the input message conflicts with the first context information if the similarity is below a threshold. For example, referring to 130 in FIG. 1, the electronic device (1801) can determine whether an input message corresponds to the first context information by using an input message, "Let's see each other at 9 o'clock today, then," and first context information stored in memory. For example, if the first context information contained in the memory of 140 in FIG. 1 contains information such as "9 o'clock PT class," the electronic device (1801) determines the similarity between the text of the input message "See you at 9 o'clock today, then." and the information "9 o'clock PT class." Since the content of "PT class" and "appointment" at 9 o'clock is different from each other, it can determine that there is a low similarity. If the similarity is below a threshold, the electronic device (1801) can determine that the input message does not correspond to the first context information and conflicts. For example, if the first context information contained in the memory of 140 in FIG. 1 contains information such as "no schedule at 9 o'clock," the electronic device (1801) determines "See you at 9 o'clock today, then."By determining the similarity between the text of the input message "" and the information "no schedule at 9 o'clock," it can be determined that there is a high degree of similarity because the content of the input message regarding "appointment possible at 9 o'clock" and the information regarding "appointment" at 9 o'clock are similar. In the case where the similarity is greater than or equal to a threshold, the electronic device (1801) can determine that the input message corresponds to the first context information and does not conflict.

[0050] According to one embodiment, an electronic device (1801) can determine whether an input message conflicts (or corresponds) to the first context information using an artificial intelligence model. To determine whether an input message conflicts (or corresponds) to the first context information, the electronic device (1801) can identify similarity by applying the input message and the first context information to a trained artificial intelligence model. Based on the identified similarity, the electronic device (1801) can determine whether the input message conflicts (or corresponds) to the first context information. For example, the electronic device (1801) identifies the relevance of the meaning of the content included in the first context information to at least a part of the content of the input message, and if the value regarding relevance is higher than a reference value, the electronic device (1801) can determine that the input message does not conflict (or corresponds) to the first context information. For example, the electronic device (1801) may request an artificial intelligence model (e.g., LLM) to check whether there is a conflict between the input message and the first context information, and may determine whether the input message conflicts with the first context information by receiving from the artificial intelligence model whether the input message and the first context information conflict (e.g., whether they are reasonable, whether they conflict, whether they are appropriate). The artificial intelligence model (e.g., LLM) may determine whether the input message and the first context information conflict by taking the input message, the first context information, and a prompt regarding the conflict determination as inputs. The artificial intelligence model may determine whether the input message and the first context information conflict based on a prompt that includes conflict determination criteria, such as determining the similarity of the information or comparing the similarity with a threshold value. However, the method described above is not limited, and the electronic device (1801) may directly obtain a determination from the artificial intelligence model regarding whether the input message corresponds to the first context information.A more detailed explanation of the operation for determining whether an input message corresponds to the first context information using an artificial intelligence model will be provided later.

[0051] According to one embodiment, an electronic device (1801) can generate a recommendation message when an input message does not correspond to context information. The electronic device (1801) can generate a recommendation message by inputting the first context information and the input message to an artificial intelligence model trained to generate a recommendation message based on the input message and the first context information. The electronic device (1801) can generate a prompt to be input to a generative artificial intelligence model and generate a recommendation message by inputting the prompt to the generative artificial intelligence model. For example, referring to 160 in FIG. 1, if the input message "Let's see each other at 9 o'clock today, then." does not correspond to the first context information, the electronic device (1801) can use the artificial intelligence model to generate a recommendation message "Let's see each other at 10 o'clock today, then." For example, if the electronic device (1801) receives a result from an artificial intelligence model (e.g., LLM) that the input message does not correspond to context information, the electronic device (1801) can use the artificial intelligence model to receive a recommendation message generated based on the input message and the first context information. A more detailed explanation of the operation of generating recommendation messages using an artificial intelligence model will be provided later.

[0052] According to one embodiment, the electronic device (1801) can generate a plurality of recommendation messages. The electronic device (1801) can generate a plurality of recommendation messages using an artificial intelligence model. For example, referring to 160 in FIG. 1, if the input message "Let's see each other at 9 o'clock today," does not correspond to the first context information, the electronic device (1801) can generate a plurality of recommendation messages such as "Let's see each other at 10 o'clock today," and "Let's see each other at 11 o'clock today."

[0053] According to one embodiment, the electronic device (1801) may not generate a recommendation message if the input message corresponds to context information. If the electronic device (1801) determines that the input message corresponds to context information, it may not perform the operation of generating a recommendation message using an artificial intelligence model. For example, referring to 150 in FIG. 1, if the input message "Let's see each other at 9 o'clock today, then." corresponds to the first context information, the electronic device (1801) may not perform the operation of generating a recommendation message and may transmit the input message "Let's see each other at 9 o'clock today, then." to another electronic device (e.g., an external electronic device) through a messenger application. However, it is obvious that it is not limited to a messenger application.

[0054] According to one embodiment, the electronic device (1801) can display a recommendation message on the execution screen of an application. The electronic device (1801) can display a plurality of recommendation messages on the execution screen of an application. The electronic device (1801) can detect an event in which a recommendation message is selected and, in response to the event, transmit the recommendation message to an external electronic device. For example, referring to 160 in FIG. 1, the electronic device (1801) can display a plurality of recommendation messages, such as "Let's see each other at 10 o'clock today, then." and "Let's see each other at 11 o'clock today, then." along with an input message, "Let's see each other at 9 o'clock today, then." The electronic device (1801) can detect an event in which the recommendation message "Let's see each other at 10 o'clock today, then." is selected and transmit the recommendation message "Let's see each other at 10 o'clock today, then." to an external electronic device. However, it is not limited to the examples described above.

[0055] According to one embodiment, if an input message conflicts (does not correspond) to the first context information, the electronic device (1801) may display a notification message on the execution screen of an application indicating that the input message does not correspond to the first context information. For example, referring to 160 in FIG. 1, the electronic device (1801) may determine that the input message "See you at 9 o'clock today, then." conflicts (does not correspond) to the first context information, and the electronic device (1801) may display a notification message "There is a PT class at 9 o'clock today" on the execution screen of an application based on the first context information. By referring to the notification message displayed on the electronic device (1801) and recognizing that the input message conflicts with the first context information, the user may be able to quickly correct the input message that does not fit the context. Even when the user directly corrects the input message, the user may be able to correct the input message by referring to the recommendation message provided by the electronic device (1801), thereby shortening the time required to correct the input message.

[0056] FIG. 2 is a flowchart (200) illustrating the operation of an electronic device (1801) generating a recommendation message using context information based on an input message, according to one embodiment of the present disclosure.

[0057] In operation 210, the electronic device (1801) can run an application.

[0058] According to one embodiment, the application may include a messenger application. The electronic device (1801) may run a messenger application capable of sending and receiving messages with another electronic device.

[0059] According to one embodiment, the application may include a memo application. The electronic device (1801) may input and store data such as text data, image data, and voice data, and may execute a memo application capable of managing or searching for data.

[0060] According to one embodiment, the application may include a schedule management application. The electronic device (1801) may execute a schedule management application that allows inputting and managing schedules, appointments, to-dos, etc.

[0061] The applications that the electronic device (1801) can execute are not limited to the examples described above and may include other applications such as email applications, applications for work collaboration, and social media applications.

[0062] In operation 220, the electronic device (1801) can obtain an input message for the application.

[0063] According to one embodiment, the electronic device (1801) may acquire input messages of various formats. The input message may be a message containing text data. The input message may be a message containing image data. The input message may be a message containing video data. The input message may be a message in the form of a file. The input message may also include URL information. The input message is not limited to the examples described above.

[0064] According to one embodiment, the electronic device (1801) can obtain an input message in which input for an application is completed. For example, the electronic device (1801) can obtain an input message by receiving input for an icon indicating that message input is completed, which is displayed on the execution screen of the application. For example, the electronic device (1801) can obtain an input message by receiving input for an icon transmitting an input message, which is displayed on the execution screen of the application. For example, after a message is entered by a user in a messenger application, the electronic device (1801) can obtain a message entered in a message input window as an input message in response to receiving user input for a button transmitting a message. A detailed description of the operation in which the electronic device (1801) obtains an input message will be provided later.

[0065] In operation 230, the electronic device (1801) can select first context information related to an input message from context information.

[0066] According to one embodiment, context information stored in memory may include information necessary for the electronic device (1801) to generate a recommendation message.

[0067] According to one embodiment, context information may be embedded and stored in memory. The embedded context information may include information resulting from the conversion of context information into a high-dimensional vector form. The embedded information may be information that can be utilized to enable fast comparison and fast searching.

[0068] According to one embodiment, context information may be information obtained by the electronic device (1801) before an input message is obtained from the electronic device (1801). For example, context information may be information stored in memory before the input message is entered. However, context information is not limited to information obtained before the input message is obtained, and may also be obtained by the electronic device (1801) after the input message is obtained.

[0069] According to one embodiment, context information may be classified according to a pre-set category and stored in memory. The category may include the type of application related to the context information, the storage period of the context information, the storage time of the context information, the format of the context information, etc. The context information may be tagged with a tag indicating a category. For example, context information acquired by the electronic device (1801) through a messenger application may be classified according to the messenger application category and stored in memory, and may be tagged with a tag indicating the messenger application. For example, context information acquired by the electronic device (1801) through a schedule management application may be classified according to the schedule management application category and stored in memory, and may be tagged with a tag indicating the schedule management application. For example, context information generated in 2023 may be classified according to the 2023 category and stored in memory, and may be tagged with a tag indicating 2023. However, this is not limited to the examples mentioned above, and a detailed explanation of the context information classified and stored in memory according to category will be provided later.

[0070] According to one embodiment, context information may be classified according to long-term storage categories and short-term storage categories and stored in memory. In one embodiment, the long-term storage category may include context information that does not require real-time capabilities, and the short-term storage category may include context information that requires real-time capabilities. For example, context information including current location information or current weather information may be classified and stored according to the short-term storage category. For example, schedule information or conversation history information of a messenger application may be classified and stored according to the long-term storage category. In one embodiment, the long-term storage category may include yearly categories, monthly categories, daily categories, etc. For example, the long-term storage category may include a 2025 category, a 2024 category, etc.

[0071] According to one embodiment, an electronic device (1801) can select first context information related to an input message within context information stored in memory. The electronic device (1801) can select the first context information from the context information through at least one of the following methods: using a tag representing a category, using a keyword search, using a vector search through embedding, or using an artificial intelligence model. According to the present disclosure, by the electronic device (1801) selecting first context information related to an input message within the context information and performing an operation to generate a recommendation message using the first context information, the effect of quickly and efficiently generating a recommendation message by selecting only the first context information necessary for the operation of the electronic device (1801) from the context information can be achieved.

[0072] According to one embodiment, an electronic device (1801) can select first context information related to an input message from context information using a tag representing a category. The electronic device (1801) can identify a first tag related to an input message using an input message and select first context information tagged with the first tag from context information. For example, the electronic device (1801) can identify a first tag representing the category of 'schedule management application' for the input message "Let's meet at 9 o'clock today, then." and select first context information tagged with the first tag 'schedule management application' from context information.

[0073] According to one embodiment, an electronic device (1801) can select first context information related to an input message from context information through keyword search. The electronic device (1801) can extract keywords from an input message and select first context information from context information based on the extracted keywords. For example, the electronic device (1801) can select first context information that includes keywords matching the keywords extracted from the input message from among the context information. For example, the electronic device (1801) can select first context information that includes keywords similar to the keywords extracted from the input message and keywords similar to the extracted keywords from among the context information. However, the operation is not limited to the above-described operation.

[0074] According to one embodiment, an electronic device (1801) can select first context information related to an input message from context information through vector search utilizing embedding. The electronic device (1801) can embed and store context information in memory, embed an input message, and use the embedded input message and the embedded context information to select first context information that is highly related to the input message from the context information. For example, the electronic device (1801) can select first context information with high similarity among context information using the embedded input message and context information converted into a vector. For example, the electronic device (1801) can select first context information among context information through vector search (e.g., cosine similarity, etc.) using the embedded input message and the embedded context information. However, the operation is not limited to the above-described operation.

[0075] According to one embodiment, the electronic device (1801) may select first context information related to an input message from context information by utilizing an artificial intelligence model. The electronic device (1801) may select first context information from context information by utilizing an artificial intelligence model trained to select first context information that is highly relevant to the input message from context information. For example, the electronic device (1801) may train the artificial intelligence model to select first context information from context information according to priority based on the input message by utilizing a machine learning model. The electronic device (1801) may train the machine learning model by providing feedback on the selection result of the machine learning model. However, the operation is not limited to the above, and the first context information related to the input message may be selected from context information through other methods using an artificial intelligence model.

[0076] According to one embodiment, the electronic device (1801) can select a first context information from the context information by selecting a category based on an input message. The electronic device (1801) can select the first context information from the context information by selecting a short-term storage category or a long-term storage category based on an input message. For example, if the input message is "The weather is nice here right now," the electronic device (1801) can select the current weather information category. At this time, the electronic device (1801) can select the first context information classified according to the short-term storage category containing the current weather information category from the context information. For example, if the input message is "It rained a lot when I went to Jeju Island last year," the electronic device (1801) can select the 2024 category and / or 2024 weather information. At this time, the electronic device (1801) can select the first context information classified according to the long-term storage category containing the 2024 category from the context information. In one embodiment, according to the method described above, the electronic device (1801) selects a category based on an input message and selects a first context information classified according to the category from among the context information, thereby enabling the rapid and efficient generation of a recommendation message using the electronic device (1801) and improving the message generation speed and contextual relevance.

[0077] In operation 240, the electronic device (1801) can determine whether the input message corresponds to the first context information. The electronic device (1801) can determine whether the input message conflicts with the first context information.

[0078] According to one embodiment, whether an input message corresponds to (or does not conflict with) the first context information may be whether the input message and the first context information conflict. If the information included in the input message and the first context information are incompatible, the input message and the first context information may not correspond. If the information included in the input message and the first context information contain information representing different actions that are incompatible at a certain time, the input message and the first context information may conflict. For example, if the input message is "Let's meet at 6:00 PM on March 1, 2025" and the first context information is "March 1, 2025, 6:00 PM Schedule: Class," the input message and the first context information are incompatible, so it may be determined that they do not correspond and conflict. However, the above examples are not limited to this.

[0079] According to one embodiment, the electronic device (1801) can identify similarity between an input message and first context information. The electronic device (1801) can identify similarity between an input message and first context information by using at least one method among a method using keywords, a method using embedded information, a method using an artificial intelligence model, and a method analyzing semantics.

[0080] According to one embodiment, if the similarity between the input message and the first context information is greater than or equal to a threshold, the electronic device (1801) may determine that the input message corresponds to the first context information or does not conflict with it. According to one embodiment, if the similarity between the input message and the first context information is less than a threshold, the electronic device (1801) may determine that the input message does not correspond to the first context information. If the similarity between the input message and the first context information is less than a threshold, the electronic device (1801) may determine that the input message conflicts with the first context information.

[0081] In one embodiment, the electronic device (1801) can identify similarity using keywords included in an input message and keywords included in first context information. The electronic device (1801) can identify keywords in the input message and keywords in the first context information. The electronic device (1801) can compare keywords in the input message and the first context information. The electronic device (1801) can identify similarity between the identified keywords and determine that if the similarity is below a threshold, the input message does not correspond to (or conflicts with) the first context information. For example, the electronic device (1801) can identify similarity between the first context information and the input message by comparing the frequency or importance of the keywords. At this time, the electronic device (1801) can identify similarity using keywords in the input message and the first context information without embedding information and determine whether the input message corresponds to the first context information by comparing it with a threshold.

[0082] In one embodiment, the electronic device (1801) can identify similarity based on an embedded input message and embedded first context information. For example, the electronic device (1801) can convert the input message and the first context information into vectors and calculate the similarity between the vectors (e.g., cosine similarity, Euclidean distance) to identify the similarity between the first context information and the input message. According to one embodiment, the electronic device can identify similarity through vector calculation of the embedded input message and the embedded first context information without using an artificial intelligence model. For example, if the similarity identified using the vector between the embedded input message and the embedded first context information is greater than or equal to a threshold, the electronic device (1801) can determine that the input message corresponds to (or does not conflict with) the first context information. For example, the electronic device (1801) may determine that the input message conflicts with (or does not correspond to) the first context information if the similarity identified using a vector between the embedded input message and the embedded first context information is less than a threshold.

[0083] In one embodiment, the electronic device (1801) can identify the similarity between an input message and first context information using an artificial intelligence model. For example, the electronic device (1801) can identify the similarity using an artificial intelligence model trained to identify the similarity between an input message and first context information by taking the input message and first context information as inputs. According to one embodiment, the electronic device (1801) can determine whether the input message conflicts (or corresponds) with the first context information by inputting the input message and first context information into an artificial intelligence model trained to determine whether the input message conflicts (or corresponds) with the first context information. For example, the electronic device (1801) can determine whether the input message conflicts (or corresponds) with the first context information by inputting the input message and first context information into an artificial intelligence model trained to calculate vector similarity using the embedded input message and the embedded first context information, receiving similarity from the artificial intelligence model, and comparing the received similarity with a threshold. For example, the electronic device (1801) can determine whether the input message conflicts with (or corresponds to) the first context information by inputting the input message and the embedded first context information, which are embedded in an artificial intelligence model trained to determine whether the input message conflicts with (or corresponds to) the first context information by comparing the similarity identified using the input message and the first context information with a threshold, and by receiving the result.

[0084] In one embodiment, the electronic device (1801) can identify similarity by analyzing the meaning of the entire sentence of the input message and the first context information, and determine whether the input message conflicts with (or corresponds to) the first context information. For example, the electronic device (1801) can identify semantic similarity between the first context information and the input message by analyzing the meaning of the entire sentence. In one embodiment, the electronic device (1801) may input the input message and the first context information into an artificial intelligence model trained to analyze the meaning of the entire sentence and determine semantic similarity, and determine whether the input message corresponds to the first context information based on the semantic similarity output by the artificial intelligence model.

[0085] However, the method by which the electronic device (1801) identifies the similarity between the input message and the first context information and compares it with a threshold is not limited to the examples described above.

[0086] According to one embodiment, the electronic device (1801) can determine whether an input message containing text data and data other than text data conflicts (or corresponds) to the first context information. For example, the electronic device (1801) can determine whether an input message containing image data conflicts (or corresponds) to the first context information. For example, the input message may include image data, voice data, video data, URL data, etc., and the electronic device (1801) can determine whether the input message corresponds to the first context information. However, the types of data included in the input message are not limited to the examples described above and may include other types of data.

[0087] In one embodiment, in operation 240, if the electronic device (1801) determines that the input message corresponds to the first context information, it may perform operation 290. In operation 240, if the electronic device (1801) determines that the input message does not conflict with the first context information, it may perform operation 290. In one embodiment, in operation 240, if the electronic device (1801) determines that the input message does not correspond to the first context information, it may perform operation 250. In operation 240, if the electronic device (1801) determines that the input message conflicts with the first context information, it may perform operation 250.

[0088] In operation 250, the electronic device (1801) can generate a recommendation message based on an input message and first context information.

[0089] According to one embodiment, an electronic device (1801) can generate a recommendation message by inputting the first context information and the input message to an artificial intelligence model trained to generate a recommendation message based on the input message and the first context information. The artificial intelligence model trained to generate a recommendation message based on the input message and the first context information may be a generative artificial intelligence model. For example, the generative artificial intelligence model may include a Large Language Model (LLM). The artificial intelligence model according to the present disclosure is not limited to a specific artificial intelligence model, and various generative artificial intelligence models may be applied.

[0090] According to one embodiment, an electronic device (1801) can generate a prompt to be input to an artificial intelligence model based on an input message and first context information. The electronic device (1801) can generate a prompt for generating a recommendation message based on the input message and according to the first context information. For example, the prompt may include an input message for an application, first context information related to the input message, and a command for generating a recommendation message. For example, the electronic device (1801) may store the prompt in a structured form using a JSON data structure and generate the prompt based on the structured form according to the JSON data structure. However, an example of a prompt according to the present disclosure will be described later in FIG. 6.

[0091] According to one embodiment, the electronic device (1801) can generate a plurality of recommendation messages based on an input message and first context information. The electronic device (1801) can generate a plurality of recommendation messages by inputting a prompt containing a command for generating a plurality of recommendation messages into a generative artificial intelligence model.

[0092] According to one embodiment, the artificial intelligence model according to operation 250 may be an artificial intelligence model different from the artificial intelligence model that performs the operation according to operation 240 and / or operation 230. However, the artificial intelligence model according to operation 250 may be included within the same artificial intelligence model as the artificial intelligence model that performs the operation according to operation 240 and / or operation 230, and may be the same artificial intelligence model.

[0093] In operation 260, the electronic device (1801) can display a recommendation message on the application's execution screen.

[0094] According to one embodiment, the electronic device (1801) can display at least one recommendation message on the execution screen of an application. For example, if the electronic device (1801) generates at least one recommendation message based on an input message and first context information, it can display at least one recommendation message together.

[0095] According to one embodiment, an electronic device (1801) can display a plurality of recommendation messages on an application execution screen according to priority. The electronic device (1801) can display a recommendation message with a higher priority among the plurality of recommendation messages at the top based on first context information. For example, the electronic device (1801) can display a recommendation message with a high similarity among the plurality of recommendation messages at the top based on first context information. For example, the electronic device (1801) can display a recommendation message containing information with high user preference at the top based on first context information related to user preference. For example, the electronic device (1801) can display a more preferred recommendation message at the top by reflecting settings or user feedback.

[0096] According to one embodiment, the electronic device (1801) may display a message indicating that an input message conflicts with (or does not correspond to) the first context information, along with at least one recommendation message. For example, referring to 160 in FIG. 1, the electronic device (1801) may display two recommendation messages, “See you at 10 o’clock today, then.” (163) and / or “See you at 11 o’clock today, then.” (165), on the execution screen of the application for an input message, “There is a PT class at 9 o’clock today.” (167), and simultaneously display a message on the execution screen of the application indicating the cause (or reason) that the input message conflicts with (or does not correspond to) the first context information. The message indicating the cause (or reason) that the input message conflicts with (or does not correspond to) the first context information may include information related to the recommendation message. For example, a message indicating the cause (or reason why the input message conflicts with the first context information) may include at least one of the reason the recommendation message was generated or information regarding the part of the input message that conflicts with the first context information. As the electronic device (1801) displays the message indicating the cause (or reason why the input message conflicts with the first context information) along with the recommendation message, the user may be able to quickly and conveniently select the recommendation message.

[0097] According to one embodiment, the electronic device (1801) can display a recommendation message on the execution screen of an application, and at the same time, display an icon on the execution screen of the application that can select the displayed recommendation message.

[0098] In operation 270, the electronic device (1801) can determine whether the recommendation message has been selected.

[0099] According to one embodiment, the electronic device (1801) may determine that a recommended message has been selected in response to identifying input for a recommended message displayed on the execution screen of an application. For example, the electronic device (1801) may determine that a recommended message has been selected in response to identifying touch input for a certain period of time or longer for a recommended message displayed on the execution screen of an application.

[0100] According to one embodiment, the electronic device (1801) may determine that a recommended message has been selected in response to identifying an input for an icon that can select a recommended message displayed on an application execution screen. For example, the electronic device (1801) may determine that a recommended message has been selected in response to identifying a touch input for an icon displayed together with the recommended message.

[0101] According to one embodiment, the electronic device (1801) may determine that one recommendation message has been selected in response to identifying an input for one recommendation message among a plurality of recommendation messages. For example, the electronic device (1801) may identify a touch input, a slide input, a push input, etc., for one recommendation message among a plurality of recommendation messages.

[0102] According to one embodiment, if it is determined in operation 270 that a recommended message has been selected, the electronic device (1801) may perform operation 280. According to one embodiment, if it is determined in operation 270 that a recommended message has not been selected, the electronic device (1801) may perform operation 290.

[0103] In operation 280, the electronic device (1801) can transmit a recommendation message.

[0104] According to one embodiment, in a messenger application, an electronic device (1801) can transmit a recommendation message to another electronic device. For example, the electronic device (1801) can transmit a recommendation message, rather than an input message, from the messenger application to another electronic device. However, operation 280 describes an example of the operation of the electronic device (1801) in the messenger application, and the application according to the present disclosure is not limited thereto.

[0105] According to one embodiment, in a messenger application, an electronic device (1801) can display a recommended message in a message input window. Upon receiving input for a message transmission icon, the electronic device (1801) can transmit the recommended message displayed in the message input window to another electronic device. For example, the electronic device (1801) can display the selected recommended message instead of the input message entered in the message input window as the recommended message is selected, and upon receiving input for a message transmission icon, transmit the selected recommended message instead of the input message entered in the message input window to another electronic device.

[0106] According to one embodiment, in a memo application, an electronic device (1801) can store recommendation messages in the memo application. For example, the electronic device (1801) can display and / or store recommendation messages generated by the electronic device (1801), rather than input messages entered into the memo application.

[0107] According to one embodiment, in a schedule management application, an electronic device (1801) can input a recommendation message into the schedule management application. For example, the electronic device (1801) can input and save an appointment based on a recommendation message, rather than an input message, as an appointment for a specific date.

[0108] In one embodiment, it will be obvious that it is not limited to the examples described above and can be applied to other applications that can apply the recommendation message generated by the electronic device (1801).

[0109] In operation 290, the electronic device (1801) can transmit an input message.

[0110] According to one embodiment, in a messenger application, an electronic device (1801) can transmit an input message to another electronic device. For example, the electronic device (1801) can transmit an input message from a messenger application to another electronic device. However, operation 290 describes an example of the operation of the electronic device (1801) in a messenger application, and the application according to the present disclosure is not limited thereto.

[0111] According to one embodiment, in a memory application, an electronic device (1801) can store an input message in the memory application. For example, the electronic device (1801) can display and / or store an input message entered into the memory application.

[0112] According to one embodiment, in a schedule management application, an electronic device (1801) can input a recommendation message into the schedule management application. For example, the electronic device (1801) can input and save an appointment according to the input message as an appointment for a specific date.

[0113] In one embodiment, it will be obvious that it is not limited to the examples described above and can also be applied to other applications that can apply an input message displayed on the execution screen of an application of the electronic device (1801).

[0114] FIG. 3 is a diagram illustrating an example of an operation in which an electronic device (1801) obtains an input message for a messenger application or a memo application according to one embodiment of the present disclosure.

[0115] According to one embodiment, the electronic device (1801) can obtain an input message for a messenger application when an input message is entered into the messenger application. For example, the electronic device (1801) can obtain the completed input message when the input of an input message is completed in the messenger application.

[0116] According to one embodiment, the electronic device (1801) can obtain an input message entered into the messenger application by identifying an input to a send button on the messenger application execution screen. For example, referring to 310 in FIG. 3, the electronic device (1801) can obtain an input message entered into a message input window in the messenger application by identifying an input to a message send icon (301) that sends a message by identifying an input to a message send icon (301) that sends a message. Referring to 320 in FIG. 3, the electronic device (1801) can obtain “See you at 9 o’clock today, then.” entered into the message input window as an input message by identifying an input to a message send icon (301) in the messenger application.

[0117] According to one embodiment, an electronic device (1801) may display an icon for determining whether an input message conflicts with (or corresponds to) the first context information in a messenger application. The electronic device (1801) may display an icon for determining whether an input message conflicts with (or corresponds to) the first context information together with a virtual keyboard. For example, referring to 310 in FIG. 3, the electronic device (1801) may display a message check icon (311) on the execution screen of the application together with a virtual keyboard. Although the message check icon (311) is illustrated in FIG. 3, it is not limited to the illustrated figure, and the message check icon (311) may be omitted and a message transmission icon (301) may be displayed instead.

[0118] According to one embodiment, an electronic device (1801) determines whether an input message in a messenger application conflicts with (or corresponds to) the first context information, and if the input message conflicts with (or does not correspond to) the first context information, it may display an icon for generating a recommendation message. For example, referring to FIG. 3, the electronic device (1801) obtains an input message “Let’s see each other at 9 o’clock today, then.” entered by a user in a messenger application, performs a verification operation to determine whether the input message conflicts with (or corresponds to) the first context information, and if it determines that the input message conflicts with the first context information, it may display a message check icon (311). In FIG. 3, the message check icon (311) is shown displayed, but is not limited thereto. When the electronic device (1801) determines that an input message conflicts with the first context information, it may display the message check icon (311), and upon receiving input for the message check icon (311), generate a recommended message based on the first context information and the input message and display the recommended message on the execution screen of the messenger application. For example, the electronic device (1801) may perform a verification operation in the background to determine whether the input message conflicts with the first context information, and if it determines that there is a conflict, it may display the message check icon (311).

[0119] According to one embodiment, in response to an input to an icon (e.g., message sending icon (301) or message check icon (311)) for determining whether an input message conflicts (or corresponds) with first context information in a messenger application, an electronic device (1801) may perform an operation to determine whether an input message conflicts (or corresponds) with first context information. For example, by identifying an input to at least one of the message sending icon (301) or the message check icon (311), an input message may be obtained, and an operation to determine whether the input message conflicts with first context information may be performed. For example, referring to 320 in FIG. 3, the electronic device (1801) may perform an operation to determine whether an input message, “Let’s see each other at 9 o’clock today.” conflicts with first context information in response to an input to the message sending icon (301). For example, referring to 320 in FIG. 3, the electronic device (1801) can perform an operation to determine whether the input message “See you at 9 o’clock today.” conflicts with the first context information in response to an input to the message check icon (311).

[0120] According to one embodiment, while the electronic device (1801) performs an operation to determine whether an input message conflicts with first context information, it may display an effect (321) indicating that it is determining whether an input message conflicts with first context information on the execution screen of an application. For example, the electronic device (1801) may indicate that it is determining whether an input message conflicts with first context information by displaying a gradient effect. For example, the electronic device (1801) may indicate that it is determining whether an input message conflicts with first context information by displaying an animation effect. For example, even while the electronic device (1801) is generating a recommendation message, the electronic device (1801) may display an effect (321) indicating that it is generating a recommendation message based on an input message on the execution screen of an application, such as 321 in FIG. 3. However, the effect indicating that the electronic device is generating a recommendation message on the execution window of a messenger application is not limited to the illustrated drawings and may be displayed using various visual elements in addition to the gradient effect.

[0121] According to one embodiment, the electronic device (1801) can determine whether the acquired input message conflicts with the first context information, separately from receiving input for the message transmission icon (301). Even without receiving input for the message transmission icon (301), the electronic device (1801) can acquire a message entered in the message input window of the messenger application as an input message, and if the acquired input message conflicts with the first context information, it can generate a recommended message and display the recommended message adjacent to the message input window. The electronic device (1801) can modify the message entered in the message input window into a recommended message in response to an input in which the recommended message displayed adjacent to the message input window is selected. For example, even without receiving input for the message transmission icon (301), the electronic device (1801) can acquire a message entered in the messenger application in real time as an input message. Referring to 310 in FIG. 3, the electronic device (1801) can acquire “See you at 9 o’clock today, then.” entered in the message input window as an input message. The electronic device (1801) determines whether an input message entered in a message input window conflicts with the first context information, and if the input message entered in the message input window conflicts with the first context information, it generates a recommended message and can display the generated recommended message adjacent to the message input window. The electronic device (1801) can display the selected recommended message in the message input window in response to an input in which the recommended message is selected.

[0122] According to one embodiment, the electronic device (1801) can obtain an input message entered into the memo application when an input message is entered into the memo application. For example, the electronic device (1801) can obtain the completed input message when the input of an input message in the memo application is completed. For example, the electronic device (1801) can obtain the message entered into the memo window as an input message by identifying the input for the icon for saving the memo when a message is entered into the memo input window in the memo application. For example, the electronic device (1801) can obtain the message entered into the memo window as an input message by identifying the input for the message check (Msg check) icon (411) when a message is entered into the memo input window in the memo application.

[0123] According to one embodiment, the electronic device (1801) may display an icon for determining whether an input message corresponds to first context information in a memo application. The electronic device (1801) may display an icon for determining whether an input message corresponds to first context information together with a virtual keyboard. For example, referring to 410 in FIG. 3, the electronic device (1801) may display a message check (Msg check) icon (411) on the execution screen of the application together with a virtual keyboard.

[0124] According to one embodiment, in response to an input for an icon for determining whether an input message in a memo application corresponds to first context information, the electronic device (1801) may perform an operation to determine whether the input message conflicts with (or corresponds to) the first context information. For example, referring to 420 in FIG. 3, the electronic device (1801) may perform an operation to determine whether the input message “Meet Minsu in Gangnam at 9 o’clock today” conflicts with (or corresponds to) the first context information in response to an input for a message check (Msg check) icon (411). While determining whether the input message corresponds to the first context information, the electronic device (1801) may display on the application execution screen, as shown in 421 in FIG. 3, that it is determining whether the input message conflicts with (or corresponds to) the first context information. Even while the electronic device (1801) is generating a recommendation message, the electronic device (1801) may display on the application execution screen, as shown in 421 in FIG. 3, that it is generating a recommendation message based on the input message. However, examples of an electronic device displaying that it is generating a recommendation message in the execution window of a memo application are not limited to the illustrated drawings.

[0125] FIG. 4 is a drawing for illustrating an example of an operation in which an electronic device (1801) acquires an input message containing image data according to one embodiment of the present disclosure.

[0126] According to one embodiment, an electronic device (1801) can acquire an input message containing image data for a messenger application. The electronic device (1801) can acquire a message entered into a message input window as an input message by identifying an input for a transmission icon that transmits a message and a message entered into a message input window in the messenger application. For example, referring to 510 in FIG. 4, the electronic device (1801) can acquire an input message containing image data entered into a message input window by identifying an input for a transmission icon (511) that transmits a message.

[0127] According to one embodiment, an electronic device (1801) can determine whether an input message containing image data corresponds to the first context information by using an input message and first context information. For example, referring to FIG. 4, the electronic device (1801) can obtain an input message (503) containing image data for a plant image. Referring to FIG. 4, the electronic device (1801) can store a received message (501) saying "Send me a picture of my nephew" as context information and select it as the first context information related to the input message. Referring to FIG. 4, the electronic device (1801) can determine that the input message does not correspond to the first context information because the input message (503) is a plant image.

[0128] According to one embodiment, an electronic device (1801) can generate a recommendation message based on an input message and first context information. For example, referring to FIG. 4, the electronic device (1801) can generate a recommendation message based on a plant image included in the input message (503) and a person image of a gallery application included in the first context information. Referring to FIG. 4, the electronic device (1801) can generate a recommendation message (521) including image data regarding a baby among the image data of a gallery application.

[0129] According to one embodiment, an electronic device (1801) can display a recommendation message on the execution screen of a message application. For example, referring to 520 in FIG. 4, the electronic device (1801) can display a recommendation message (521) corresponding to a photo of a nephew on the execution screen of a message application. The electronic device (1801) can transmit the recommendation message (521) to another electronic device in response to the recommendation message (521) being selected.

[0130] FIG. 5 is a flowchart (600) illustrating the operation of an electronic device (1801) generating a prompt according to one embodiment of the present disclosure. FIG. 5 may be described with reference to the operation of FIG. 2. FIG. 6 is a diagram illustrating an example of a prompt input to an artificial intelligence model of an electronic device (1801) according to one embodiment of the present disclosure.

[0131] In operation 210, the electronic device (1801) can execute an application. Operation 210 of FIG. 5 may be an operation corresponding to operation 210 of FIG. 2.

[0132] In operation 220, the electronic device (1801) can obtain an input message for an application. Operation 220 of FIG. 5 may be an operation corresponding to operation 220 of FIG. 2.

[0133] In operation 230, the electronic device (1801) may select first context information related to an input message from context information. Operation 230 of FIG. 5 may be an operation corresponding to operation 230 of FIG. 2.

[0134] In operation 240, the electronic device (1801) can determine whether the input message corresponds to the first context information. Operation 240 of FIG. 5 may be an operation corresponding to operation 240 of FIG. 2.

[0135] According to one embodiment, if the electronic device (1801) determines in operation 240 that the input message does not correspond to the first context information, operation 610 may be performed.

[0136] In operation 610, the electronic device (1801) can generate a prompt based on an input message and first context information.

[0137] According to one embodiment, the prompt may include an input message, a category of first context information, and a command for generating a recommendation message.

[0138] According to one embodiment, a prompt generated by an electronic device (1801) may include a category of first context information selected by the electronic device (1801). The prompt may include a category related to the first context information selected by the electronic device (1801) among a plurality of categories related to context information. By including a category related to the first context information in the prompt generated by the electronic device (1801), the amount of unnecessary information that the artificial intelligence model must process can be reduced, and the response speed can be improved by minimizing the computational burden of the artificial intelligence model (e.g., LLM). For example, referring to 710 in FIG. 6, the prompt input to the artificial intelligence model may include “I think I will be free at 9 o’clock today” (711) as an input message, “CHAT” and “CALENDAR” as categories (713) of the first context information, and “Generate recommendation message” (714) as a command for generating a recommendation message. The prompt may include first context information classified according to a long-term storage category (712) in memory. The long-term storage category (712) may be a category for classifying non-volatile information stored in memory. In 710 of FIG. 6, the input message is a message regarding a schedule, and the prompt may include first context information classified according to a messenger application and a schedule management application.

[0139] According to one embodiment, the prompt may include a command for determining whether an input message corresponds to first context information and a command for generating a recommendation message. For example, referring to 720 in FIG. 6, the prompt may include a command (723) for determining whether an input message (721) corresponds to first context information (722) (determining whether it deviates from the context). Referring to 720 in FIG. 6, the prompt may include a command (724) for generating a recommendation message (generating a recommendation message if it deviates) when the input message does not correspond to the first context information. However, the information included in the prompt is not limited to the examples described above, and may additionally include the output format, priority, response method, etc. of the recommendation message.

[0140] According to one embodiment, the electronic device (1801) may be defined to generate a prompt using a specific data structure. For example, the electronic device (1801) may be defined to generate a prompt using a JSON format-based data structure. For example, referring to 710 in FIG. 6, the electronic device (1801) may generate a prompt in a structured format including an input message, a category of first context information, and a command to generate a recommendation message. By defining the generation of a prompt to be input into an artificial intelligence model using a specific data structure, the storage, modification, and expansion of the prompt are facilitated, and the effect of facilitating smooth data exchange with an external system can also be achieved. The specific data structure for generating the prompt is not limited to a JSON format-based data structure and may include other data structures such as XML and YAML.

[0141] According to one embodiment, the electronic device (1801) may include not only a single prompt but also multiple prompts when generating a prompt. By generating multiple prompts, the electronic device (1801) can enable a generative artificial intelligence model to generate recommendation messages that reflect richer context using multiple prompts. For example, referring to 720 in FIG. 6, the electronic device (1801) may generate a prompt that includes multiple prompts such as "determine if it is out of sync" and "if it is out of sync, generate a recommendation message for it."

[0142] In operation 620, the electronic device (1801) can input a prompt into a generative artificial intelligence model.

[0143] According to one embodiment, an electronic device (1801) inputs a prompt into a generative artificial intelligence model and can obtain a recommendation message from the generative artificial intelligence model. The generative artificial intelligence model receives a prompt from the electronic device (1801) and can generate a recommendation message based on an input message, first context information, and a command included in the prompt.

[0144] According to one embodiment, the electronic device (1801) may obtain a recommendation message from a generative artificial intelligence model after performing operation 620 and display the recommendation message. Operation 260 of FIG. 5 may be an operation corresponding to operation 260 of FIG. 2.

[0145] In operation 630, the electronic device (1801) can confirm the recommendation message and deliver the recommendation message.

[0146] According to one embodiment, the electronic device (1801) can confirm a recommendation message obtained from a generative artificial intelligence model. For example, the electronic device (1801) can receive a recommendation message generated by a generative artificial intelligence model and confirm the recommendation message.

[0147] According to one embodiment, an electronic device (1801) can transmit a recommendation message obtained and confirmed from a generative artificial intelligence model to a specific application. The electronic device (1801) can transmit a recommendation message generated by a generative artificial intelligence model to an application that has obtained an input message. For example, the electronic device (1801) can confirm a recommendation message generated by a generative artificial intelligence model and transmit it to a messenger application. For example, the electronic device (1801) can confirm a recommendation message generated by a generative artificial intelligence model and transmit it to a certain application. However, the type of application is not limited to the examples described above, and the electronic device (1801) can transmit the confirmed recommendation message to an application that has obtained an input message.

[0148] FIG. 7 is a drawing for illustrating an example of the configuration of an electronic device (1801) including an artificial intelligence model according to one embodiment of the present disclosure.

[0149] According to one embodiment, the prompt interface unit (813) of the electronic device (1801) can receive an input message (815) for the application from the application. The prompt interface unit (813) of the electronic device (1801) can select first context information related to the input message from context information stored in the context DB (814) of memory. The prompt interface unit (813) of the electronic device (1801) can select first context information related to the input message from context information stored in the context DB (814).

[0150] According to one embodiment, the prompt interface unit (813) of the electronic device (1801) can determine whether the input message corresponds to the first context information using the input message and the first context information. For example, the electronic device (1801) can determine whether the input message (815) corresponds to the first context information directly through the prompt interface unit (813).

[0151] According to one embodiment, the prompt interface portion (813, 823) of the electronic device (1801) can embed an input message (815, 825) and input it to an artificial intelligence model (811, 821, 822) along with first context information. The artificial intelligence model (811, 821) receives the input message (815, 825) and the first context information, receives a query regarding whether the input message (815, 825) corresponds to the first context information, and can output whether the input message (815, 825) corresponds to the first context information. For example, referring to 810 in FIG. 7, the prompt interface unit (813) of the electronic device (1801) can input the first context information and the embedded input message (815) to the artificial intelligence model (811) to query whether the input message (815) corresponds to the first context information, and the prompt interface unit (813) of the electronic device (1801) can receive a response to the query from the artificial intelligence model (811). For example, referring to 820 in FIG. 7, the prompt interface unit (823) of the electronic device (1801) can input the first context information and the embedded input message (825) to the artificial intelligence model (821) to query whether the input message (825) corresponds to the first context information. The prompt interface unit (823) of the electronic device (1801) can receive a response to the query from the artificial intelligence model (821).

[0152] According to one embodiment, the prompt interface unit (813) of the electronic device (1801) may input an input message (815) and first context information to the artificial intelligence model (811), and transmit a query regarding whether the input message (815) corresponds to the first context information, and a query requesting that a recommendation message (816) be output if the input message (815) does not correspond to the first context information. For example, referring to 810 of FIG. 7, the prompt interface unit (813) of the electronic device (1801) may input the first context information and the embedded input message (815) to the artificial intelligence model (811), and may make a query regarding whether the input message (815) corresponds to the first context information and a query requesting that a recommendation message (816) be output if the input message (815) does not correspond to the first context information. The prompt interface part (813) of the electronic device (1801) can receive responses to two queries from the artificial intelligence model (811).

[0153] According to one embodiment, the prompt interface unit (823) of the electronic device (1801) inputs an input message (825) and first context information to an artificial intelligence model (821), transmits a query regarding whether the input message (825) corresponds to the first context information, and transmits a query to another artificial intelligence model (822) to output a recommendation message (826) if the input message does not correspond to the first context information. For example, referring to 820 in FIG. 7, the prompt interface unit (823) of the electronic device (1801) may input the first context information and the embedded input message (825) to the artificial intelligence model (821) and ask whether the input message (825) corresponds to the first context information, and the prompt interface unit (823) of the electronic device (1801) may ask another artificial intelligence model (822) to output a recommendation message (826) if the input message (825) does not correspond to the first context information. The prompt interface unit (823) of the electronic device (1801) may receive responses to the two queries from the artificial intelligence model (821) and the other artificial intelligence model (822). Although the artificial intelligence model (821) and the other artificial intelligence model (822) are depicted as different models in the example above, they are not limited thereto and may be conceptually separated within a single artificial intelligence model.

[0154] According to one embodiment, the prompt interface portion (823) of the electronic device (1801) inputs an input message (825) and first context information to an artificial intelligence model (821), transmits a query regarding whether the input message (825) corresponds to the first context information, and transmits a candidate generation prompt to another artificial intelligence model (822) requesting that a recommendation message (826) indicating candidates for modifying the input message (825) be output if the input message does not correspond to the first context information. For example, referring to 820 in FIG. 7, the prompt interface unit (823) of the electronic device (1801) can input the first context information and the embedded input message (825) into the artificial intelligence model (821) and query whether the input message (825) corresponds to the first context information, and the prompt interface unit (823) of the electronic device (1801) can query using a candidate generation prompt that includes a query to output a recommendation message (826) indicating a candidate that can modify the input message (825) if the input message (825) does not correspond to the first context information. The prompt interface unit (823) of the electronic device (1801) may receive a response to a candidate generation prompt query from another artificial intelligence model (822). If there are changes to the numerical value in the response received from the other artificial intelligence model (822), the prompt interface unit (823) may determine the response in the format of the numerical response. The prompt interface unit (823) may display a plurality of recommendation messages (826) and, when user input is received, determine the response by switching the numerical value. For example, it may receive a slide input, a flip of the electronic device (1801), a slide of the electronic device (1801), or a touch input, and select a recommendation message from among the plurality of recommendation messages (826) by switching the numerical value.

[0155] According to one embodiment, the prompt interface portion (813, 823) of the electronic device (1801) can transmit recommendation messages (816, 826) to an application. For example, referring to FIG. 7, the prompt interface portion (813, 823) of the electronic device (1801) can transmit recommendation messages (816, 826) received from an artificial intelligence model (811, 822) to an application.

[0156] In one embodiment, the prompt interface section (813, 823) may be included in the processor of the electronic device (1801). For example, the processor of the electronic device (1801) may perform the operation of the prompt interface section (813, 823) of FIG. 7.

[0157] FIG. 8 is a diagram illustrating the type of context information stored in the memory of an electronic device (1801) according to one embodiment of the present disclosure.

[0158] According to one embodiment, context information stored in the context DB of the electronic device (1801) may include information acquired by the electronic device (1801). The context DB constructed by the electronic device (1801) using information from memory may include context information acquired by the electronic device (1801). For example, the context information may include information acquired through an application of the electronic device (1801). The context information may include information acquired by the electronic device (1801) through an external server and / or an external electronic device. For example, the electronic device (1801) may not merely store simple information in the context DB, but may control and manage information to store it in the context DB, and the context DB may correspond to a set of context information acquired by the electronic device (1801). For example, referring to FIG. 8, context information may include transmitted and received message information, profile information, transmitted and received file information, transmitted and received image data information, etc., obtained by the electronic device (1801) from the messenger application of the electronic device (1801). For example, referring to FIG. 8, context information may include schedule information, schedule information, etc., obtained by the electronic device (1801) from the schedule management application of the electronic device (1801). For example, referring to FIG. 8, context information may include location information, current location information of the electronic device (1801), route information, search information, place information, etc., obtained by the electronic device (1801) from the map application of the electronic device (1801). For example, referring to FIG. 8, weather information may include weather information, current weather information, past weather information, weather forecast information, weather information for an area of ​​interest, etc., obtained by the electronic device (1801) from the weather application of the electronic device (1801). Contextual information is not limited to the examples mentioned above.

[0159] According to one embodiment, the context DB may include context information, and the electronic device (1801) may control and manage the information included in the context DB. For example, the electronic device (1801) may construct a context DB using information in memory and perform management such as deleting, adding, creating, or editing the information included in the context DB over time. For example, the electronic device (1801) may generate a query using the first context information included in the context DB and store the query in the context DB.

[0160] According to one embodiment, context information stored in the memory of an electronic device (1801) may include volatile information and non-volatile information. In one embodiment, volatile information may be information where real-time is important, and may be information that requires immediate use by reflecting the current state. Volatile information may be stored for a short period and deleted or converted into non-volatile information after a certain period of time has passed or the environment has changed. Among the context information, volatile information may be classified according to a short-term storage category and stored in memory. For example, among the context information, volatile information may include current location information, current weather information, real-time traffic information, real-time user status, real-time financial information, etc. For example, referring to FIG. 8, among the context information, current location information and current weather information may be volatile information and may be classified according to a short-term storage category and stored in memory. In one embodiment, non-volatile information may be information where real-time is not important, and may be information that can be used for a long period of time. Non-volatile information may be stored for a long period of time and maintained even after a certain period of time has passed or the environment has changed. Among the context information, non-volatile information may be classified according to a long-term storage category and stored in memory. For example, among the context information, non-volatile information may include messenger conversation history, schedule information, user profile information, photo information, document information, etc. For example, referring to FIG. 8, among the context information, messenger conversation information and schedule information are non-volatile information and can be classified according to long-term storage categories and stored in memory. Short-term storage categories and long-term storage categories will be described later in FIG. 9.

[0161] According to one embodiment, the memory of the electronic device (1801) may store embedded context information. The context information may be embedded so that it can be input into an artificial intelligence model. For example, the electronic device (1801) may acquire context information and embed the acquired context information to store it in the memory of the electronic device (1801). For example, the electronic device (1801) may embed and store the context information so that it can input the context information into an artificial intelligence model to generate a recommendation message. For example, the electronic device (1801) may embed and store the context information so that it can input the context information into an artificial intelligence model to determine whether an input message corresponds to the first context information. As another example, the electronic device (1801) may embed and store the context information so that it can determine whether an input message corresponds to the first context information using the embedded context information even without inputting it into an artificial intelligence model.

[0162] According to one embodiment, the electronic device (1801) can select a first context information using embedded context information stored in memory. The electronic device (1801) can determine whether an input message corresponds to the first context information using the embedded context information.

[0163] According to one embodiment, the memory of the electronic device (1801) may store context information that is not embedded. Although FIG. 8 illustrates that context information is stored in memory by embedding it, this is exemplary, and in one embodiment, the electronic device (1801) may store context information in memory without embedding it. The electronic device (1801) may select a first context information using the context information that is not embedded. The electronic device (1801) may determine whether an input message corresponds to the first context information using the context information that is not embedded. For example, the electronic device (1801) may select the first context information within the context information through keyword search using the context information that is not embedded.

[0164] FIG. 9 is a diagram illustrating an example of context information classified and stored according to the storage period of context information according to one embodiment of the present disclosure.

[0165] According to one embodiment, the electronic device (1801) can classify and store context information according to the storage period of the context information. The electronic device (1801) can classify context information with a long storage period according to a long-term storage category and store it in the long-term DB (1010) of memory, and the electronic device (1801) can classify context information with a short storage period according to a short-term storage category and store it in the short-term DB (1020) of memory. For example, referring to FIG. 9, the electronic device (1801) can classify context information obtained from a messenger application and context information obtained from a schedule management application as information that needs to be stored for a long period and store it in the long-term DB (1010). For example, referring to FIG. 9, the electronic device (1801) can classify current location information obtained from a map application and current weather information obtained from a weather application as information that is stored for a short period where real-time is important and store it in the short-term DB (1020) according to a short-term storage category. However, it is not limited to the examples described above.

[0166] According to one embodiment, the electronic device (1801) may classify and store context information according to variability or volatility. The electronic device (1801) may classify and store context information according to real-time. In one embodiment, volatile information may be information where real-time is important, and may be information that requires immediate use by reflecting the current state. Volatile information may be stored for a short period and deleted or converted into non-volatile information after a certain period of time or when the environment changes. Among the context information, volatile information may be classified according to the short-term storage category and stored in the short-term DB (1020) of memory. For example, referring to FIG. 9, the electronic device (1801) may classify the current location information (1021) and current weather information (1022), which are volatile information, according to the short-term storage category and store them in the short-term DB (1020). In one embodiment, non-volatile information may be information where real-time is not important, and may be information that can be utilized for a long period of time. Non-volatile information may be stored for a long period of time and maintained even after a certain period of time or when the environment changes. Among the context information, non-volatile information can be classified according to long-term storage categories and stored in a long-term DB (1010) in memory. For example, referring to FIG. 9, the electronic device (1801) can classify non-volatile information, such as conversation content information (1011) of a messenger application and schedule information (1012) of a schedule management application, according to long-term storage categories and store them in a long-term DB (1010).

[0167] According to one embodiment, the context DB may include a long-term DB and a short-term DB, and the electronic device (1801) may store volatile context information in the short-term DB (1020) of memory and non-volatile context information in the long-term DB (1010) of memory. According to another embodiment, the electronic device (1801) may store volatile context information in a short-term storage memory (e.g., ram) and non-volatile context information in a long-term storage memory (e.g., rom). However, it is not limited to the examples described above, and may be stored in a single memory with only different categories.

[0168] According to one embodiment, by classifying and storing context information according to long-term storage categories and short-term storage categories, the electronic device (1801) can efficiently and quickly select first context information related to an input message from a vast amount of context information. Additionally, by storing volatile context information in a fast short-term DB and non-volatile context information in a long-term DB capable of storing a large amount of information, speed can be optimized and the storage space of the electronic device (1801)'s memory can be efficiently utilized. Furthermore, by storing context information classified by category within memory, data protection and recovery are facilitated, and the effect of quickly generating recommendation messages can be achieved.

[0169] According to one embodiment, context information is classified according to categories, and may be tagged with a tag representing the category. For example, referring to FIG. 9, context information classified according to long-term storage categories may be stored by embedding, and may be tagged with a tag representing the long-term storage category in natural language. For example, referring to FIG. 9, context information classified according to short-term storage categories may be stored by embedding, and may be tagged with a tag representing the short-term storage category in natural language.

[0170] Although FIG. 9 illustrates that context information is stored by being classified according to the storage period, it is not limited thereto. According to one embodiment, the electronic device (1801) may classify the context information according to categories based on its attributes and store it in a context DB in memory. Before receiving an input message, the electronic device (1801) classifies the context information according to its attributes into separate categories and stores it in a separate DB included in the context DB. By selecting the first context information classified according to the category of attributes related to the input message, it can determine whether the input message corresponds to the first context information and utilize only the information essential for generating a recommendation message. According to one embodiment, the attributes of the context information may include temporal attributes, purpose of use, data type, security level, creator, associated target, etc. For example, the context information may be classified according to categories such as volatility, non-volatility, short-term storage, and long-term storage based on temporal attributes. For example, the context information may be classified according to categories such as for searching and log tracking based on attributes regarding the purpose of use. For example, context information can be classified according to categories such as public information, one-time information, and restricted access information based on attributes regarding security levels. For example, context information can be classified according to categories such as user-generated information, system-generated information, and external source information based on attributes regarding the creator. However, it is not limited to the examples described above.

[0171] FIG. 10 is a diagram illustrating an example of context information classified and stored according to the time of creation of context information, according to one embodiment of the present disclosure.

[0172] According to one embodiment, the electronic device (1801) can classify context information according to the time of creation of the context information and store it in a DB in memory. The electronic device (1801) can classify and store context information according to the time of acquisition of the context information. For example, referring to FIG. 10, the electronic device (1801) can classify context information generated or acquired from a messenger application in 2024 according to the 2024 category and store it in the 2024 DB (1130). For example, referring to FIG. 10, the electronic device (1801) can classify context information generated or acquired from a messenger application in 2025 according to the 2025 category and store it in the 2025 DB (1120). However, the example of classifying and storing according to yearly categories is not limited to this, and it may also be classified and stored according to monthly categories or daily categories.

[0173] According to one embodiment, regarding context information stored in a DB of memory classified according to a category, the electronic device (1801) can change the category and store it in another DB of memory. The electronic device (1801) can change the context information stored in a short-term DB (1110) classified according to a short-term storage category to a long-term storage category and store it in a long-term DB of memory. For example, referring to FIG. 10, regarding context information classified according to a short-term storage category, the electronic device (1801) can change the category to a long-term storage category or a 2025 category so that it is classified according to a long-term storage category or a 2025 category and stored in a 2025 DB (1120) of memory. As the category is changed, the context information can be moved from the context DB to a DB corresponding to the changed category and stored.

[0174] According to one embodiment, an electronic device (1801) may classify context information stored in a short-term DB (1110) that is classified according to a short-term storage category, and among the context information whose data period has expired, classify context information according to a long-term storage category and store it in a long-term DB (1120, 1130). The data period may refer to a period set for the context information to be classified and stored in memory according to the corresponding category. The data period may be a pre-set period, or it may be determined variably depending on the characteristics, importance, frequency of use, storage capacity of memory, etc. of the context information. For example, context information regarding the current location of an electronic device obtained from a map application may be classified according to a short-term storage category and stored in the short-term DB (1110) of memory. The data period for context information regarding the current location may be set to 1 hour. As the data period for context information regarding the current location expires, the category of the context information regarding the current location may be changed from the short-term storage category to the 2025 category, classified according to the 2025 category, and stored in the 2025 DB (1120) of memory. For example, referring to FIG. 10, context information stored in the short-term DB (1110) in memory, classified according to the short-term storage category, can be moved to and stored in the 2025 DB (1120) in memory, classified according to the 2025 category.

[0175] According to one embodiment, the electronic device (1801) can convert context information classified according to a short-term storage category into context information classified according to a long-term storage category by changing a tagged tag for context information whose data expiration date has expired among context information classified according to a short-term storage category. The electronic device (1801) can convert context information classified according to a short-term storage category into context information classified according to a year category by changing a tagged tag for context information classified according to a short-term storage category. For example, referring to FIG. 10, the electronic device (1801) can convert context information tagged with a short-term storage tag into context information classified according to a 2025 category by changing a tag tagged with a 2025 tag for context information tagged with a short-term storage tag.

[0176] According to one embodiment, the electronic device (1801) can select first context information related to an input message from context information stored in memory. The electronic device (1801) can select first context information from the DB in memory based on a category related to the input message. For example, regarding an input message "The weather is nice here right now," the electronic device (1801) can identify that the input message related to current weather information is related to a short-term storage category. The electronic device (1801) can select first context information related to weather from among the context information stored in the short-term DB (1110) in memory, which is classified according to the short-term storage category. As another example, regarding an input message "It rained a lot when I went to Jeju Island last year," the electronic device (1801) can identify that the input message is related to a long-term storage category regarding the year 2024. The electronic device (1801) can identify that it is related to the 2024 category among the long-term storage categories, and can select first context information related to weather and / or Jeju Island from the context information stored in the 2024 DB (1130) of memory that is classified according to the 2024 category. As described above, the electronic device (1801) can quickly and efficiently select first context information from each DB of memory based on the category related to the input message, and can perform operations based on optimized first context information by selecting only the first context information related to the input message.

[0177] FIG. 11 is a drawing for illustrating an example in which an electronic device (1801) generates a recommendation message (1216) based on a search result through a server (1210) connected to the electronic device (1801), according to one embodiment of the present disclosure.

[0178] According to one embodiment, an electronic device (1801) can generate a search term based on an input message (1215) and first context information. The electronic device (1801) can generate a search term using keywords of the input message (1215). The electronic device (1801) can generate a search term to obtain information not present in the first context information. For example, if the input message (1215) is “KIM was born in the 70s, isn’t he?”, the electronic device (1801) can generate a search term “KIM’s age” based on the input message (1215) and first context information.

[0179] According to one embodiment, an electronic device (1801) may acquire an input message (1215) and determine whether it is necessary to generate a search term. The electronic device (1801) may determine that it is necessary to generate a search term if there is no first context information related to the input message (1215). The electronic device (1801) may determine that it is necessary to generate a search term if the first context information related to the input message is not sufficient to generate a recommendation message (1216). However, it is not limited to the above-described embodiment, and the electronic device (1801) may determine that it is necessary to generate a search term at any time regardless of whether there is first context information related to the input message (1215). For example, if the input message (1215) is “KIM was born in the 70s, isn’t he?”, the electronic device (1801) may determine that it is necessary to generate a search term if there is no “KIM’s age information” in the context information stored in the context DB (1214) of memory. As another example, if the input message (1215) is “KIM was born in the 70s,” the electronic device (1801) may determine that it needs to generate a search term even if there is “KIM’s age information” in the context information stored in the context DB (1214) of memory.

[0180] According to one embodiment, an electronic device (1801) can perform a search operation using a search term through a connected server (1210) and obtain a search result from the connected server (1210). The electronic device (1801) can perform a search operation using a search term generated through the server (1210) connected to the electronic device (1801). The electronic device (1801) can transmit a search request containing a search term to an external server (1210). The external server (1210) can search for related information by utilizing an internal algorithm based on the received search term. The electronic device (1801) can obtain a search result obtained using a search term from the connected server (1210). For example, referring to FIG. 11, the electronic device (1801) can transmit a search request containing a search term to the server (1210) connected to the electronic device (1801) and obtain a search result from the server (1210). For example, an electronic device (1801) can transmit a search request containing a search term to an external search site (1210), perform a search operation using the search term at the external search site (1210), and obtain a search result from the external search site (1210). For example, an electronic device (1801) can transmit a search request containing a search term to an API (Application Programming Interface) server (1210), perform a search operation using the search term at the API server (1210), and obtain a search result from the API server (1210).

[0181] According to one embodiment, the search operation performed by the electronic device (1801) through the server (1210) and the operation of determining whether an input message corresponds to the first context information may be performed simultaneously. For example, the electronic device (1801) may generate a search term and transmit the search term to an external server (1210) to perform a search operation, while the electronic device (1801) may perform an operation of determining whether an input message (1215) corresponds to the first context information using an artificial intelligence model (1211). In one embodiment, the search operation performed by the electronic device (1801) through the server (1210) and the operation of determining whether an input message (1215) corresponds to the first context information may be performed sequentially. For example, after the electronic device (1801) performs an operation to determine whether an input message (1215) conflicts with first context information using an artificial intelligence model (1211), if it determines that the input message (1215) conflicts with first context information, it may perform a search operation to obtain additional context information from an external server (1210). However, it is not limited to the example described above.

[0182] According to one embodiment, an electronic device (1801) can generate a recommendation message (1216) based on an acquired search result, an input message (1215), and first context information. The electronic device (1801) can generate a recommendation message (1216) by inputting the search result, an input message (1215), and first context information acquired from a connected server (1210) into an artificial intelligence model (1210). For example, the electronic device (1801) can generate a recommendation message (1216) based on an input message (1215) and a search result through an external server (1210) by inputting a search result including an input message (1215) saying “KIM was born in the 70s,” first context information, and KIM’s age information into an artificial intelligence model (1210).

[0183] According to one embodiment, the electronic device (1801) can modify the recommendation message (1216) generated based on the input message (1215) using search results obtained from a server connected to the electronic device (1801). For example, referring to FIG. 11, when the electronic device (1801) generates a recommendation message and then receives search results from the server (1210), the electronic device (1801) can generate a modified recommendation message (1216) based on the recommendation message and the search results.

[0184] FIG. 12 is a diagram illustrating an example of generating a recommendation message using context information based on a message transmitted by an electronic device (1801) to another electronic device according to one embodiment of the present disclosure.

[0185] According to one embodiment, the electronic device (1801) can transmit a message to an external electronic device and then collect and store context information. The electronic device (1801) can transmit a recommendation message obtained using an artificial intelligence model to an external electronic device and then update the context information. For example, the electronic device (1801) can transmit a recommendation message, “See you at 9 o’clock today, then.” to an external electronic device and then collect context information, “9 o’clock today: PT class,” from a schedule management application and store it in memory. At this time, the electronic device may need to generate a recommendation message based on the updated context information.

[0186] According to one embodiment, the electronic device (1801) can receive input for a message transmitted to an external electronic device. The electronic device (1801) can receive touch input for a message transmitted to an external electronic device. For example, referring to 1310 in FIG. 12, the electronic device (1801) can receive touch input for a message “See you at 9 o’clock today, then.” transmitted to an external electronic device. However, input for a message transmitted to an external electronic device is not limited to the example described above. In one embodiment, while the electronic device (1801) is checking an input message and generating a recommendation message, it can display an effect indicating that the input message is being checked. For example, referring to 1320 in FIG. 12, when the electronic device (1801) is performing at least one of the operations of determining whether an input message corresponds to first context information or generating a recommendation message based on the first context information and the input message, the electronic device (1801) may display a visual effect including a gradient effect or a shading effect for an input message such as “Let’s see each other at 9 o’clock today.” as in 1320 of FIG. 12.

[0187] According to one embodiment, an electronic device (1801) can acquire a message transmitted to an external electronic device as an input message. The electronic device (1801) can acquire a message transmitted to an external electronic device as an input message in response to identifying an input for a message transmitted to an external electronic device. The electronic device (1801) can select first context information related to the input message. For example, referring to FIG. 12, the electronic device (1801) can select first context information related to the input message “See you at 9 o’clock today, then.”

[0188] According to one embodiment, an electronic device (1801) can determine whether an input message corresponds to the first context information by using an input message and first context information. For example, referring to 1330 in FIG. 12, the electronic device (1801) can determine that the input message does not correspond to the first context information by using an input message, “See you at 9 o’clock today, then.” and first context information, “9 o’clock today: PT class.” Specific operations by which the electronic device (1801) determines whether an input message corresponds to the first context information by using an input message and first context information may correspond to the operations described above in the present disclosure.

[0189] According to one embodiment, the electronic device (1801) may not generate a recommendation message if the input message corresponds to the first context information using the input message and the first context information. For example, referring to 1350 in FIG. 12, the electronic device (1801) may check the input message and, if the input message corresponds to the first context information, not generate a recommendation message and not transmit an additional message.

[0190] According to one embodiment, the electronic device (1801) can generate a recommendation message using an artificial intelligence model when the input message does not correspond to the first context information using an input message and first context information. For example, referring to 1360 in FIG. 12, if the electronic device (1801) determines that the input message does not correspond to the first context information using an input message “See you at 9 o’clock today, then.” and first context information “9 o’clock today: PT class,” it can generate a recommendation message “Sorry, but I forgot I have an appointment. See you at 9 o’clock tomorrow.” using the input message and first context information. Specific operations of the electronic device (1801) generating a recommendation message using an artificial intelligence model may correspond to the operations described above in the present disclosure.

[0191] According to one embodiment, after the electronic device (1801) transmits a message to an external electronic device, if the context DB is updated, it can determine whether the transmitted message conflicts with the context information by using the context information included in the updated context DB. Even if the electronic device (1801) transmits a message to an external electronic device after performing an input message check, if the electronic device (1801) receives new context information and the context DB is updated accordingly, it can determine whether the transmitted message conflicts with the first context information by using the first context information including the newly received context information and the transmitted message. For example, if, after checking the input message according to 1320 of FIG. 13, the message “Let’s see each other at 9 o’clock today” is transmitted to an external electronic device, and then context information indicating that an appointment has been made at 9 o’clock today is received, the electronic device (1801) can determine at 1330 that the input message “Let’s see each other at 9 o’clock today” conflicts with the first context information “9 o’clock appointment today” included in the updated context DB, by using the first context information and the input message based on the updated context DB, and determine that the input message conflicts with the first context information by using an artificial intelligence model, and can transmit the recommendation message “Sorry, but I forgot I have an appointment. Let’s see each other at 9 o’clock tomorrow.” to the external electronic device.

[0192] FIG. 13 is a flowchart (1400) illustrating an operation in which an electronic device (1801) according to one embodiment of the present disclosure generates a recommendation message using context information based on an unidentified message among messages transmitted to another electronic device.

[0193] In operation 210, the electronic device (1801) can run an application.

[0194] According to one embodiment, the application may include a messenger application. However, it is not limited to a messenger application.

[0195] According to one embodiment, the operation 210 of FIG. 13 may be an operation corresponding to the operation 210 of FIG. 2.

[0196] In operation 1410, the electronic device (1801) can determine whether there are any unacknowledged messages among the messages transmitted to the external electronic device that have not been acknowledged by the user of the external electronic device.

[0197] According to one embodiment, the electronic device (1801) can determine that among the messages transmitted to the external electronic device, the user of the external electronic device has not acknowledged the message. For example, if the number 1 displayed next to the message transmitted to the external electronic device from a messenger application does not disappear, the electronic device (1801) can determine that the message is not acknowledged by the user of the external electronic device. For example, the electronic device (1801) can determine that the user of the external electronic device has not acknowledged the message through an acknowledgment log for the message transmitted to the external electronic device. For example, the electronic device (1801) can determine that the user of the external electronic device has not acknowledged the message by using a server-based acknowledgment signal function.

[0198] According to one embodiment, if there is an unacknowledged message among the messages transmitted to the external electronic device that has not been acknowledged by the user of the external electronic device, the electronic device (1801) may perform operation 1420. According to one embodiment, if there is no unacknowledged message among the messages transmitted to the external electronic device that has not been acknowledged by the user of the external electronic device, the electronic device (1801) may terminate operation.

[0199] In operation 1420, the electronic device (1801) can obtain an unidentified message as an input message.

[0200] According to one embodiment, the electronic device (1801) can acquire an unacknowledged message as an input message among the messages transmitted to an external electronic device that has not been acknowledged by the user of the external electronic device. The electronic device (1801) can acquire the unacknowledged message as an input message in response to identifying that there is an unacknowledged message.

[0201] According to one embodiment, the electronic device (1801) identifies an input for selecting one unidentified message among a plurality of unidentified messages, and as the input for selecting one unidentified message is identified, one unidentified message can be obtained as an input message.

[0202] In operation 230, the electronic device (1801) can select first context information related to an input message from context information.

[0203] According to one embodiment, the operation 230 of FIG. 13 may be an operation corresponding to the operation 230 of FIG. 2.

[0204] In operation 240, the electronic device (1801) can determine whether the input message corresponds to the first context information.

[0205] According to one embodiment, if the input message does not correspond to the first context information, the electronic device (1801) may perform operation 250. According to one embodiment, if the input message corresponds to the first context information, the electronic device (1801) may terminate the operation.

[0206] According to one embodiment, the operation 240 of FIG. 13 may be an operation corresponding to the operation 240 of FIG. 2.

[0207] In operation 250, the electronic device (1801) can generate a recommendation message based on an input message and first context information.

[0208] According to one embodiment, the operation 250 of FIG. 13 may be an operation corresponding to the operation 250 of FIG. 2.

[0209] In operation 260, the electronic device (1801) can display a recommendation message on the application's execution screen.

[0210] According to one embodiment, operation 260 of FIG. 13 may be an operation corresponding to operation 260 of FIG. 2.

[0211] In operation 270, the electronic device (1801) can determine whether a recommendation message displayed on the execution screen of the application has been selected.

[0212] According to one embodiment, when a recommendation message is selected, the electronic device (1801) can perform operation 1430. According to one embodiment, when a recommendation message is not selected, the electronic device (1801) can terminate the operation.

[0213] According to one embodiment, the operation 270 of FIG. 13 may be an operation corresponding to the operation 270 of FIG. 2.

[0214] In operation 1430, the electronic device (1801) can cancel the transmission of an unacknowledged message sent to an external electronic device and send a recommendation message to an external electronic device.

[0215] According to one embodiment, the electronic device (1801) may cancel the transmission of an unacknowledged message sent to an external electronic device and, in response to the cancellation of the transmission of the unacknowledged message, transmit a recommendation message to the external electronic device. For example, the electronic device (1801) may send a request to a server to cancel the transmission of an unacknowledged message and, in response to the server acknowledging that the request to cancel the transmission of the unacknowledged message has been approved, transmit a generated recommendation message to the external electronic device. The electronic device (1801) may transmit a recommendation message to the external electronic device in place of the unacknowledged message.

[0216] FIG. 14 is a diagram illustrating an example of an operation to generate a recommendation message using context information based on an unidentified message among messages transmitted by an electronic device (1801) to another electronic device according to one embodiment of the present disclosure.

[0217] According to one embodiment, the electronic device (1801) can determine whether the message transmitted to the external electronic device is a message that has not been checked by the user of the external electronic device. For example, in 1510 of FIG. 14, the electronic device (1801) can determine whether the message transmitted to the external electronic device, “Let’s see each other at 9 o’clock today, then.” is a message that has not been checked by the user of the external electronic device, and determine whether it is a message that has not been checked by the user of the external electronic device.

[0218] According to one embodiment, the electronic device (1801) may receive input for a message that has not been acknowledged by the user of the external electronic device among the messages transmitted to the external electronic device. The electronic device (1801) may receive touch input for a message that has not been acknowledged by the user of the external electronic device. For example, referring to 1510 in FIG. 14, the electronic device (1801) may receive touch input for an unacknowledged message, “See you at 9 o’clock today, then.” transmitted to the external electronic device. However, input for an unacknowledged message transmitted to the external electronic device is not limited to the example described above.

[0219] According to one embodiment, the electronic device (1801) may acquire a message transmitted to an external electronic device as an input message. The electronic device (1801) may acquire an unidentified message as an input message in response to identifying an input to a message transmitted to an external electronic device. The electronic device (1801) may select first context information related to the input message. For example, referring to 1520 in FIG. 14, the electronic device (1801) may acquire a message transmitted to an external electronic device, “See you at 9 o’clock today, then.” as an input message. The electronic device (1801) may select first context information related to the input message. For example, referring to 1520 in FIG. 14, the electronic device (1801) may display an effect indicating that it is performing an operation to select first context information related to the input message.

[0220] According to one embodiment, an electronic device (1801) can determine whether an input message corresponds to the first context information by using an input message and first context information. For example, referring to 1530 in FIG. 14, the electronic device (1801) can determine that the input message does not correspond to the first context information by using an input message, “See you at 9 o’clock today, then.” and first context information, “9 o’clock today: PT class.” Specific operations by which the electronic device (1801) determines whether an input message corresponds to the first context information by using an input message and first context information may correspond to the operations described above in the present disclosure. For example, referring to 1520 in FIG. 14, the electronic device (1801) may indicate that it is determining whether an input message corresponds to the first context information by displaying a gradient effect for the input message.

[0221] According to one embodiment, the electronic device (1801) may not generate a recommendation message if the input message corresponds to the first context information using the input message and the first context information. For example, referring to 1550 in FIG. 14, the electronic device (1801) may check the input message and, if the input message corresponds to the first context information, not generate a recommendation message and not transmit an additional message instead of an unconfirmed message.

[0222] According to one embodiment, the electronic device (1801) can generate a recommendation message using an artificial intelligence model when the input message does not correspond to the first context information using an input message and first context information. For example, referring to 1560 in FIG. 14, if the electronic device (1801) determines that the input message does not correspond to the first context information using an input message “See you at 9 o’clock today, then.” and first context information “9 o’clock today: PT class,” it can generate a recommendation message “Sorry, but I forgot I have an appointment. See you at 9 o’clock tomorrow.” using the input message and first context information. Specific operations of the electronic device (1801) generating a recommendation message using an artificial intelligence model may correspond to the operations described above in the present disclosure.

[0223] According to one embodiment, the electronic device (1801) may cancel the transmission of an unacknowledged message and transmit a recommendation message. For example, referring to 1560 in FIG. 14, the electronic device (1801) may delete the unacknowledged message “Let’s see each other at 9 o’clock today,” and transmit a recommendation message “Sorry, but I forgot I have an appointment. Let’s see each other at 9 o’clock tomorrow” to another electronic device.

[0224] FIG. 15 is a diagram illustrating the operation of an electronic device (1801) generating a recommendation message using a context selector according to one embodiment of the present disclosure.

[0225] According to one embodiment, a context selector (1603) of an electronic device (1801) can select a DB of memory to be used when generating a recommendation message (1607) using the electronic device (1801). The context selector (1603) can select a category for classifying context information. The context selector (1603) can select a first context included in the context DB. For example, referring to FIG. 15, the context selector (1603) can select first context information related to an input message (1606), and the context selector (1603) can input the selected first context information and the input message (1606) into an artificial intelligence model (1601). For example, referring to FIG. 15, the context selector (1603) can select context information stored in a long-term DB (1610) classified according to a long-term storage category as the first context information. For example, referring to FIG. 15, the context selector (1603) can select context information stored in the short-term DB (1620), classified according to the short-term storage category, as the first context information. For example, referring to FIG. 15, the context selector (1603) can select context information classified according to the short-term storage category and context information classified according to the long-term storage category as the first context information.

[0226] According to one embodiment, the context selector (1603) can check the first tag in the context information and select the first context information to which the first tag is tagged. For example, if only context information regarding a specific date is needed from the context information obtained from a messenger application, the context selector (1603) can select the first context information to which a tag regarding the specific date is tagged from the context information stored in the context DB (1610, 1620) in memory. In one embodiment, the first tag is not limited to the example described above and may include a tag regarding a date, a tag regarding a date, a tag regarding user information, and a tag regarding application information.

[0227] According to one embodiment, the context selector (1603) can quickly select only the first context information necessary to generate a recommendation message (1607) from a vast amount of context information by performing the operation of selecting a first context information from context information in which a plurality of categories exist and inputting the first context information into an artificial intelligence model (1601).

[0228] According to one embodiment, the electronic device (1801) may set categories and / or tags when storing information in memory and / or a context DB of memory. The electronic device (1801) may periodically change or reset categories and / or tags for context information stored in the context DB. For example, when the electronic device (1801) constructs a context DB using information stored in memory, it may set categories and / or tags for context information stored in the context DB. Alternatively, the electronic device (1801) may periodically update tags and / or categories for context information stored in a context DB that has been previously set and stored.

[0229] According to one embodiment, context information may be classified according to pre-set categories, and categories may be subdivided and tagged according to tags. For example, context information may be classified according to long-term storage categories and short-term storage categories. Context information stored in the long-term DB (1610) according to the long-term storage category may be stored in the long-term DB (1610) with tags tagged by date. Context information stored in the long-term DB (1610) may be stored in the long-term DB (1610) with tags tagged by the creation date. In another embodiment, tags may be natural language, and categories may be machine language. For example, an electronic device (1801) may classify context information based on categories classified according to machine language and store it in memory, and the electronic device (1801) may display tags tagged in context information to a user based on natural language tagged in context information. However, it is not limited to the examples described above.

[0230] According to one embodiment, the context selector (1603) may be included in the processor (1605). Although the context selector (1603) is depicted as a separate component from the processor (1605) in FIG. 15, it is not limited to the configuration depicted in FIG. 15. The context selector (1603) may be included in the processor (1605) and may be included as a component of the processor.

[0231] FIG. 16 is a diagram showing a system including a generative artificial intelligence model according to one embodiment of the present disclosure.

[0232] Referring to FIG. 16, the User Query / Response Interface (1710) can receive user input. The user input may be in the form of natural language, images, and / or videos. Additionally, context information may be transmitted along with the user input. Context information may include various additional information at the time of user input. For example, information about the application currently being used by the user or the user's location information. Furthermore, user input may be in a mixed form of the aforementioned natural language, images, sounds, and context information. Additionally, user input may be in a non-natural language form, such as selecting a menu. The User Query / Response Interface (1710) can output results from the generative artificial intelligence system to the user. The output may be in the form of natural language or specific content, and may also be provided in the form of actions requested by the user.

[0233] The AI ​​framework (1720) can receive input from the user and coordinate and control each component necessary to perform the user's intent based on the user's query.

[0234] User input received from the User Query / Response Interface (1710) can be sent to the Prompt design component (1721). The Prompt design component (1721) can be used to generate a prompt suitable for inputting user input into a Large Language Model (LLM) or Large Multimodal Models (LMM). The Prompt design component (1721) may be an AI component that uses machine learning algorithms or neural networks to develop better prompts over time. Based on user input, the Prompt design component (1721) can generate a prompt by accessing a knowledge component (e.g., knowledge repositories (1740)) containing user preference data, a prompt library, and prompt examples, and can pass the generated prompt to the LLM or LMM.

[0235] The API / Plug-in management component (1723) can perform the role of communicating with external information when there is a request for additional information when passing user input as input to a generative model. The API / Plug-in management component (1723) establishes a channel to communicate with the outside of the AI ​​Interface via the API, and through the established channel, it can enable access to various data sources (e.g., knowledge repositories (1740)). Additionally, if the application or service needs to perform an action that executes the user input as a final step rather than an intermediate result, the API / Plug-in management component (1723) can request that action from the application / service component (1730) via the API. The information obtained from the outside can be used to generate a prompt in the Prompt design component (1721) along with the user input, or it can be passed as input to the generative model.

[0236] The Refiner component (e.g., output modification component (1725)) allows for detailed tuning of the output from a generative model. For instance, the Refiner component can verify whether the content generated by LLM and / or LMM is irrelevant, contains biased content, or includes harmful content. Additionally, the Refiner component can determine the extent to which the output matches the user's desired outcome and, if necessary, proceed with additional processing. Furthermore, the Refiner component can configure and provide hints to the user to help avoid unwanted outputs.

[0237] A Generative AI Model (1750) generally refers to an artificial intelligence neural network that generates new forms of data based on user input information. A Generative AI Model (1750) may include models that generate images and / or models that generate language. Models that generate images include, but are not limited to, GANs (generative adversarial networks) and VAEs (variational autoencoders), and examples include Diffusion-based generative models that use VAEs and Transformer structures. Models that generate language are models trained to output the most statistically appropriate output value based on input values, and examples include models such as CHAT-GPT 3 and CHAT-GPT 4. There are also LMMs that can recognize various forms of data input, such as text, images, and voice, and generate new data corresponding to them.

[0238] According to one embodiment, the electronic device (1801) of FIGS. 1 to 17 may be configured to include at least some of the User Query / Response Interface (1710) AI framework (1720), application / service component (1730), knowledge repositories (1740), or Generative AI Model (1750) of FIG. 16. According to one embodiment, at least some of the User Query / Response Interface (1710) AI framework (1720), application / service component (1730), knowledge repositories (1740), or Generative AI Model (1750) of FIG. 16 may be included in another electronic device (e.g., an external electronic device and / or server).

[0239] FIG. 17 is a block diagram of an electronic device in a network environment according to various embodiments.

[0240] FIG. 17 is a block diagram of an electronic device (1801) in a network environment (1800) according to various embodiments. Referring to FIG. 17, in the network environment (1800), the electronic device (1801) may communicate with an electronic device (1802) through a first network (1898) (e.g., a short-range wireless communication network) or may communicate with at least one of an electronic device (1804) or a server (1808) through a second network (1899) (e.g., a long-range wireless communication network). According to one embodiment, the electronic device (1801) may communicate with the electronic device (1804) through a server (1808). According to one embodiment, the electronic device (1801) may include a processor (1820), memory (1830), input module (1850), sound output module (1855), display module (1860), audio module (1870), sensor module (1876), interface (1877), connection terminal (1878), haptic module (1879), camera module (1880), power management module (1888), battery (1889), communication module (1890), subscriber identification module (1896), or antenna module (1897). In some embodiments, at least one of these components (e.g., connection terminal (1878)) may be omitted from the electronic device (1801), or one or more other components may be added. In some embodiments, some of these components (e.g., sensor module (1876), camera module (1880), or antenna module (1897)) may be integrated into a single component (e.g., display module (1860)).

[0241] The processor (1820) can, for example, execute software (e.g., program (1840)) to control at least one other component (e.g., hardware or software component) of the electronic device (1801) connected to the processor (1820) and can perform various data processing or operations. According to one embodiment, as at least part of the data processing or operations, the processor (1820) can store commands or data received from other components (e.g., sensor module (1876) or communication module (1890)) in volatile memory (1832), process the commands or data stored in volatile memory (1832), and store the resulting data in non-volatile memory (1834). According to one embodiment, the processor (1820) may include a main processor (1821) (e.g., a central processing unit or an application processor) or an auxiliary processor (1823) that can operate independently or together with it (e.g., a graphics processing unit, a neural processing unit (NPU), an image signal processor, a sensor hub processor, or a communication processor). For example, if the electronic device (1801) includes a main processor (1821) and an auxiliary processor (1823), the auxiliary processor (1823) may be configured to use less power than the main processor (1821) or to be specialized for a specified function. The auxiliary processor (1823) may be implemented separately from the main processor (1821) or as part thereof.

[0242] The auxiliary processor (1823) may control at least some of the functions or states associated with at least one component of the electronic device (1801) (e.g., display module (1860), sensor module (1876), or communication module (1890)) on behalf of the main processor (1821) while the main processor (1821) is in an inactive (e.g., sleep) state, or together with the main processor (1821) while the main processor (1821) is in an active (e.g., application execution) state. According to one embodiment, the auxiliary processor (1823) (e.g., image signal processor or communication processor) may be implemented as part of another functionally related component (e.g., camera module (1880) or communication module (1890)). According to one embodiment, the auxiliary processor (1823) (e.g., neural network processing unit) may include a hardware structure specialized for processing an artificial intelligence model. The artificial intelligence model may be generated through machine learning. Such learning may be performed, for example, on the electronic device (1801) itself where the artificial intelligence model is executed, or through a separate server (e.g., server (1808)). The learning algorithm may 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 may include a plurality of artificial neural network layers.An artificial neural network may be a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a restricted Boltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), a deep Q-network, or a combination of two or more of the above, but is not limited to the examples described above. In addition to the hardware structure, the artificial intelligence model may include a software structure, either additionally or substantially.

[0243] The memory (1830) may store various data used by at least one component of the electronic device (1801) (e.g., a processor (1820) or a sensor module (1876)). The data may include, for example, input data or output data for software (e.g., a program (1840)) and related commands. The memory (1830) may include volatile memory (1832) or non-volatile memory (1834).

[0244] The program (1840) may be stored as software in memory (1830) and may include, for example, an operating system (1842), middleware (1844), or an application (1846).

[0245] The input module (1850) can receive commands or data to be used for a component of the electronic device (1801) (e.g., processor (1820)) from outside the electronic device (1801) (e.g., user). The input module (1850) may include, for example, a microphone, a mouse, a keyboard, a key (e.g., a button), or a digital pen (e.g., a stylus pen).

[0246] The sound output module (1855) can output a sound signal to the outside of the electronic device (1801). The sound output module (1855) may include, for example, a speaker or a receiver. The speaker may be used for general purposes, such as multimedia playback or recording playback. The receiver may be used to receive incoming calls. According to one embodiment, the receiver may be implemented separately from the speaker or as part thereof.

[0247] The display module (1860) can visually provide information to an external (e.g., user) of the electronic device (1801). The display module (1860) may include, for example, a display, a holographic device, or a projector and a control circuit for controlling said device. According to one embodiment, the display module (1860) may include a touch sensor configured to detect a touch, or a pressure sensor configured to measure the intensity of the force generated by said touch.

[0248] The audio module (1870) can convert sound into an electrical signal or, conversely, convert an electrical signal into sound. According to one embodiment, the audio module (1870) can acquire sound through the input module (1850) or output sound through the sound output module (1855) or an external electronic device (e.g., electronic device (1802)) (e.g., speaker or headphones) connected directly or wirelessly to the electronic device (1801).

[0249] The sensor module (1876) can detect the operating state of the electronic device (1801) (e.g., power or temperature) or the external environmental state (e.g., user state) and generate an electrical signal or data value corresponding to the detected state. According to one embodiment, the sensor module (1876) may include, for example, a gesture sensor, a gyroscope sensor, a barometric pressure sensor, a magnetic sensor, an accelerometer sensor, a grip sensor, a proximity sensor, a color sensor, an IR (infrared) sensor, a biosensor, a temperature sensor, a humidity sensor, or an illuminance sensor.

[0250] The interface (1877) may support one or more specified protocols that can be used for the electronic device (1801) to be connected directly or wirelessly to an external electronic device (e.g., electronic device (1802)). According to one embodiment, the interface (1877) may include, for example, a high definition multimedia interface (HDMI), a universal serial bus (USB) interface, an SD card interface, or an audio interface.

[0251] The connection terminal (1878) may include a connector through which the electronic device (1801) can be physically connected to an external electronic device (e.g., electronic device (1802)). According to one embodiment, the connection terminal (1878) may include, for example, an HDMI connector, a USB connector, an SD card connector, or an audio connector (e.g., a headphone connector).

[0252] The haptic module (1879) can convert an electrical signal into a mechanical stimulus (e.g., vibration or movement) or an electrical stimulus that the user can perceive through tactile or kinesthetic senses. According to one embodiment, the haptic module (1879) may include, for example, a motor, a piezoelectric element, or an electric stimulation device.

[0253] The camera module (1880) can capture still images and video. According to one embodiment, the camera module (1880) may include one or more lenses, image sensors, image signal processors, or flashes.

[0254] The power management module (1888) can manage the power supplied to the electronic device (1801). According to one embodiment, the power management module (1888) can be implemented, for example, as at least part of a power management integrated circuit (PMIC).

[0255] The battery (1889) can supply power to at least one component of the electronic device (1801). According to one embodiment, the battery (1889) may include, for example, a non-rechargeable primary battery, a rechargeable secondary battery, or a fuel cell.

[0256] The communication module (1890) can support the establishment of a direct (e.g., wired) communication channel or a wireless communication channel between an electronic device (1801) and an external electronic device (e.g., electronic device (1802), electronic device (1804), or server (1808)), and the performance of communication through the established communication channel. The communication module (1890) may include one or more communication processors that operate independently of the processor (1820) (e.g., application processor) and support direct (e.g., wired) communication or wireless communication. According to one embodiment, the communication module (1890) may include a wireless communication module (1892) (e.g., cellular communication module, short-range wireless communication module, or GNSS (global navigation satellite system) communication module) or a wired communication module (1894) (e.g., LAN (local area network) communication module, or power line communication module). The corresponding communication module among these communication modules can communicate with an external electronic device (1804) via a first network (1898) (e.g., a short-range communication network such as Bluetooth, WiFi (wireless fidelity) direct, or IrDA (infrared data association)) or a second network (1899) (e.g., a legacy cellular network, a 5G network, a next-generation communication network, the Internet, or a computer network (e.g., a LAN or WAN)). These various types of communication modules may be integrated into a single component (e.g., a single chip) or implemented as multiple separate components (e.g., multiple chips). The wireless communication module (1892) can identify or authenticate the electronic device (1801) within a communication network such as the first network (1898) or the second network (1899) using subscriber information (e.g., International Mobile Subscriber Identifier (IMSI)) stored in the subscriber identification module (1896).

[0257] The wireless communication module (1892) can support 5G networks and next-generation communication technologies following 4G networks, for example, new radio access technology. NR access technology can support high-speed transmission of high-capacity data (enhanced mobile broadband (eMBB)), minimization of terminal power and connection of multiple terminals (massive machine type communications (mMTC)), or high reliability and low latency (ultra-reliable and low-latency communications (URLLC)). The wireless communication module (1892) can support a high-frequency band (e.g., mmWave band) to achieve a high data transmission rate, for example. The wireless communication module (1892) can support various technologies for securing performance in the high-frequency band, such as beamforming, massive MIMO (multiple-input and multiple-output), full-dimensional MIMO (FD-MIMO), array antenna, analog beam-forming, or large-scale antenna. The wireless communication module (1892) can support various requirements specified in the electronic device (1801), external electronic device (e.g., electronic device (1804)), or network system (e.g., second network (1899)). According to one embodiment, the wireless communication module (1892) can support a Peak data rate (e.g., 20 Gbps or more) for realizing eMBB, loss coverage (e.g., 164 dB or less) for realizing mMTC, or U-plane latency (e.g., downlink (DL) and uplink (UL) each 0.5 ms or less, or round trip 1 ms or less) for realizing URLLC.

[0258] An antenna module (1897) can transmit a signal or power to or from an external source (e.g., an external electronic device). According to one embodiment, the antenna module (1897) may include an antenna comprising a radiator made of a conductor or a conductive pattern formed on a substrate (e.g., a PCB). According to one embodiment, the antenna module (1897) 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 a first network (1898) or a second network (1899), may be selected from the plurality of antennas, for example, by a communication module (1890). A signal or power may be transmitted or received between the communication module (1890) and an external electronic device through the selected at least one antenna. According to some embodiments, in addition to the radiator, other components (e.g., a radio frequency integrated circuit (RFIC)) may be additionally formed as part of the antenna module (1897).

[0259] According to various embodiments, the antenna module (1897) may form a mmWave antenna module. According to one embodiment, the mmWave antenna module may include a printed circuit board, an RFIC disposed on or adjacent to a first surface (e.g., bottom surface) of the printed circuit board and capable of supporting a specified high frequency band (e.g., mmWave band), and a plurality of antennas (e.g., array antennas) disposed on or adjacent to a second surface (e.g., top surface or side surface) of the printed circuit board and capable of transmitting or receiving a signal of the specified high frequency band.

[0260] At least some of the above components can be connected to each other via a communication method between peripheral devices (e.g., bus, GPIO (general purpose input and output), SPI (serial peripheral interface), or MIPI (mobile industry processor interface)) and exchange signals (e.g., commands or data) with each other.

[0261] According to one embodiment, commands or data may be transmitted or received between an electronic device (1801) and an external electronic device (1804) through a server (1808) connected to a second network (1899). Each of the external electronic devices (1802, or 1804) may be the same or a different type of device as the electronic device (1801). According to one embodiment, all or part of the operations performed on the electronic device (1801) may be performed on one or more of the external electronic devices (1802, 1804, or 1808). For example, if the electronic device (1801) needs to perform a function or service automatically or in response to a request from a user or another device, the electronic device (1801) may request one or more external electronic devices to perform at least part of the function or service instead of performing the function or service itself or additionally. One or more external electronic devices that receive the above request may execute at least part of the requested function or service, or additional function or service related to the request, and transmit the result of the execution to the electronic device (1801). The electronic device (1801) may provide the result as is or additionally processed as at least part of the response to the request. For this purpose, for example, cloud computing, distributed computing, mobile edge computing (MEC), or client-server computing technology may be used. The electronic device (1801) may provide ultra-low latency services using, for example, distributed computing or mobile edge computing. In another embodiment, the external electronic device (1804) may include an Internet of Things (IoT) device. The server (1808) may be an intelligent server using machine learning and / or neural networks.According to one embodiment, an external electronic device (1804) or server (1808) may be included within the second network (1899). The electronic device (1801) may be applied to intelligent services (e.g., smart home, smart city, smart car, or healthcare) based on 5G communication technology and IoT-related technology.

[0262] An electronic device according to various embodiments of the present disclosure verifies an input message before transmitting a message and, if inaccurate information is included or there is an error in the message, generates and provides a recommended message to the user, thereby allowing the user to have the opportunity to correct the inaccurate information before transmitting it and enabling more accurate and reliable message delivery.

[0263] An electronic device according to various embodiments of the present disclosure determines whether an input message corresponds to context information based on preset rules or an artificial intelligence model, and, if necessary, can generate a recommendation message containing accurate information or a user-customized message. This can provide the effect of improving the reliability of the user's communication and reducing information distortion.

[0264] An electronic device according to various embodiments of the present disclosure automatically provides recommended messages, thereby enabling a user to modify an input message and select a recommended message based on reliable information without separate searching or modification. This allows the user to easily transmit an accurate message by selecting a recommended message without having to directly review complex information.

[0265] The electronic device according to various embodiments of the present disclosure can be applied in various environments, such as messenger applications, memo applications, schedule management applications, social media applications, and work collaboration tools. Through this, users can prevent recording or sharing incorrect information due to mistakes and are provided with the opportunity to utilize more accurate information. Furthermore, by reducing the burden on the user during the input process through an automatic recommendation message function and supporting communication and schedule management based on highly reliable information, the effect of maximizing efficiency in both personal and business environments can be achieved.

[0266] The effects obtainable from the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art to which the present disclosure belongs from the description below.

[0267] A method for generating a message using an electronic device according to various embodiments of the present disclosure may include: an operation of executing an application; an operation of obtaining an input message for the application; an operation of selecting a first context information related to the input message from context information obtained by the electronic device prior to the input message and stored in the memory of the electronic device; an operation of determining whether the input message conflicts with the first context information using the input message and the first context information; an operation of generating a recommendation message based on the first context information and the input message when the input message conflicts with the first context information; and an operation of displaying the recommendation message on an execution screen of the application.

[0268] In one embodiment, the operation of generating the recommendation message may include the operation of generating the recommendation message by inputting the first context information and the input message to a first artificial intelligence model trained to generate the recommendation message based on the input message and the first context information.

[0269] In one embodiment, in a method for an electronic device to generate a message, the context information including the first context information is embedded so as to be input into the first artificial intelligence model, and the context information including the first context information is classified according to a pre-set category, and may be a first tag tagged among tags representing the category.

[0270] In one embodiment, in a method for an electronic device to generate a message, the first artificial intelligence model includes a generative artificial intelligence model, and the operation of generating the recommendation message includes: an operation of generating a prompt to be input to the generative artificial intelligence model based on the input message and the first context information; and an operation of inputting the prompt to the generative artificial intelligence model, wherein the prompt may include the input message, a category of the first context information, and a command for generating the recommendation message.

[0271] In one embodiment, in a method for an electronic device to generate a message, the operation of selecting the first context information related to the input message from the context information may include the operation of identifying the first tag related to the input message using the input message, and the operation of selecting the first context information tagged with the first tag from the context information stored in the memory.

[0272] In one embodiment, in a method for an electronic device to generate a message, the category may be classified according to at least one of the type of application associated with the context information, the storage period of the context information, or the format of the context information.

[0273] In one embodiment, in a method for an electronic device to generate a message, the operation of determining whether the input message conflicts with the first context information may include: identifying a similarity between the input message and the first context information; determining that the input message does not conflict with the first context information when the similarity is greater than or equal to a threshold; and determining that the input message conflicts with the first context information when the similarity is less than or equal to the threshold.

[0274] In one embodiment, in a method for an electronic device to generate a message, the operation of determining whether the input message conflicts with the first context information may include the operation of identifying the similarity by inputting the input message and the first context information into a second artificial intelligence model trained to determine whether the input message conflicts with the first context information.

[0275] In one embodiment, in a method for an electronic device to generate a message, at least one of the first artificial intelligence model or the second artificial intelligence model may be an artificial intelligence model included in an external electronic device.

[0276] In one embodiment, a method for an electronic device to generate a message may further include: generating a search term based on the input message and the first context information; obtaining a search result searched using the search term through a server connected to the electronic device; and generating a recommendation message based on the input message, the first context information, and the search result.

[0277] In one embodiment, in a method for an electronic device to generate a message, the application may include at least one of a messenger application, a memo application, or a schedule management application.

[0278] In one embodiment, a method for an electronic device to generate a message may further include: an application, which is a messenger application that transmits and receives messages with an external electronic device; an operation of detecting an event in which the recommended message displayed on the execution screen of the application is selected; and an operation of transmitting the recommended message to the external electronic device in response to the event.

[0279] In one embodiment, a method for an electronic device to generate a message, wherein the operation of acquiring the input message may include: receiving a first input for a transmission icon that transmits a message displayed on the execution screen of the messenger application; and acquiring the input message input to the messenger application in response to the first input for the transmission icon in order to determine whether the input message conflicts with the first context information.

[0280] In one embodiment, a method for an electronic device to generate a message, wherein the operation of acquiring the input message may include: receiving a second input that selects a message transmitted to the external electronic device through the messenger application; and acquiring the message transmitted to the external electronic device as the input message in response to the second input.

[0281] An electronic device according to various embodiments of the present disclosure may include a memory storing context information and instructions; and at least one processor. The instructions may be executed by the at least one processor so that the electronic device: executes an application, obtains an input message for the application, selects a first context information related to the input message from the context information obtained by the electronic device prior to the input message and stored in the memory of the electronic device, determines whether the input message conflicts with the first context information using the input message and the first context information, and if the input message conflicts with the first context information, generates a recommendation message based on the first context information and the input message, and displays the recommendation message on the execution screen of the application.

[0282] In one embodiment, the instructions are executed by the at least one processor, so that the electronic device can generate the recommendation message by inputting the first context information and the input message to a first artificial intelligence model trained to generate the recommendation message based on the input message and the first context information.

[0283] In an electronic device according to one embodiment, the context information including the first context information may be embedded so as to be input into the first artificial intelligence model, and the context information including the first context information may be classified according to a pre-set category, and the first tag may be tagged among tags representing the category.

[0284] In an electronic device according to one embodiment, the first artificial intelligence model may include a generative artificial intelligence model. The instructions are executed by the at least one processor so that the electronic device: generates a prompt to be input to the generative artificial intelligence model based on the input message and the first context information, and inputs the prompt to the generative artificial intelligence model to generate the recommendation message. The prompt may include the input message, a category of the first context information, and a command for generating the recommendation message.

[0285] In one embodiment, the instructions are executed by the at least one processor, so that the electronic device: identifies the similarity between the input message and the first context information; determines that the input message does not conflict with the first context information if the similarity is greater than or equal to a threshold, and determines that the input message conflicts with the first context information if the similarity is less than or equal to the threshold.

[0286] In one embodiment, the instructions are executed by the at least one processor, so that the electronic device can identify the similarity by inputting the input message and the first context information to a second artificial intelligence model trained to determine whether the input message conflicts with the first context information.

[0287] In one embodiment, the instructions are executed by the at least one processor so that the electronic device: generates a search term based on the input message and the first context information, obtains a search result using the search term through a server connected to the electronic device, and generates a recommendation message based on the input message, the first context information, and the search result.

[0288] A method for generating a message by an electronic device according to various embodiments of the present disclosure may include: displaying a message input window for receiving a message to be transmitted from a user to an external electronic device; receiving an input message through the message input window; determining whether the input message conflicts with the first context information using the input message and first context information stored in the memory of the electronic device; if the input message conflicts with the first context information: displaying at least one recommendation message generated based on the first context information and the input message adjacent to the message input window; displaying a message indicating the cause of the input message conflicting with the first context information adjacent to the recommendation message and the message input window; transmitting the selected recommendation message to the external electronic device as one of the displayed recommendation messages is selected; and if the input message does not conflict with the first context information: transmitting the input message to the external electronic device.

[0289] Methods according to the claims or embodiments described in the specification of the present disclosure may be implemented in the form of hardware, software, or a combination of hardware and software.

[0290] When implemented in software, a computer-readable storage medium may be provided for storing one or more programs (software modules). One or more programs stored in the computer-readable storage medium are configured for execution by one or more processors within an electronic device. One or more programs include instructions that cause the electronic device to execute methods according to the claims or embodiments described in the specification of this disclosure.

[0291] In the present disclosure, the function or operation performed by an electronic device may be performed by one or more processors executing one or more instructions stored in memory. The function or operation of the electronic device mentioned in the present disclosure may be performed by a single processor executing one or more instructions, or by a combination of multiple processors executing one or more instructions. A processor mentioned in the present disclosure is understood to include a circuit for performing operations or controlling other components of the electronic device. For example, the one or more processors may include a central processing unit (CPU), a micro-processor unit (MPU), an application processor (AP), a communication processor (CP), a neural processing unit (NPU), a system on chip (SoC), or an integrated circuit (IC) configured to execute one or more instructions. The one or more processors may be configured to perform the operation of the electronic device described above.

[0292] In the present disclosure, a program (software module, software) may be stored in a random access memory, a non-volatile memory including flash memory, a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic disc storage device, a compact disc-ROM (CD-ROM), digital versatile discs (DVDs), or other forms of optical storage devices, or a magnetic cassette. Alternatively, it may be stored in a memory composed of some or all of these. The memory may be composed of a single storage medium or a combination of multiple storage media. The one or more instructions may be stored in a single storage medium or distributed across multiple storage media.

[0293] Additionally, the above program may be stored on an attachable storage device that can be accessed via a communication network such as the Internet, Intranet, LAN (local area network), WLAN (wide LAN), or SAN (storage area network), or a combination thereof. Such a storage device may be connected to a device performing an embodiment of the present disclosure through an external port. Additionally, a separate storage device on a communication network may be connected to a device performing an embodiment of the present disclosure.

[0294] The electronic device according to the various embodiments disclosed in this document may be of various forms. The electronic device may include, for example, a portable communication device (e.g., a smartphone), a computer device, a portable multimedia device, a portable medical device, a camera, a wearable device, or a consumer electronics device. The electronic device according to the embodiments of this document is not limited to the devices described above.

[0295] The various embodiments of this document and the terms used therein are not intended to limit the technical features described in this document to specific embodiments, and should be understood to include various modifications, equivalents, or substitutions of said 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 said items unless the relevant context clearly indicates otherwise. In this document, phrases such as "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" may each include any one of the items listed together in the corresponding phrase, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used simply to distinguish said components from other said components and do not limit said components in any other aspect (e.g., importance or order). Where any (e.g., 1st) component is referred to as “coupled” or “connected” to another (e.g., 2nd) component, with or without the terms “functionally” or “communicationly,” it means that said any component may be connected to said other component directly (e.g., via a wire), wirelessly, or through a third component.

[0296] The term “module” as used in the 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, for example. A module may be a component formed integrally, or a minimum unit of said component or a part thereof 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).

[0297] Various embodiments of the present document may be implemented as software (e.g., program (2040)) comprising one or more instructions stored in a storage medium (e.g., internal memory (2036) or external memory (2038)) readable by a machine (e.g., electronic device (2001)). For example, a processor (e.g., processor (2020)) of the machine (e.g., electronic device (2001)) may call at least one of the one or more instructions stored from the storage medium and execute it. This enables the machine to be operated 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 that can be executed by an interpreter. The storage medium readable by the machine may be provided in the form of a non-transitory storage medium. Here, 'non-temporary' simply means that the storage medium is a tangible device and does not contain a signal (e.g., electromagnetic waves), and the term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily.

[0298] According to one embodiment, the method according to the various embodiments disclosed herein may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)) or an application store (e.g., Play Store). TM It can be distributed online (e.g., downloaded or uploaded) through ) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created on a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0299] According to various embodiments, each component (e.g., module or program) of the components described above may include a singular or multiple entities, and some of the multiple entities may be separated and placed in other components. According to various embodiments, one or more of the components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Generally or additionally, multiple components (e.g., module or program) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the multiple components in the same or similar manner as those performed by the corresponding component among the multiple components prior to integration. According to various embodiments, operations performed by the module, program, or other components 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.

[0300] The specific embodiments described in this disclosure are merely examples and do not limit the scope of this disclosure in any way. For the sake of brevity, descriptions of prior electronic configurations, control systems, software, and other functional aspects of said systems may be omitted.

[0301] Meanwhile, although specific embodiments have been described in the detailed description of the present disclosure, it is understood that various modifications are possible within the scope of the present disclosure. Therefore, the scope of the present disclosure should not be limited to the described embodiments, but should be defined by the claims set forth below as well as equivalents thereof.

Claims

1. In a method for an electronic device to generate a message, The action of running an application; An operation to obtain an input message for the above application; An operation of selecting first context information related to the input message from context information obtained by the electronic device and stored in the memory of the electronic device prior to the input message being obtained; An operation to determine whether the input message conflicts with the first context information using the input message and the first context information; An operation to generate a recommendation message based on the first context information and the input message when the input message conflicts with the first context information; A method comprising the operation of displaying the recommendation message on the execution screen of the above application.

2. In Paragraph 1, The operation of generating the above recommendation message is, A method comprising the operation of generating the recommendation message by inputting the first context information and the input message into a first artificial intelligence model trained to generate the recommendation message based on the input message and the first context information.

3. In Paragraph 2, The context information including the first context information is embedded so that it can be input into the first artificial intelligence model, and A method in which the context information including the first context information is classified according to a pre-set category, and the first tag is tagged among the tags representing the category.

4. In Paragraph 3, The above-mentioned first artificial intelligence model includes a generative artificial intelligence model, and The operation of generating the above recommendation message is, An operation to generate a prompt to be input to the generative artificial intelligence model based on the input message and the first context information; and Includes the operation of inputting the above prompt into the above generative artificial intelligence model, A method comprising the above prompt, the input message, the category of the first context information, and a command for generating the recommendation message.

5. In Paragraph 3, The operation of selecting the first context information related to the input message from the above context information is, An operation of identifying the first tag related to the input message using the input message, and A method comprising the operation of selecting the first context information tagged with the first tag from the context information stored in the memory.

6. In Paragraph 3, A method in which the above categories are classified according to at least one of the type of application related to the context information, the storage period of the context information, or the format of the context information.

7. In Paragraph 2, The operation of determining whether the above input message conflicts with the above first context information is, An operation to identify the similarity between the above input message and the above first context information, An operation of determining that the input message does not conflict with the first context information when the similarity is greater than or equal to a threshold, and A method comprising determining that the input message conflicts with the first context information when the similarity is less than the threshold.

8. In Paragraph 7, The operation of determining whether the above input message conflicts with the above first context information is, A method comprising the operation of identifying the similarity by inputting the input message and the first context information into a second artificial intelligence model trained to determine whether the input message conflicts with the first context information.

9. In Paragraph 8, A method in which at least one of the first artificial intelligence model or the second artificial intelligence model is an artificial intelligence model included in an external electronic device.

10. In Paragraph 2, An operation to generate a search term based on the above input message and the above first context information; The operation of obtaining search results using the search term through a server connected to the electronic device; and A method further comprising the operation of generating the recommendation message based on the input message, the first context information, and the search result.

11. In Paragraph 2, The above application is a messenger application that sends and receives messages with an external electronic device, and An action of displaying a message indicating the cause of the input message conflicting with the first context information, along with the recommendation message, on the execution screen of the above messenger application; An operation to detect an event in which the recommended message displayed on the execution screen of the above messenger application is selected; The operation of transmitting the recommendation message to the external electronic device in response to the above event; and A method further comprising the operation of transmitting the input message to the external electronic device when the input message does not conflict with the first context information.

12. In Paragraph 11, The operation of obtaining the above input message is, The operation of receiving a first input for a transmission icon that transmits a message displayed on the execution screen of the messenger application; and A method comprising the operation of acquiring the input message entered into the messenger application in response to the first input for the transmission icon, in order to determine whether the input message conflicts with the first context information.

13. In Paragraph 11, The operation of obtaining the above input message is, The operation of receiving a second input selecting a message transmitted to the external electronic device through the messenger application; and A method comprising the operation of acquiring a message transmitted to the external electronic device as the input message in response to the second input.

14. In electronic devices, Memory for storing context information and instructions; and It includes at least one processor, The above instructions are executed by the above at least one processor, and the electronic device: Run the application, Obtain an input message for the above application, and From the context information obtained by the electronic device and stored in the memory of the electronic device prior to the acquisition of the input message, a first context information related to the input message is selected, Using the above input message and the above first context information, determine whether the above input message conflicts with the above first context information, and If the above input message conflicts with the above first context information, a recommendation message is generated based on the above first context information and the above input message, and An electronic device that displays the recommendation message on the execution screen of the above application.

15. In a method for an electronic device to generate a message, An operation of displaying a message input window to receive a message to be transmitted from a user to an external electronic device; The operation of receiving an input message through the above message input window; An operation to determine whether the input message conflicts with the first context information using the input message and the first context information stored in the memory of the electronic device; If the above input message conflicts with the above first context information: The operation of displaying at least one recommendation message generated based on the first context information and the input message adjacent to the message input window; An operation of displaying a message indicating the cause of the above input message conflicting with the above first context information adjacent to the recommendation message and the message input window; The operation of transmitting the selected recommendation message to the external electronic device as one of the above-displayed recommendation messages is selected; If the above input message does not conflict with the above first context information: A method comprising the operation of transmitting the above input message to the above external electronic device.

Images