Online advertisement content platform providing system

Abstract

According to various embodiments, an online advertising content platform providing system includes a server and an electronic device, in which the server is configured to acquire at least one content of a specific product from the electronic device, in which the at least one content includes an image of the specific product, and obtaining product attribute information of the specific product based on the at least one content, the product attribute information corresponding to a plurality of product features, and providing celebrity image recommendation information determined based on a plurality of weight values associated with the plurality of product features to the electronic device based on the product attribute information, wherein the celebrity image recommendation information includes at least one celebrity image, and the electronic device is configured to display a first screen including the at least one celebrity image and a predetermined number of celebrity features among a plurality of celebrity features based on the provided celebrity image recommendation information, wherein the predetermined number of celebrity features includes a first group of celebrity features in the plurality of celebrity features based on the specific product being a first product, and the predetermined number of celebrity features includes a second group of celebrity features in the plurality of celebrity features based on the specific product being a second product, and the second set of celebrity features may be at least partially different from the first set of celebrity features.

Description

Technical Field

[0001] The present disclosure relates to an online advertising content platform providing system, an electronic device included in the system, and an operating method of the electronic device. Background Art

[0002] The online display advertising market is growing. In online advertising, celebrities are often used to attract consumers' attention. Therefore, the advertising invitation market for celebrities is growing.

[0003] In this case, advertisers face the cost of using celebrities for advertising, while celebrities face the problem of quickly entering the advertising market while their fame lasts. Summary of the invention Technical problem to be solved by the invention

[0004] In order to quickly implement online advertising, it is necessary to build an intermediary platform between the advertiser who sells the product, the creator who makes the advertisement, and the celebrity who appears in the advertisement. When an online advertising content platform intends to advertise a specific product online, the advertising effect can only be improved when a celebrity who is compatible with the attributes of the product to be advertised appears in the advertisement.

[0005] According to various embodiments, by providing a system for providing an online advertising content platform, an electronic device included in the system, and an operating method of the electronic device, a suitable celebrity may be recommended according to a specific product. Technical solutions to technical problems

[0006] An online advertising content platform providing system includes a server and an electronic device, wherein the server is configured to obtain at least one content of a specific product from the electronic device, wherein the at least one content includes an image of the specific product, obtain product attribute information of the specific product based on the at least one content, wherein the product attribute information corresponds to a plurality of product features, and provide celebrity image recommendation information determined based on a plurality of weight values ​​associated with the plurality of product features to the electronic device, wherein the celebrity image recommendation information includes at least one celebrity image, wherein the electronic device is configured to display a first screen including the at least one celebrity image and a predetermined number of celebrity features from a plurality of celebrity features based on the provided celebrity image recommendation information, wherein based on the specific product being a first product, the predetermined number of celebrity features includes a first group of celebrity features from the plurality of celebrity features, wherein based on the specific product being a second product, the predetermined number of celebrity features includes a second group of celebrity features from the plurality of celebrity features, wherein the second group of celebrity features is at least partially different from the first group of celebrity features.

[0007] According to various embodiments, an electronic device includes: a display, a memory and at least one processor, wherein the at least one processor can obtain at least one content of a specific product, wherein the at least one content includes an image of the specific product, obtain product attribute information of the specific product based on the at least one content, wherein the product attribute information corresponds to multiple product features, obtain multiple celebrity features of each of the multiple celebrity images based on multiple celebrity images, wherein the multiple celebrity features of each of the multiple celebrity images include a first feature of a first image of a specific celebrity and a second feature of a second image of the specific celebrity, wherein the first feature and the second feature are at least partially different, based on the product attribute information, obtain at least one celebrity image of the multiple celebrity images and a predetermined celebrity feature of a plurality of features corresponding to the at least one celebrity image, obtain celebrity image recommendation information including the at least one celebrity image and the predetermined celebrity feature; and based on the celebrity image recommendation information, display a first screen including at least one celebrity image and the predetermined celebrity feature on the display.

[0008] According to various embodiments, an operating method of an electronic device includes: obtaining at least one content of a specific product, wherein the at least one content includes an image of the specific product; obtaining product attribute information of the specific product based on the at least one content, wherein the product attribute information corresponds to multiple product features; obtaining multiple celebrity features of each celebrity image in the multiple celebrity images based on multiple celebrity images, wherein the multiple celebrity features of each celebrity image include a first feature of a first image of a specific celebrity and a second feature of a second image of the specific celebrity, and the first feature and the second feature are at least partially different; based on the product attribute information, obtaining at least one celebrity image in the multiple celebrity images and a predetermined celebrity feature among a plurality of features corresponding to the at least one celebrity image; obtaining celebrity image recommendation information including the at least one celebrity image and the predetermined celebrity feature; and based on the celebrity image recommendation information, displaying a first screen including the at least one celebrity image and the predetermined celebrity feature. Beneficial Effects

[0009] According to various embodiments, an operating method of an online advertising content platform providing system and an operating method of an online advertising content platform providing system may provide an intermediary platform between advertisers selling products, creators making advertisements, and celebrities appearing in the advertisements.

[0010] Therefore, celebrities can provide the effect of increasing advertising revenue at a fixed cost, as well as the effect of expanding the scope of advertising exposure and receiving advertisements for various products at the same time. Quick settlement based on simple execution can minimize the risk of advertising contracts, and celebrities can have the final say when requesting the use of portrait rights, thereby preventing advertising exposure and abuse in advance.

[0011] Advertisers can quickly execute advertisements by accessing a built-in library of celebrity photos and videos, and can maximize the advertising effect by executing the advertisements during the celebrity's popularity. A system that automatically recommends suitable celebrities for products to be advertised can provide an advantage of using the best celebrities for advertising at the best cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a block diagram of devices included in the system according to various embodiments.

[0013] Figure 2 is a block diagram of an electronic device according to various embodiments.

[0014] Figure 3 is a diagram illustrating an example of an artificial intelligence model according to various embodiments.

[0015] Figure 4 is a diagram illustrating learning data according to various embodiments.

[0016] Figure 5 is a diagram illustrating a learned artificial intelligence model according to various embodiments.

[0017] Figure 6 is a flowchart illustrating an operation method of an online advertising content platform providing system according to various embodiments.

[0018] Figure 7 is a flowchart illustrating an operation method of an online advertising content platform providing system according to various embodiments.

[0019] Figure 8 is a diagram illustrating an example of a graphic object included in a celebrity recommendation screen according to various embodiments.

[0020] Fig. 9 An example of a screen for receiving information of a specific product according to various embodiments is shown.

[0021] Fig.10 An example of a screen for receiving product category information according to various embodiments is shown.

[0022] Fig.11 An example of a celebrity recommendation screen is shown in accordance with various embodiments.

[0023] Fig.12a An example of a profile of a particular celebrity is shown in accordance with various embodiments.

[0024] Figure 12b An example of a profile of a particular celebrity is shown in accordance with various embodiments.

[0025] Fig.13 Examples of interface screens for mediating negotiations between advertisers and celebrities are shown in accordance with various embodiments.

[0026] Fig.14 An example of a creator recommendation screen is shown in accordance with various embodiments.

[0027] Fig.15 Examples of profiles of specific authors are shown in accordance with various embodiments.

[0028] Fig.16 Examples of interface screens for mediating negotiations between advertisers and creators are shown in accordance with various embodiments.

[0029] Fig.17 An example of a results screen of an advertising mediation platform according to various embodiments is shown.

[0030] Fig.18 Examples of advertisement execution results according to various embodiments are shown. DETAILED DESCRIPTION

[0031] Figure 1 is a block diagram of devices included in system 100 according to various embodiments.

[0032] According to various embodiments, the system 100 may be an online advertising content platform providing system. For example, the system 100 may provide a platform for mediating between advertisers selling products, creators making advertisements, and celebrities appearing in advertisements. However, those skilled in the art may appreciate that the service areas that the system 100 may provide are not limited.

[0033] Reference Figure 1 , the system 100 may include an electronic device 101, a first external electronic device 110, a second external electronic device 120, and / or a server 130. For example, the electronic device 101 may be an electronic device of an advertiser. For example, the first external electronic device 110 may be an electronic device of a celebrity. For example, the second external electronic device 120 may be an electronic device of a creator. For example, the server 130 may be a server that mediates the electronic device 101, the first external electronic device 110, and the second external electronic device 120. Although Figure 1, each device is shown as one, but this is only an example, and there is no limit to the number of each of the electronic device 101, the first external electronic device 110, and the second external electronic device 120 mediated by the server 130. For example, the number of electronic devices 101 may be as many as the number of advertisers, the number of first external electronic devices 110 may be as many as the number of celebrities, and the number of second external electronic devices 120 may be as many as the number of creators.

[0034] According to various embodiments, Figure 1 The components shown in (e.g., the electronic device 101, the first electronic device 110, and the second electronic device 120 and / or the server 130) can be connected via wired communication and / or wireless communication, and can exchange various information and / or data described in the present disclosure with each other through communication.

[0035] Figure 2 is a block diagram of an electronic device 101 according to various embodiments.

[0036] According to various embodiments, other components included in the system 100 (eg, the first electronic device 110, the second electronic device 120, and / or the server 130) include Figure 2 The same components as those shown in can be implemented to perform at least a portion of the operations of the electronic device 101 described in the present disclosure.

[0037] Reference Figure 2 , the electronic device 101 may include a data acquisition device 210 such as a camera 211, a microphone 213, and a communication circuit 215, a memory 230 storing a plurality of applications 231 and a plurality of artificial intelligence models 233, a processor 240 including an optimization module 243, a configuration value database (DB) 245, and a compiler 241, and a display 250. The type of the display 250 is not limited. Below, each component included in the electronic device 101 is described.

[0038] First, hereinafter, the data acquisition device 210 in various embodiments will be described.

[0039] According to various embodiments, the data collection device 210 may be interpreted as a logical concept for classifying devices for collecting content among the devices included in the electronic device 101. In addition to the camera 211, the microphone 213, and the communication circuit 215, the data collection device 210 may further include various types of content.

[0040] According to various embodiments, the data acquisition device 210 may acquire various types of data (or content) for processing based on the artificial intelligence model 233 described later. In an embodiment, the data acquisition device 210 may be driven to acquire various types of data according to the execution and / or driving of the process, program and / or application installed (or stored) in the electronic device 101. For example, when executing and / or driving a camera application, the electronic device 101 may drive the camera 211 (e.g., to perform an operation to control the image sensor to read out) to acquire images and / or videos as data. For example, when executing and / or driving a recording application, the electronic device 101 may drive the microphone 213 to acquire audio data such as the user's voice and / or ambient sound as data. For example, when executing and / or driving a network-based application, the electronic device 101 may use the communication circuit 215 to establish a communication connection with a media server and acquire media data such as images, videos and audio data. Examples of each data acquisition device 210 will be introduced below.

[0041] According to various embodiments, the camera 211 (e.g., at least one front camera and at least one rear camera) can capture still images (or images) and moving images. According to an embodiment, at least one camera 211 may include one or more lenses, an image sensor, an image signal processor, or a flash. In one embodiment, the electronic device 101 may include cameras 211 with different properties or functions (or uses). For example, at least one camera 211 may include cameras with different viewing angles. For example, the viewing angle may include an ultra-wide angle of 114° to 94°, a wide angle of 84° to 63°, a telephoto of 28° to 8°, and an ultra-telephoto of 6° to 3°. For example, as described above, at least one camera 211 may include at least one front camera disposed on a front surface for capturing images and / or videos, and at least one rear camera disposed on a rear surface for capturing images and / or videos.

[0042] According to various embodiments, the microphone 213 may receive sounds from outside the electronic device 101. For example, the electronic device 101 (such as the processor 240) may drive the microphone 213 to receive sounds generated from the outside through the microphone 213. The sounds generated from the outside may include the voice (or speech) of the speaker (e.g., the user and / or another speaker (or other companion)), life noise, and environmental (or background) noise. In one embodiment, the microphone 213 may include a plurality of microphones. The electronic device 101 (e.g., the processor 240) may form beamforming from the sounds received using the plurality of microphones 213 to receive sounds generated in a specified direction from the electronic device 101. The sounds in the specified direction acquired based on the received sounds may be defined as sub-sounds. Each of the plurality of microphones 213 may be arranged in the electronic device 101 so as to be spaced apart by a predetermined distance and a time or phase associated with acquiring the sounds. Since the beamforming technology is a well-known technology, a detailed description will be omitted.

[0043] According to various embodiments, the communication circuit 215 can form a communication connection with an external electronic device (e.g., another electronic device 110 or 120 or a server 130) through various communication methods, and send and / or receive data. As described above, the communication method may include a communication method for setting a direct communication connection (e.g., Bluetooth and Wi-Fi Direct), a communication method using an access point (AP) (e.g., Wi-Fi communication), or a communication method for cellular communication using a base station (e.g., 3G, 4G / LTE, 5G).

[0044] A plurality of artificial intelligence models 233 according to various embodiments will be described below.

[0045] Figure 3 is a diagram illustrating an example of an artificial intelligence model 233 according to various embodiments.

[0046] According to various embodiments, the artificial intelligence model 233 may be stored in the electronic device 101, the first external electronic device 110, and / or the second external electronic device 120. According to various embodiments, the server 130 may generate (or update) the artificial intelligence model 233, and perform various operations described in the present disclosure using the generated artificial intelligence model 233. According to an embodiment, the server 130 may provide the generated artificial intelligence model 233 to the electronic device 101, and perform various operations described in the present disclosure using the artificial intelligence model 233 provided (or downloaded) by the electronic device 101. According to an embodiment, the electronic device 101 may generate (or update) the artificial intelligence model 233, and perform various operations described in the present disclosure using the generated artificial intelligence model 233.

[0047] According to various embodiments, the artificial intelligence model 233 is a model pre-learned based on a specified type of learning algorithm, and may be a pre-implemented artificial intelligence model to receive and calculate various data (or content) and output (or obtain) result data. For example, learning may be performed based on a machine learning algorithm or a deep learning algorithm in the server 130 (or electronic device 101) to output a specific type of result data as output data by using a specific type of data as input data, and multiple artificial intelligence models 233 (e.g., machine learning models and deep learning models) may be generated. The machine learning algorithm may include supervised algorithms such as linear regression, logistic regression, unsupervised algorithms such as clustering, visualization and dimensionality reduction, association rule learning, and reinforcement algorithms, and the deep learning algorithm may include an artificial neural network (ANN), a deep neural network (DNN), and a convolutional neural network (CNN), and may further include various learning algorithms, but are not limited to the content. The learned artificial intelligence model 233 may include at least one matrix (e.g., a convolutional layer and a pooling layer) for calculating input data, and may output result data by calculating input data based on the matrix. As an example, Figure 3 As shown, the artificial intelligence model 233 (e.g., CNN model) can be generated based on CNN by using image data and information related to the object included in the image data as training data (e.g., setting the image data as input data and setting the information of the object (or subject) as output data), and the generated artificial intelligence model 233 can be implemented to output output content 303 (e.g., result data) including information about the object (e.g., instance segmentation image data) by calculating the input image data using a convolutional layer, an activation function (e.g., sigmoid, Relu), and a pooling layer in response to receiving content 301. At the same time, although Figure 3 It is not shown, but the deep learning model based on CNN learning may include more layers in addition to the convolutional layers and pooling layers shown, and its detailed description will be omitted because it is well known. As another example, the artificial intelligence model can generate an artificial intelligence model based on other learning algorithms by using audio data and information about the speaker as learning data (for example, setting the audio data as input data and setting the information about the speaker as output data), and other artificial intelligence models can be implemented to output an identifier (for example, a unique ID) for identifying the speaker as result data in response to receiving audio data (for example, a sentence).

[0048] According to various embodiments, the artificial intelligence model 233 may be implemented as a weight and activation function including matrices corresponding to a 32-bit specified computing capability (hereinafter referred to as a calculation value) (e.g., a weight calculation value (e.g., a weight precision)) and a 32-bit activation function calculation value (e.g., an activation function precision). For example, the calculation value may be a value used to calculate parameters such as weights or activation functions (e.g., sigmoid, Relu), and may include a weight calculation value (e.g., a weight precision) and an activation function calculation value (e.g., an activation precision).

[0049] Meanwhile, when the server 130 generates (or stores) the artificial intelligence model 233, the server 130 may receive various contents from the electronic device 101, the first external electronic device 110, and / or the second external electronic device 120, and provide the results processed using the artificial intelligence model 233 to the electronic device 101. When the electronic device 101 generates (or stores) the artificial intelligence model 233, the electronic device 101 may process the acquired various contents using the artificial intelligence model 233 and output the results.

[0050] Figure 4 is a diagram illustrating learning data according to various embodiments.

[0051] refer to Figure 4 , the artificial intelligence model 233 may be trained by the first training data 401 and the second training data 402. The first training data 401 may be input data, and the second training data 402 may be data output by the artificial intelligence model 233. For example, the artificial intelligence model 233 trained by the first training data 401 and the second training data 402 may receive the same type of input data as the first training data (learning data) 401, and output the same type of output data as the second training data (learning data) 402.

[0052] According to various embodiments, the artificial intelligence model 233 may be implemented as multiple. For example, when the first training data 401 includes information of a product described later and the second training data 402 includes product features described later and / or product feature values ​​corresponding to the product features, the first artificial intelligence model may be generated based on the learning data related to the product. When the first training data 401 includes information of a celebrity described later and the second training data 402 includes celebrity features described later and / or celebrity feature values ​​corresponding to the celebrity features, the second artificial intelligence model may be generated based on the learning data related to the celebrity.

[0053] According to an embodiment, a plurality of artificial intelligence models may be generated based on the type of celebrity (such as race (white, yellow, black), gender (male, female), etc.). For example, a third artificial intelligence model may be generated based on information related to a celebrity of a first race (or first gender) (such as celebrity information, celebrity features, and celebrity feature values), and a fourth artificial intelligence model may be generated based on information related to a celebrity of a second race (or second gender) (such as celebrity information, celebrity features, and celebrity feature values). In this case, based on the celebrity type (such as race, gender, etc.) selected by the electronic device 101 (i.e., the advertiser), at least one celebrity belonging to the selected celebrity type may be provided as a recommendation result through an artificial intelligence model corresponding to the selected celebrity type.

[0054] In the following, for the convenience of explanation, the product-related artificial intelligence model and the celebrity-related artificial intelligence model are described using the same figure mark 233, but as mentioned above, those skilled in the art can understand that the product-related artificial intelligence model and the celebrity-related artificial intelligence model can be implemented and used separately.

[0055] According to various embodiments, the artificial intelligence model 233 (e.g., a first artificial intelligence model) may use product information as first training data 401, and use product features and / or P feature values ​​corresponding to product features (e.g., product feature vectors (a matrix of size P×1)) as second learning data 402.

[0056] For example, the product information may include an image, name, category, description of the product, manufacturer information, or URL of the manufacturer, but there is no limitation on the product information used as the first learning data 401. The product information may be provided by an advertiser (such as the advertiser's electronic device 101) or collected by an external server (such as an external cloud server including various information of the product).

[0057] For example, the second learning data 402 may include multiple product features and multiple feature values ​​corresponding to the multiple product features. "Product features" may be specific features related to a product. For example, the multiple product features may include "innovation", "originality", "economy", "fashion", "unique", etc. According to various embodiments, multiple product features may be preset or changed, and the number of multiple product features is not limited. According to one embodiment, multiple product features may be determined based on the first learning data (e.g., information about a product). For example, when there is a post that publishes information about a specific product in a social networking service (SNS), additional information (e.g., topic tags) with a high frequency in multiple products may be determined as a candidate for a product feature based on additional information (e.g., topic tags) related to the information about the specific product (e.g., frequently mentioned). According to various embodiments, the multiple feature values ​​corresponding to the multiple product features may be numerical values ​​corresponding to each of the multiple product features. For example, the second learning data 402 may include multiple feature values ​​corresponding to each of the multiple product features (e.g., "innovation", "originality", "economy", "fashion", "unique", etc.), as shown in mathematical formula (1). The form of Mathematical Formula 1 (Fp) (such as vector and / or matrix form) is just an example, and the form of the second learning data 402 is not limited.

[0058] Mathematical formula 1

[0059] In mathematical formula 1, p1 may represent a first eigenvalue (numerical value) corresponding to a first product feature (such as “innovation”), p2 may represent a second eigenvalue (numerical value) corresponding to a second product feature (such as “originality”), p3 may represent a third eigenvalue (numerical value) corresponding to a third product feature (such as “economic”), p4 may represent a fourth eigenvalue (numerical value) corresponding to a fourth product feature (such as “fashionable”), p5 may represent a fifth eigenvalue (numerical value) corresponding to a fifth product feature (such as “unique”), and product attribute information (such as a product feature vector (Fp)) may include p eigenvalues ​​corresponding to a total of p product features.

[0060] In the present disclosure, for convenience of explanation, the feature value is described as a value in the range of 0 to 10, but the feature value can be defined in various ranges outside the above range (0 to 10). For example, the feature value can be set to 0 or 1. In the present disclosure, a set of product feature values ​​(e.g., vector and / or matrix form data (Fp)) is described as "features of a product" (or product attribute information).

[0061] According to various embodiments, the artificial intelligence model 233 (e.g., the second artificial intelligence model) may use information of a celebrity as first training data 401 and receive celebrity features and / or feature values ​​corresponding to the celebrity features as second training data 402 for learning.

[0062] For example, the celebrity information used as the first learning data 401 may include an image (and / or video), name, age, height, weight, or description of the celebrity (famous person). The celebrity information used as the first learning data 401 is not limited. The celebrity information may be provided by the celebrity (e.g., the first external electronic device 110 of the celebrity) or collected from an external server (e.g., an external cloud server including various information of the celebrity).

[0063] For example, the second learning data 402 may include multiple celebrity features and / or multiple C feature values ​​corresponding to the multiple celebrity features (e.g., celebrity feature vectors (matrices of size C×1). "Celebrity features" may be specific features related to a certain person. For example, the multiple celebrity features may include "pretty", "cute", "cool", "gorgeous", "luxury", "healthy", etc. The multiple celebrity features may be preset or changed, and the number of the multiple celebrity features is not limited. According to an embodiment, the celebrity features may be determined based on the first learning data (e.g., celebrity information). For example, when there is a post including specific celebrity information in an SNS, additional information (e.g., topic tags) related to the specific celebrity (e.g., frequently mentioned) may be determined as a candidate for the celebrity feature. According to various embodiments, the multiple celebrity features may correspond to each of the multiple celebrity features. For example, the second learning data 402 may include multiple feature values ​​corresponding to the multiple celebrity features (e.g., "pretty", "cute", "cool", "gorgeous", "luxury", "healthy"), respectively, as shown in Mathematical Formula 2. The form of Mathematical Formula 2 (Fc) (e.g., vector and / or matrix form) is only an example, and there is no limitation on the form of the second learning data 402.

[0064] Mathematical formula 2

[0065] In mathematical formula 2, c1 represents the first eigenvalue (numerical value) corresponding to the first celebrity characteristic (such as "beautiful"), c2 represents the second eigenvalue (numerical value) corresponding to the second celebrity characteristic (such as "cute"), c3 represents the third eigenvalue (numerical value) corresponding to the third celebrity characteristic (such as "cool"), c4 represents the fourth eigenvalue (numerical value) corresponding to the fourth celebrity characteristic (such as "gorgeous"), c5 represents the fifth eigenvalue (numerical value) corresponding to the fifth celebrity characteristic (such as "luxurious"), c6 represents the sixth eigenvalue (numerical value) corresponding to the sixth celebrity characteristic (such as "healthy"), and celebrity attribute information (such as celebrity feature vector Fc) may include c eigenvalues ​​corresponding to a total of c celebrity characteristics.

[0066] Although the eigenvalues ​​are described as values ​​in the range of 0 to 10 for ease of description, the eigenvalues ​​may be defined in various ranges outside the above range (0 to 10). For example, the eigenvalues ​​may be set to 0 or 1. In the present disclosure, a set of celebrity eigenvalues ​​(e.g., vector and / or matrix data Fc) will be described as “attributes of a celebrity” (or celebrity attribute information).

[0068] According to various embodiments, the celebrity attribute information may be determined to correspond to an image (and / or video) of a celebrity used as the first learning data. For example, even if they are the same celebrity, the celebrity attribute information of the first image of the celebrity and the celebrity attribute information of the second image of the celebrity may be different.

[0069] According to one embodiment, at least some of the above-mentioned product features and celebrity features can be repeated. For example, product features such as "health" can be used as celebrity features to represent the attributes of the celebrity, and the product features and celebrity features are not necessarily set to be different, as shown in the examples described herein.

[0070] Figure 5 is a diagram illustrating a trained artificial intelligence model according to various embodiments.

[0071] Reference Figure 5 , the artificial intelligence model 233 can receive the first data 501 and output the second data 502. For example, the first data 501 can be related to Figure 4 The first training data 401 in the same type of data, the second data 502 can be Figure 4 The second learning data 402 in the same type of data.

[0072] According to various embodiments, the artificial intelligence model 233 may receive product information as first data 501, and output product features and / or feature values ​​corresponding to the product features as second data 502. Figure 4description to understand the celebrity information, celebrity features and feature values ​​corresponding to the celebrity features.

[0073] According to various embodiments, the artificial intelligence model 233 may receive celebrity information as first data 501, and receive celebrity features and feature values ​​corresponding to the celebrity features as second data 502. Figure 4 By describing the information, celebrity features, and feature values ​​corresponding to celebrity characteristics, one can understand.

[0074] Figure 6 is a flowchart illustrating an operation method of an online advertising content platform providing system according to various embodiments.

[0075] Figure 6 Operation and Figure 7 The operation can be carried out organically, and Figure 6 Operation and Figure 7 At least some of the operations can be omitted.

[0076] Hereinafter, operations 601 to 605 are described as being performed by the server 130 , but when the electronic device 101 includes the artificial intelligence model 233 , at least some of the operations 601 to 605 may be performed by the electronic device 101 .

[0077] According to various embodiments, in operation 601, the server 130 may acquire at least one content of a specific product. For example, the server 130 may acquire (e.g., receive) at least one content of a specific product from the electronic device 101. For example, the at least one content may include information (e.g., image, name, category, description) of the specific product. According to various embodiments, the server 130 may receive and acquire at least one content of a specific product input through the electronic device 101. At least some of the at least one content of a specific product may be collected by an external server (e.g., an external cloud server including various products).

[0078] According to various embodiments, in operation 603, the server 130 may acquire product attribute information. For example, the server 130 may acquire product attribute information of a specific product (e.g., a product feature value of a specific product in the form of mathematical formula 1) using the artificial intelligence model 233 based on the at least one content acquired. As another example, when the electronic device 101 acquires product attribute information of a specific product using the artificial intelligence model 233, the acquired product attribute information may be provided to the server 130. According to an embodiment, even for the same celebrity, the celebrity attribute information may be different depending on the image (photo and / or dynamic image) of the celebrity. For example, for a specific celebrity, a first image (e.g., a smiling image) may have a high feature value of a first celebrity feature (e.g., "beautiful"), while a second image (e.g., an exercise image) may have a high feature value of a sixth celebrity feature (e.g., "healthy").

[0079] According to various embodiments, in operation 605 , the server 130 may provide celebrity image recommendation information determined based on a weight value associated with product attribute information.

[0080] Next, an example of a method in which the server 130 recommends at least one celebrity image will be described.

[0081] According to various embodiments, the server 130 may determine a weight value associated with the product attribute information. For example, the weight value may be represented in the form of a matrix of size 1xC (hereinafter referred to as a weight matrix (W)).

[0082] Mathematical formula 3

[0083] According to various embodiments, the weight values ​​w1, w2, ..., w c The weight values ​​w1, w2, ..., w c They can correspond to the celebrity feature values ​​in the same row. For example, the nth weight value w in the nth row n It can correspond to the nth celebrity feature value c in the nth row n .

[0084] According to various embodiments, the weight matrix (W) may be calculated based on the product attribute information and the calculation results of the reference matrix (M).

[0085] Mathematical formula 4 W=(F p T ·M) T

[0086] Mathematical formula 5

[0087] Referring to Mathematical Formula (4), a weight matrix (W) associated with product attribute information of a specific product may be calculated as a matrix calculation result of a product feature vector (Fp) and a reference matrix (M). Referring to Mathematical Formula 5, a reference matrix (M) may be defined as a matrix of size pxc. The reference matrix (M) may be general and independent of a specific product (e.g., independent of product attribute information), and a weight matrix (W) associated with a specific product may be calculated as a calculation result of a product feature vector of the specific product in the same reference matrix (M).

[0088] According to various embodiments, reference information (eg, reference matrix (M)) may be updated according to the results of advertisement execution, and detailed description will be described later.

[0089] According to various embodiments, the server 130 may provide at least one celebrity image from a plurality of celebrity images as a result of recommending a celebrity image based on the weight matrix (W). For example, the server 130 may calculate the brand fit (or expected advertising effect (e.g., expected click-through rate (CTR))) from the celebrity attribute information of the celebrity image by using the determined weight matrix (W) associated with the product attribute information of the specific product, and the calculation method will be described in more detail with reference to the accompanying drawings described below. The server 130 may provide the electronic device 101 with information including at least one celebrity image with high brand fit (or expected advertising effect).

[0090] According to various embodiments, in operation 607, the electronic device 101 may display a screen including a celebrity image and a predetermined number of celebrity features. According to various embodiments, the electronic device 101 may receive information including at least one celebrity image from the server 130, and display at least one recommended celebrity image according to the received information. According to various embodiments, the electronic device 101 may identify a specified number of celebrity features among a plurality of celebrity features. For example, the server 130 may identify a weight value of a weight matrix (W) associated with product attribute information of a particular product. The server 130 may identify a specified number of weight values ​​among the weight values. For example, the server 130 may compare the size of the weight values ​​to determine a specified number of weight values ​​having a larger size. The server 130 may provide (e.g., send) information including a predetermined number of weight values ​​determined, a predetermined number of celebrity features (e.g., located in the same row) corresponding to the predetermined number of weight values, and / or feature values ​​of a predetermined number of celebrity features to the electronic device 101. The electronic device 101 may display a specified number of celebrity features according to the information provided by the server 130. According to an embodiment, the electronic device 101 may display the celebrity feature values ​​of a specified number of celebrity features together, which will be described in more detail later with reference to the accompanying drawings. According to an embodiment, the electronic device 101 may display at least one recommended celebrity image together with another image of a celebrity corresponding to the recommended celebrity image, and will be described in more detail with reference to the following drawings.

[0091] Figure 7 is a flowchart illustrating an operation method of an online advertising content platform providing system according to various embodiments.

[0092] Hereinafter, operations 701 to 709 are described as being performed by the server 130 , but when the electronic device 101 includes the artificial intelligence model 233 , at least some of the operations 701 to 709 may be performed by the electronic device 101 .

[0093] Figure 6 Operation and Figure 7 The operation can be carried out organically, Figure 6 Operation and Figure 7 At least some of the operations may be omitted.

[0094] According to various embodiments, the server 130 may acquire at least one content of a specific product in operation 701. Operation 701 may be operation 601, and descriptions within an overlapping range are omitted.

[0095] According to various embodiments, the server 130 may acquire product attribute information in operation 703. Operation 703 may be operation 603, and descriptions within an overlapping range are omitted.

[0096] According to various embodiments, the server 130 may determine a weight value associated with the product attribute information in operation 705. For example, the server 130 may identify a weight value (e.g., a weight matrix (W) of Math. Formula 4) associated with the product attribute information of a specific product based on the acquired product attribute information of the specific product (e.g., Fp of Math. Formula 1) and reference information (e.g., a reference matrix (M) of Math. Formula 5).

[0097] According to various embodiments, the server 130 may determine the brand fit (suitability) (or expected advertising effect) of each celebrity image based on the determined weight value in operation 707. According to various embodiments, for a plurality of celebrity images, the server 130 may store each celebrity image in association with celebrity attribute information of the celebrity image (e.g., Fc of Mathematical Formula 2). The server 130 may determine the brand fit (or expected advertising effect) of each celebrity image based on the calculation result of the celebrity attribute information of each celebrity image and the determined weight value associated with a specific product (e.g., the weight matrix (W) of Mathematical Formula 4).

[0098] Mathematical formula 6 S=W T ·F c

[0099] Referring to Mathematical Formula 6, the brand suitability (S) may be calculated as a 1×1 size value (i.e., a scalar value) as a result of a matrix multiplication operation of a 1×1 size weight matrix (W) and a C×1 size celebrity feature vector Fc. The brand suitability (S) of each celebrity image may be displayed by the electronic device 101 in the form of a number in percentage form and / or a graphic object, and will be described in more detail with reference to the drawings described later.

[0100] According to various embodiments, the server 130 may recommend at least one celebrity image associated with a specific product based on the determined brand suitability (or expected advertising effect) in operation 709. For example, the server 130 may recommend at least one celebrity image with a higher brand suitability (or expected advertising effect) among a plurality of celebrity images as a celebrity image associated with a specific product based on the brand suitability (or expected advertising effect) of each celebrity image.

[0101] Figure 8 is a diagram illustrating an example of a graphic object included in a celebrity recommendation screen according to various embodiments.

[0102] According to various embodiments, the celebrity recommendation screen may include a graphic object representing a specified number of celebrity features and / or feature values ​​of celebrity features (hereinafter referred to as celebrity feature values).

[0103] According to various embodiments, the electronic device 101 may display a specified number of celebrity features through a graphic object. For example, the displayed celebrity features may be celebrity features corresponding to a specified number of weight values ​​having a larger size (e.g., located in the same row) in the weight values ​​associated with the product attribute information of a specific product (e.g., the weight matrix (W) of Mathematical Formula 4).

[0104] According to various embodiments, the order in which celebrity features are displayed may be determined based on the size of the corresponding weight values. For example, the order from celebrity features corresponding to relatively large weight values ​​to celebrity features corresponding to relatively small weight values ​​may be displayed in a first direction (e.g., clockwise).

[0105] According to various embodiments, the electronic device 101 may display celebrity feature values ​​through graphic objects. For example, for each celebrity image, in the celebrity attribute information of each celebrity image, celebrity feature values ​​corresponding to a specified number of celebrity features displayed through the graphic object may be displayed.

[0106] Based on the above description, Figure 8 (a) to 8(c) are described.

[0107] See also Figure 8 (a) to Figure 8 (c), with respect to the image of a specific celebrity, the celebrity's features of "gorgeous", "luxury", "healthy", "cool" and "pretty" are displayed through graphic objects. This indicates that among the weight values ​​determined based on the product attribute information of a specific product, the weight values ​​corresponding to celebrity features such as "gorgeous", "luxury", "healthy", "cool" and "pretty" are relatively high. If all celebrity features are composed in the order of the first celebrity feature ("pretty"), the second celebrity feature ("cute"), the third celebrity feature ("cool"), the fourth celebrity feature ("gorgeous"), the fifth celebrity feature ("luxury"), and the sixth celebrity feature ("healthy"), it can be indicated that the fourth weight value is the largest, and the order of the following weight values ​​is the fifth weight value, the sixth weight value, and the first weight value.

[0108] See also Figure 8 (a) and Figure 8 (c) The numerical value (e.g., 9, 8, 7, 6, 5) corresponding to each celebrity feature is displayed by a graphic object. This indicates the celebrity feature value corresponding to each of the celebrity features "gorgeous", "luxurious", "healthy", "cool", and "beautiful". Figure 8 (a) and Figure 8 As shown in (b), the electronic device 101 can display the digital values ​​(e.g., celebrity feature values) in the form of a radial graph. Figure 8As shown in (c), the electronic device 101 can display the numerical value in the form of a polygon (such as a pentagon), and the polygon connects the points representing the numerical value (such as the celebrity feature value). Figure 8 As shown in (b), the numerical value corresponding to each celebrity feature (such as 9, 8, 7, 6, 5) can be omitted.

[0109] As described above, the features displayed by the graphic object may be determined based on the weight value associated with the product attribute information (e.g., according to the size of the weight value). That is, since the product attribute information may differ depending on the advertising target product, the set of celebrity features displayed by the graphic object may also differ depending on the advertising target product. In other words, when the advertising target product is the first product, the celebrity features of the first group may be displayed, and when the advertising target product is the second product having different product attribute information from the first product, the celebrity features of the second group may be displayed, and all or part of the celebrity features of the first group and the celebrity features of the second group may be different.

[0110] As described above, the numerical value displayed by the graphic object can be determined based on the celebrity feature value of each celebrity image (e.g., according to the size of the celebrity feature value). That is, since the celebrity attribute information may be different for each celebrity image, the numerical value (and / or the form of the radial graph) displayed by the graphic object may also be different for each celebrity image.

[0111] Fig. 9 Screens for receiving information on a specific product according to various embodiments are shown.

[0112] According to various embodiments, the electronic device 101 may display a screen 900 for receiving the content of a specific product to be advertised. For example, the advertiser may input information of the product to be advertised through the screen 900 displayed on the electronic device 101. For example, the screen 900 may include an interface for receiving product-related information (such as name, price, category, description, image) and information related to a preferred advertising model (such as gender, category, description). For example, the screen 900 may include an interface (such as 910 and 920) for receiving a product image and displaying the received image. For example, the screen 900 may further include an interface for specifying a country in which the advertiser wants to advertise. According to various embodiments, the electronic device 101 may provide the content of the product received through the screen 900 (for example, the name, price, category, description, image related to the product) to the server 130 to obtain product attribute information (for example, feature values ​​corresponding to multiple products) of the specific product that the advertiser wants to advertise, or obtain product attribute information (for example, feature values ​​corresponding to multiple product features) of the specific product that the advertiser wants to advertise based on the product content (for example, the name, price, category, description, image related to the product) received through the screen 900.

[0113] Fig.10 An example of a screen for receiving product category information according to various embodiments is shown.

[0114] According to various embodiments, the electronic device 101 may display a screen 1000 including a search interface 1010. For example, the electronic device 101 may receive a product category through the search interface 1010, and provide celebrity (or celebrity image) information related to the received product category. For example, as shown in the figure, when "cosmetics" is input through the search interface 1010, celebrities (or celebrity images) related to "cosmetics" may be searched. For example, at least one celebrity image of a celebrity input as the preferred product category "cosmetics", or a celebrity or at least one celebrity image with a large amount or high advertising performance for a specific product with the product category "cosmetics" may be searched. According to an embodiment, a search may be displayed based on a search performed through the screen 1000. Fig.11 a screen (eg, screen 1100) or Fig.14 screen (eg, screen 1400).

[0115] Fig.11 An example of a celebrity recommendation screen is shown in accordance with various embodiments.

[0116] According to various embodiments, the electronic device 101 may display a screen 1100 for recommending at least one celebrity image. For example, through the method described in the above figures, when at least one content of a specific product is input, the electronic device 101 may display a screen for recommending at least one celebrity image through a process of acquiring product attribute information of the specific product. Or, as described above with reference to Fig.10 As described above, based on the product category input through the search interface 1010, a recommendation result of at least one celebrity or at least one celebrity image associated with the input product category may be displayed.

[0117] According to various embodiments, the celebrity-related information displayed may include image, brand fit, popularity, advertising contract price, and exclusion-related information.

[0118] According to various embodiments, product suitability may be calculated based on each celebrity image and / or a weight value associated with each celebrity image.

[0119] According to various embodiments, "popularity" may be an indicator of the popularity of a celebrity in real time or at a future point in time. For example, "popularity" may be confirmed in real time based on Internet search volume data. For example, "popularity" may be calculated or predicted based on a future point in time based on the celebrity's future work plans. For example, the "popularity" of a celebrity may be confirmed in different ways (e.g., popularity in each country) depending on the country in which the advertisement is to be made.

[0120] According to various embodiments, the advertising contract price (Price) may represent the cost per click (CPC) of the celebrity. The advertising contract price may be preset and may be determined based on the advertising performance and / or real-time popularity of the advertisements previously executed by the celebrity. According to an embodiment, when the advertiser inputs an advertising execution period and / or an advertising budget, celebrities whose advertising contract prices do not fall within the range calculated according to the advertising execution period and / or the advertising budget may be excluded from the recommendation results.

[0121] According to various embodiments, the exclusion may indicate the product category that the celebrity excludes. For example, when a specific celebrity registers "alcohol" as an excluded product, when the category of the specific product is "alcohol", the specific celebrity may be excluded from the recommendation results.

[0122] According to various embodiments, the reference information obtained by searching for at least one celebrity image may be displayed in an area (eg, a left area) of the screen 1100. Figure 8 As shown, information related to celebrity features corresponding to relatively high specified number of weight values ​​can be provided. For example, celebrity features can be in text form and / or similar to Figure 8 The graphics object of the pentagon is provided.

[0123] According to various embodiments, the displayed celebrity-related information may include images 1110a, 1120a, and 1130a with the highest brand suitability among celebrity images of the celebrity. According to an embodiment, a plurality of images with high brand suitability among celebrity images of the celebrity may be displayed.

[0124] According to various embodiments, the displayed celebrity information may include graphic objects 1110b, 1120b, and 1130b representing celebrity feature values ​​for each celebrity (or each celebrity image). According to various embodiments, each graphic object may represent a celebrity feature value for each celebrity corresponding to a relatively large number of designated weight values. Figure 8 According to an embodiment, although celebrity features such as "gorgeous" may not be displayed in graphic objects 1110b, 1120b, and 1130b, celebrity feature values ​​corresponding to the celebrity feature values ​​may also be displayed, such as Figure 8 shown.

[0125] Fig.12a An example of a profile of a particular celebrity is shown in accordance with various embodiments. Figure 12b An example of a profile of a particular celebrity is shown in accordance with various embodiments.

[0126] According to various embodiments, when a specific celebrity (or an image of a specific celebrity) is selected from the recommendation results of celebrities, the electronic device 101 may display a screen 1210 including a profile of the selected celebrity.

[0127] According to various embodiments, the screen 1210 may include a representative image 1211, a name 1212, popularity 1213, a description 1214, an object 1215 (e.g., a radial graph) representing celebrity characteristics, a portfolio 1216, and an object representing a country 1217. According to various embodiments, the representative image 1211 may be an image displayed in a celebrity image recommendation screen (e.g., an image with high brand suitability), or may be an image preset by a celebrity.

[0128] According to various embodiments, the collection 1216 may include at least one image of the selected celebrity. According to an embodiment, when the selected celebrity has registered multiple images, the electronic device 101 may include a specified number of images with high brand suitability among the multiple images in the collection 1216 (or place them at the top of the collection 1216).

[0129] According to various embodiments, Fig.12a The popularity 1213 displayed on the screen 1210 of the display may be displayed as a graph according to time points. The popularity 1213 displayed on the screen 1210 may be determined according to at least one country represented by the object 1216 representing a country displayed on the screen 1210. For example, the object 1216 representing the country displayed on the screen 1210 may be an object 1217 representing the country in which the product is expected to be advertised.

[0130] According to various embodiments, Figure 12b The popularity 1223 displayed on the screen 1220 of the embodiment may be displayed as a graph according to time points. The popularity 1223 displayed on the screen 1220 may be determined based on at least one country represented by the object 1226 representing a country displayed on the screen 1220. For example, Figure 12b The country object 1227 displayed on the screen 1220 of the embodiment may be an object 1227 representing a specific country (eg, a country) in which the product is intended to be advertised, and accordingly, Fig.12a The popularity 1213 and Figure 12b The popularity 1223 displayed on the screen 1220 may be different.

[0131] According to various embodiments, objects 1215 and 1225 may represent a predetermined number (e.g., five) of celebrity features and their feature values ​​having a higher celebrity feature value among a plurality of celebrity features, and may be represented, for example, in the form of a radial graph. According to an embodiment, object 1215 may represent celebrity features corresponding to a predetermined number of weight values ​​with high weight values ​​associated with a specific product, similar to Figure 8 The graphical objects shown in and / or Fig.11 Graphic objects 1110a, 1110b and 1110c in.

[0132] Fig.13 Examples of interface screens for mediating negotiations between advertisers and celebrities are shown in accordance with various embodiments.

[0133] According to various embodiments, when a specific celebrity (or an image of a specific celebrity) is selected from the celebrity recommendation results, the electronic device 101 may display a screen 1300 including a negotiation mediation interface. The screen 1300 may include celebrity features and feature values, an advertising contract fee (or range) required by the celebrity, and / or a message box.

[0134] According to various embodiments, the electronic device 101 may provide an interface for negotiating with a selected celebrity. For example, the advertiser may set an advertising contract fee, input information to be sent to the celebrity through a message box, and allow the advertising contract fee and message to be sent to the celebrity for negotiation.

[0135] Fig.14 An example of a screen displaying a creator recommendation screen according to various embodiments is shown.

[0136] According to various embodiments, the electronic device 101 may display a screen 1400 for recommending at least one creator. For example, the screen 1400 may include an image, nickname, and specifications of the creator (e.g., technical fields that the creator can provide, advertising contract fees, and history). According to an embodiment, the at least one creator displayed may include information about the creator and / or specific products that the advertiser wants to advertise related to the input product category (e.g., "cosmetics"). The "pending" displayed on the left side of the screen 1400 may indicate the process of negotiating with the celebrity selected by the advertiser, and when the celebrity accepts the advertising contract fee proposed by the advertiser, etc., the "pending" may be changed to text such as "accepted".

[0137] Fig.15 Examples of profiles of specific authors are shown in accordance with various embodiments.

[0138] According to various embodiments, when a specific creator (or an image of a creator) is selected from the recommendation results of creators, the electronic device 101 may display a screen 1500 including a profile of the selected creator.

[0139] According to various embodiments, screen 1500 may include a representative image 1510 of the selected creator, a name (or nickname) 1520, specifications 1530 of the creator (e.g., technical fields that the creator can provide), popularity 1540, description 1550, an object 1560 representing the creator's capabilities (e.g., a radial graph), a portfolio 1570 (e.g., previous work results), and an object 1580 representing a country associated with the creator (e.g., the creator's nationality or the country in which the creator is active).

[0140] According to various embodiments, the collection 1570 may include at least one previous work product of the selected creator. According to an embodiment, when the selected creator has registered multiple works, the electronic device 101 may include a specified number of works with a higher degree of brand fit among the multiple works in the collection 1570 (or place them at the top of the collection 1570).

[0141] Fig.16 Examples of interface screens for mediating negotiations between advertisers and creators are shown in accordance with various embodiments.

[0142] According to various embodiments, the electronic device 101 may display a screen 1600 including a negotiation mediation interface. For example, the screen 1600 may include an image 1610 of a specific product that the advertiser wants to advertise, an image 1620 of a celebrity selected by the advertiser, information 1630 of the creator (e.g., the image, nickname, and specifications of the creator), an object 1640 for specifying a unit price of an advertising contract, and a message input 1650. The advertiser may negotiate with the selected creator through the screen 1600 displayed on the electronic device 101. The creator may determine the image 1610 of the specific product that the advertiser wants to advertise, the image 1620 of a celebrity selected by the advertiser, the advertising contract fee desired by the advertiser, and / or the message input by the advertiser, and the negotiation between the advertiser and the creator may be conducted accordingly.

[0143] Fig.17 An example of a results screen of an advertising mediation platform according to various embodiments is shown.

[0144] According to various embodiments, the electronic device 101 may display a result screen 1700 of the advertising mediation platform. For example, the progress status of the advertisement production work may be displayed through the screen 1700. For example, when an advertisement image is produced and registered based on an image 1710 of a specific product and an image 1720 of a celebrity selected by an advertiser received by the creator, the produced advertisement image 1740 may be displayed through the screen 1700.

[0145] According to various embodiments, the screen 1700 may further include an object 1730 (eg, a radial graph) representing celebrity characteristics, a conversation interface 1750 , a first indicator 1760 , and / or a second indicator 1770 .

[0146] According to various embodiments, the first indicator 1760 may be an object indicating whether the celebrity accepts the contract. Fig.17 Screen 1700 of shows that the celebrity has accepted the contract ("Accepted").

[0147] According to various embodiments, the second indicator 1770 may be an object indicating whether the creator accepts the contract. Fig.17Screen 1700 shows that the creator has not yet accepted the contract ("pending").

[0148] According to various embodiments, a conversation may be conducted between a celebrity or creator and an advertiser via the conversation interface 1750 .

[0149] Fig.18 Examples of advertisement execution results according to various embodiments are shown.

[0150] According to various embodiments, the electronic device 101 may display a screen indicating a result of advertisement execution (e.g., advertisement performance) 1800. For example, the screen 1800 may include a first image 1810 (e.g., an image of a celebrity used in an advertisement or a representative image of a celebrity), a second image 1820 (e.g., an advertisement image of advertisement execution), a profit / cost graph 1830, an object 1840 indicating an advertisement execution date, an object 1850 indicating an advertisement target country, and an object 1860 indicating advertisement performance.

[0151] According to various embodiments, the screen 1800 may include only one of the first image 1810 and the second image 1820, or may include both the first image 1810 and the second image 1820. The electronic device 101, the first external electronic device 110, or the second external electronic device 120 may determine which image of the first image 1810 and the second image 1820 to include depending on whether it is a screen provided to an advertiser, a screen provided to a celebrity, or a screen provided to a creator.

[0152] According to various embodiments, the object 1860 representing the advertising performance may include an object representing "click rate", "retention rate" and / or "sales" for each media (e.g., SNS). The electronic device 101 may check the advertising performance of the advertising image and display the object 1860 representing the advertising performance on the screen 1800. According to various embodiments, the electronic device 101 may update a reference value (e.g., reference matrix (M)) for determining the celebrity features of the celebrity (or celebrity image) used in the advertising image, the product attribute information of the target product of the advertising image, or the weight value of a specific product (or product attribute information) based on the advertising performance corresponding to the advertising image. For example, the server 130 may use the reference value (e.g., reference matrix (M)) in common independently of the specific product (or product attribute information), and may check the advertising performance of each advertising product based on the advertising image prepared for recommending the celebrity image for each advertising product based on the specific reference value (e.g., reference matrix (M)). The server 130 may update (or adjust) the specific reference value (e.g., reference matrix (M)) based on the advertising performance so that it reaches the reference value (e.g., reference matrix (M)) that maximizes the advertising performance.

[0153] Understandably, Figures 1 to 18The embodiments can be organically applicable to each other within the scope of application.

[0154] According to various embodiments, the online advertising content platform providing system includes a server and an electronic device, wherein the server is configured to obtain at least one content of a specific product from the electronic device, wherein the at least one content includes an image of the specific product, obtain product attribute information of the specific product based on the at least one content, wherein the product attribute information corresponds to a plurality of product features, and based on the product attribute information, provide celebrity image recommendation information determined based on a plurality of weight values ​​associated with the plurality of product features to the electronic device, wherein the celebrity image recommendation information includes at least one celebrity image, wherein the electronic device is configured to display a first screen including the at least one celebrity image and a predetermined number of celebrity features among a plurality of celebrity features based on the provided celebrity image recommendation information, wherein based on the specific product being the first product, the predetermined number of celebrity features includes a first group of celebrity features among the plurality of celebrity features, and wherein based on the specific product being the second product, the predetermined number of celebrity features includes a second group of celebrity features among the plurality of celebrity features.

[0155] According to various embodiments, the predetermined number of celebrity features may correspond to a predetermined number of weight values, respectively.

[0156] According to various embodiments, the first screen may include celebrity feature values ​​corresponding to the predetermined number of celebrity features corresponding to each of the at least one celebrity image.

[0157] According to various embodiments, the first screen may include at least one other celebrity image of a celebrity for each celebrity image included in the celebrity image recommendation information.

[0158] According to various embodiments, the electronic device may be further configured to, if a first celebrity image is selected among the at least one celebrity image displayed on the first screen, provide an intermediary interface with a first celebrity of the first celebrity image.

[0159] According to various embodiments, the electronic device may be further configured to display a second screen if a first celebrity image is selected from among the at least one celebrity image displayed on the first screen, the second screen providing an image of the specific product, the selected first celebrity image, and a larger predetermined number of weight values ​​among the multiple weight values ​​determined based on a result of the comparison.

[0160] According to various embodiments, the server may be configured to compare the sizes of the plurality of weight values, and determine a predetermined number of weight values ​​from among a plurality of larger weight values ​​based on a result of the comparison.

[0161] According to various embodiments, the server may be further configured to determine celebrity attribute information of a plurality of celebrity images; calculate the plurality of weight values ​​and the determined celebrity attribute information; and determine the at least one celebrity image based on a result of the calculation.

[0162] According to various embodiments, the multiple weight values ​​can be determined based on the specific product being the first product, based on a reference matrix and first product attribute information of the first specific product, and based on the specific product being the second product, based on the reference matrix and second product attribute information of the second specific product; and wherein the first product attribute information and the second product attribute information are at least partially different.

[0163] According to various embodiments, an operation method of an online advertising content platform providing system includes: obtaining at least one content of a specific product from the electronic device through the server, wherein the at least one content includes an image of the specific product; obtaining product attribute information of the specific product based on the at least one content through the server, wherein the product attribute information corresponds to a plurality of product features; providing celebrity image recommendation information determined based on a plurality of weight values ​​associated with a plurality of product features to the electronic device through the server based on the product attribute information, wherein the celebrity image recommendation information includes at least one celebrity image; and displaying a first screen including the at least one celebrity image and a predetermined number of celebrity features based on the celebrity image recommendation information through the electronic device, wherein based on the specific product being the first product, the predetermined number of celebrity features include a first group of celebrity features among the plurality of celebrity features, wherein based on the specific product being the second product, the predetermined number of celebrity features include a second group of celebrity features among the plurality of celebrity features, and the second group of celebrity features are at least partially different from the first group of celebrity features.

[0164] According to various embodiments, the predetermined number of celebrity features respectively correspond to a predetermined number of weight values.

[0165] According to various embodiments, the first screen may include celebrity feature values ​​corresponding to the predetermined number of celebrity features of each of the at least one celebrity image.

[0166] According to various embodiments, the first screen may further include at least one other celebrity image of a celebrity for each celebrity image included in the celebrity image recommendation information.

[0167] According to various embodiments, the operating method of the online advertising content platform providing system may further include an operation in which the electronic device provides an intermediary interface with the first celebrity of the first celebrity image when a first celebrity image is selected from the at least one celebrity image displayed on the first screen.

[0168] According to various embodiments, the operating method of the online advertising content platform providing system may further include, when the first celebrity image is selected among the at least one celebrity image displayed on the first screen, an operation of displaying a second screen through the electronic device, the second screen providing together an image of a specific product, the selected first celebrity image, and an advertising image generated based on the image of the specific product and the selected first celebrity image.

[0169] The electronic device according to various embodiments disclosed herein may be various types of devices. For example, the electronic device may include a portable communication device (such as a smart phone), a computer device, a portable multimedia device, a portable medical device, a camera, a wearable device, or a household appliance. The electronic device according to the embodiments of this document is not limited to the above devices.

[0170] It should be understood that the various embodiments of this article and the terms used in this article are not intended to limit the technical features described herein to specific embodiments, but include various changes, equivalents or substitutes of the corresponding embodiments. With respect to the description of the drawings, similar reference numerals may be used for similar or related components. Unless otherwise clearly indicated by the relevant context, the singular form of the noun corresponding to the project may include one or more projects. In this article, each of the 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", "at least one of A, B or C" may include any one of the items listed in the corresponding phrase in the phrase, or any and all combinations thereof. The terms "1st", "2nd" or "first" or "second" may be used to simply distinguish the corresponding component from other corresponding components, and the corresponding components are not limited in other aspects (such as importance or order). When a component (such as a first component) is referred to as "coupled" or "connected", whether or not the terms "functionally" or "communicationally" are used, it means that the component can be connected to other components directly (such as wired), wirelessly or through a third component.

[0171] The term "module" used in the embodiments herein may include units implemented in hardware, software or firmware, and may be used interchangeably with terms such as logic, logic block, component or circuit. A module may be an integrated component or a minimum unit or a portion thereof that performs one or more functions. For example, according to an embodiment, a module may be implemented in the form of an application specific integrated circuit (ASIC).

[0172] Various embodiments of the present disclosure may be implemented in the form of software (such as a program), and the software includes one or more instructions stored in a storage medium (such as an internal memory or an external memory) readable by a machine (such as an electronic device). For example, a processor (such as a processor) of a machine (such as an electronic device) may call at least one of the one or more instructions stored in the storage medium and execute the instruction. In this way, the machine can perform at least one function according to at least one called instruction. The one or more instructions may include code generated by a compiler or code executable by an interpreter. The machine-readable storage medium may be provided in the form of a non-temporary storage medium. Here, the term "non-temporary" means that the storage medium is a tangible device and does not include signals (such as electromagnetic waves), and the term does not distinguish between the case where data is semi-permanently stored in the storage medium and the case where data is temporarily stored.

[0173] According to an embodiment, the method according to various embodiments disclosed in the present disclosure may be included and provided in a computer program product. The computer program product can be traded between a seller and a buyer as a product. The computer program product can be distributed in the form of a machine-readable storage medium (such as a compact disc read-only memory (CD-ROM)), or distributed online (such as downloading or uploading) through an application store (such as PlayStoreTM), or distributed directly between two user devices (such as smartphones). In the case of online distribution, at least a portion of the computer program product can be temporarily generated or at least temporarily stored in a machine-readable storage medium, such as a memory of a manufacturer's server, an application store's server, or a relay server.

[0174] According to various embodiments, each component (such as a module or a program) in the above-mentioned components may include a single entity or multiple entities, and some of the multiple entities may be arranged in other components respectively. According to various embodiments, one or more components or operations in the above-mentioned corresponding components may be omitted, and one or more other components or operations may be added. Alternatively or in addition, multiple components (such as modules or programs) may also be integrated into one component. In this case, the integrated component may perform one or more functions of each component in the multiple components in the same or similar manner as the corresponding components in the multiple components before integration. According to various embodiments, the operations performed by modules, programs or other components may be performed sequentially, in parallel, repeatedly or heuristically, and may add one or omit one or more other operations in different sequential executions.

[0176] [reference numerals list]

[0177] 100: System

[0178] 101: Electronic devices

[0179] 110: First external electronic device

[0180] 120: Second external electronic device

[0181] 130: Server

[0182] 233: Artificial Intelligence Model

Claims

1. An online advertising content platform providing system, comprising: server; and Electronic devices, The server is constructed as follows: acquiring at least one content of a specific product from the electronic device, wherein the at least one content includes an image of the specific product, acquiring product attribute information of the specific product based on the at least one content, wherein the product attribute information corresponds to a plurality of product features, Based on the product attribute information, celebrity image recommendation information determined based on a plurality of weight values ​​associated with a plurality of product features is provided to the electronic device, wherein the celebrity image recommendation information includes at least one celebrity image, The electronic device is constructed as follows: Based on the provided celebrity image recommendation information, displaying a first screen including the at least one celebrity image and a predetermined number of celebrity features among the plurality of celebrity features, Wherein, based on the fact that the specific product is the first product, the predetermined number of celebrity features includes a first group of celebrity features among the plurality of celebrity features, wherein based on the fact that the specific product is the second product, the predetermined number of celebrity features includes a second group of celebrity features among the plurality of celebrity features, The second set of celebrity features is at least partially different from the first set of celebrity features.

2. The online advertising content platform providing system according to claim 1, The predetermined number of celebrity features respectively correspond to a predetermined number of weight values.

3. The online advertising content platform providing system according to claim 1, The first screen includes celebrity feature values ​​corresponding to the predetermined number of celebrity features of each of the at least one celebrity image.

4. The online advertising content platform providing system according to claim 1, Wherein the first screen further includes at least one other celebrity image of the celebrity for each celebrity image included in the celebrity image recommendation information.

5. The online advertising content platform providing system according to claim 1, The electronic device is further configured to provide an intermediary interface with a first celebrity of the first celebrity image if a first celebrity image is selected from the at least one celebrity image displayed on the first screen.

6. The online advertising content platform providing system according to claim 1, in, The electronic device is further configured to display a second screen, if a first celebrity image is selected from among the at least one celebrity image displayed on the first screen, the second screen providing an image of the specific product, the selected first celebrity image, and an advertising image generated based on the image of the specific product and the selected first celebrity image.

7. The online advertising content platform providing system according to claim 1, The server is constructed as follows: comparing the sizes of the multiple weight values, and A larger predetermined number of weight values ​​among the plurality of weight values ​​are determined based on a result of the comparison.

8. The online advertising content platform providing system according to claim 1, The server is further configured to: determining celebrity attribute information of a plurality of celebrity images; Calculating the plurality of weight values ​​and the determined celebrity attribute information; and The at least one celebrity image is determined based on a result of the calculation.

9. The online advertising content platform providing system according to claim 1, in, The multiple weight values ​​are: Based on the specific product being the first product, the determination is based on a reference matrix and first product attribute information of the first specific product, and Based on the specific product being the second product, the determination is based on the reference matrix and the second product attribute information of the second specific product; and The first product attribute information and the second product attribute information are at least partially different.

10. An electronic device, comprising: monitor; Memory; and at least one processor, wherein the at least one processor is configured to: acquiring at least one content for a specific product, wherein the at least one content includes an image of the specific product; acquiring product attribute information of the specific product based on the at least one content, wherein the product attribute information corresponds to a plurality of product features; Acquire a plurality of celebrity features of each of the plurality of celebrity images based on the plurality of celebrity images, wherein the plurality of celebrity features of each celebrity image includes a first feature of a first image of a specific celebrity and a second feature of a second image of the specific celebrity, and the first feature and the second feature are at least partially different; Based on the product attribute information, acquiring at least one celebrity image among the plurality of celebrity images and a predetermined celebrity feature among a plurality of features corresponding to the at least one celebrity image; Acquiring celebrity image recommendation information including the at least one celebrity image and the predetermined celebrity feature; as well as Based on the celebrity image recommendation information, a first screen including the at least one celebrity image and the predetermined celebrity feature is displayed on the display.

11. The electronic device according to claim 10, wherein the at least one processor is configured to determine a plurality of weight values ​​corresponding to each of the plurality of product features based on the product attribute information, and The at least one celebrity image and the predetermined celebrity feature corresponding to the at least one celebrity image are acquired based on the multiple weight values.

12. The electronic device according to claim 11, The predetermined celebrity feature corresponds to a predetermined number of weight values ​​among the plurality of weight values.

13. The electronic device according to claim 10, The first screen includes celebrity feature values ​​corresponding to the predetermined celebrity feature of each celebrity image in the at least one celebrity image.

14. The electronic device according to claim 10, Wherein the first screen further includes at least one other celebrity image of the celebrity for each celebrity image included in the celebrity image recommendation information.

15. The electronic device according to claim 10, The at least one processor is configured to provide an intermediary interface with the first celebrity of the first celebrity image if a first celebrity image is selected from among the at least one celebrity image displayed on the first screen.

16. The electronic device according to claim 10, in, If the first celebrity image is selected among the at least one celebrity image displayed on the first screen, the at least one processor is configured to display a second screen that provides an image of the specific product, the selected first celebrity image, and an advertising image generated based on the image of the specific product and the selected first celebrity image.

17. The electronic device according to claim 11, wherein the at least one processor is configured to: comparing the sizes of the multiple weight values, and A larger predetermined number of weight values ​​among the plurality of weight values ​​are determined based on a result of the comparison.

18. The electronic device according to claim 11, wherein the at least one processor is configured to: determining celebrity attribute information of a plurality of celebrity images, Calculating the plurality of weight values ​​and the determined celebrity attribute information, and The at least one celebrity image is determined based on a result of the calculation.

19. The electronic device according to claim 11, The multiple weight values ​​are: Based on the specific product being the first product, the determination is based on a reference matrix and first product attribute information of the first specific product, and Based on the specific product being a second product, the determination is based on the reference matrix and second product attribute information of the second specific product; and in, The first product attribute information and the second product attribute information are at least partially different.

20. A method for operating an electronic device, the method comprising: acquiring at least one content for a specific product, wherein the at least one content includes an image of the specific product; acquiring product attribute information of the specific product based on the at least one content, wherein the product attribute information corresponds to a plurality of product features; Acquire a plurality of celebrity features of each of the plurality of celebrity images based on the plurality of celebrity images, wherein the plurality of celebrity features of each celebrity image includes a first feature of a first image of a specific celebrity and a second feature of a second image of the specific celebrity, and the first feature and the second feature are at least partially different; Based on the product attribute information, acquiring at least one celebrity image among the plurality of celebrity images and a predetermined celebrity feature among a plurality of features corresponding to the at least one celebrity image; Acquiring celebrity image recommendation information including the at least one celebrity image and the predetermined celebrity feature; as well as Based on the celebrity image recommendation information, a first screen including the at least one celebrity image and the predetermined celebrity feature is displayed.

Images