KR20220158654A - Method for supporting creating, server and terminal performing the method using neural networks - Google Patents

Abstract

The present invention relates to a creating assistance method using a neural network. The creating assistance method according to embodiments may comprise the operations of: receiving, by a transceiver, personal data of a creator account from a terminal; pre-processing, by a data collection unit, the personal data and classifying the personal data of the creator account into one of creator propensity information, creator capability information, content information, scenario information, and guest information; extracting, by a creator feature extraction unit, a feature vector from the creator propensity information and the creator capability information by using a creating resource recommendation model including a neural network; obtaining, by a creating resource recommendation unit, an output vector from the feature vector by using the creating resource recommendation model, and recommending a creating resource suitable for the output vector by comparing the output vector with a previously learned reference vector; collecting, by a subscriber information analyzing unit, personal data of a subscriber account and account information of the creator account, and analyzing the personal data of the subscriber account and the account information of the creator account; calculating, by a commerce element recommendation unit, product reliability for each of a plurality of products related to an uploaded video of the creator account based on an analysis result; and recommending, by the commerce element recommendation unit, a related product having the highest product reliability among the plurality of related products. Accordingly, content quality and commerce efficiency can be improved.

Description

Creation assisting method using neural network, server and terminal performing the method

Embodiments of the present invention relate to a technique of assisting creation, and to a technique of assisting creation by using a neural network.

MCN (Multi-Channel Network) refers to an organization that supports production, funding, promotion, copyright management, revenue creation/sales, customer acquisition, etc. in partnership with creators of video platforms such as YouTube, and shares a portion of channel revenue. do. As the way and form of video consumption diversifies, the status of one-person content creators, called creators, rises. As such, the channel network business that manages these one-person creators is MCN.

The MCN industry emerged as media technology developed. The fast speed of wired networks and the mobility of wireless networks and the advancement of high-speed broadband have expanded the basis for seamless video service to users. The video production environment has also improved in terms of technology, such as cheaper shooting equipment, easier usage, studio rental for single producers, and video production using smartphones.

Above all, with the expansion of video sharing platforms such as YouTube and video viewing through mobile devices, UCC (User Created Contents or UGC: User Generated Contents) content has emerged as a new media, especially among people in their teens and 20s, rather than conventional TV broadcasting. The change in video viewing patterns, such as the increase in video viewing through mobile devices, is considered to be the biggest factor in the growth background of the MCN industry.

In this way, as one-person media becomes a trend, creating your own content is changing into an era where you can make money beyond a simple hobby. The popularity of one-person broadcasting video content centered on YouTube, a representative content distribution channel and platform, is reorganizing the existing media market and content industry along with the growth of creators. As the market grew, single content creators were called creators and took a step further from gaining individual brand value to making profits.

The background art described above is possessed or acquired by the inventor in the process of deriving the disclosure of the present application, and cannot necessarily be said to be known art disclosed to the general public prior to the present application.

Embodiments may provide a technique for recommending elements necessary for the creator's content quality and commerce improvement.

Technical tasks to be achieved in the embodiments are not limited to those mentioned above, and other technical tasks not mentioned will be considered by those skilled in the art from various embodiments to be described below. can

A creation assistance method according to embodiments may include receiving, by a transceiver, personal data of a creator account from a terminal; pre-processing the personal data and classifying the personal data of the creator account into one of creator tendency information, creator ability information, content information, scenario information, and guest information, by a data collection unit; extracting, by a creator feature extractor, a feature vector from the creator propensity information and the creator capability information by using a creating resource recommendation model constructed of a neural network; A creating resource recommender obtains an output vector from the feature vector using the creating resource recommendation model, and compares the output vector with a pre-learned reference vector to create a resource suitable for the output vector. action to recommend; collecting personal data of the subscriber account and account information of the creator account and analyzing the personal data of the subscriber account and the account information of the creator account, by the subscriber information analyzer; calculating, by a commerce element recommender, product reliability for each of a plurality of related products related to the uploaded video of the creator account based on the analysis result; and recommending, by the commerce element recommendation unit, a related product having the highest product reliability among the plurality of related products.

The classifying operation may include: pre-processing the personal data for each type by the data collecting unit and outputting a pre-processing vector for each type; obtaining a normalized vector by normalizing the preprocessing vector for each type by the data collection unit; and classifying the normalization vector into one of creator propensity information, creator capability information, content information, scenario information, and guest information.

The extracting of the feature vector may include obtaining a creator propensity vector from the creator propensity information using a creating resource recommendation model; obtaining a creator capability vector from the creator capability information by using a creating resource recommendation model; and obtaining the feature vector by performing concatenation on the creator propensity vector and the creator capability vector.

The recommending of the creating resource may include calculating a distance between the output vector and each of a plurality of pre-learned reference vectors; outputting a reference vector corresponding to a reference vector having the smallest distance based on the calculation result; and recommending a combination of content information, scenario information, and guest information corresponding to the reference vector.

The creative resource recommendation model includes an input layer, one or more hidden layers, and an output layer, each of the plurality of learning data is composed of learning creator tendency information, learning creator ability information, and a correct answer criterion vector, and the correct answer label is content Corresponding to a combination of information, scenario information and guest information, each learning data is input to the input layer of the creating resource recommendation model and passes through the one or more hidden layers and output layers to output an output vector; An output vector is input to a loss function layer connected to the output layer, and the loss function layer outputs a loss value using a loss function that compares the output vector with a correct answer reference vector, and the parameter of the creating resource recommendation model. Is learned in a direction in which the loss value decreases,

[mathematical expression]

The loss function follows the above equation, where n is the number of training data for each criterion, y and j are identifiers representing the criterion, C is a constant value, M is the number of criterion, x_y is the training data is criterion y A probability value belonging to , x_j may mean a probability value that training data belongs to criterion j, and L may mean a loss value.

A learning feature vector is obtained from learning data belonging to a first criterion among a plurality of learning data by using a consulting creating resource recommendation model, and each reference vector of a criterion different from the first criterion is obtained by using the creating resource recommendation model. Residuals between the learning feature vectors are obtained, and the high-intensity learning parameter is based on the reference residuals between the first reference vector of the first reference and the second reference vector of the second reference and the feature residuals between the first reference vector and the learning feature vector. Is calculated, and a loss value is calculated through a loss function including residuals between feature vectors and each reference vector of a residual and a criterion different from the first criterion to which a high-intensity learning parameter is applied to the feature residual, and based on the loss value Parameters or one or more reference vectors of the creating resource recommendation model may be learned.

The operation of analyzing the personal data of the subscriber account and the account information of the creator account includes the average of the classification identifiers of the videos watched by the subscriber at the current time and the subscriber's personal data from the personal data of the subscriber account. obtaining an average of identifiers of related products of videos viewed by the user; And from the creator's account information, the number of videos uploaded by the creator, the number of views of the uploaded video by the creator, the number of comments on the uploaded video by the creator, the video length, the number of advertisements included in the uploaded video, An operation of acquiring a total number of times that the uploaded video of the creator has been linked, a classification identifier of the video uploaded by the creator, and an identifier of a product related to the uploaded video may be included.

The operation of calculating the product reliability includes the number of videos uploaded by the creator, the number of views of the uploaded video by the creator, the number of comments on the uploaded video by the creator, the length of the video, the number of advertisements included in the uploaded video, The number of times the uploaded video of the creator has been linked, the classification identifier of the video uploaded by the creator, the average of the classification identifier of the video watched by the subscriber at the current time, the identifier of the related product of the uploaded video, the subscriber watched Calculating the product reliability based on an average of identifiers of related products of one video, an average of classification identifiers of videos watched by the subscriber at a current time, and an average of identifiers of related products of videos watched by the subscriber can do.

In the operation of calculating the product reliability, the product reliability is calculated according to the following equation,

[mathematical expression]

In the above equation, n is the number of uploaded videos of the creator, k is a variable identifying the uploaded videos of the creator, γ_k is a normalized value of the number of views of the uploaded videos of the creator, and f_k is the creator Represents a normalized value of the total sum of the number of comments, the length of the video, the number of included advertisements, and the number of times the video was linked, T is the classification identifier of the video uploaded by the creator, t is the current time An average of classification identifiers of videos watched by the subscriber, R may represent an identifier of a product related to the uploaded video, and r may represent an average of identifiers of a product related to the video watched by the subscriber.

A creation assistance method according to an embodiment may include transmitting, by a transceiver, personal data of a creator account stored in a memory to a server; receiving, by the transceiver, information about a related product recommended from the server; and displaying information on the related product by means of a display, wherein the server pre-processes the personal data to convert the personal data of the creator account to creator preference information, creator ability information, content information, and scenario information. and guest information, and a feature vector is extracted from the creator propensity information and the creator capability information using a creative resource recommendation model composed of a neural network, and the feature vector is extracted using the creating resource recommendation model. An output vector is obtained from, a creative resource suitable for the output vector is recommended by comparing the output vector with a pre-learned reference vector, personal data of the subscriber account and account information of the creator account are collected, and the subscriber account 's personal data and account information of the creator account are analyzed, and product reliability for each of a plurality of related products related to the uploaded video of the creator account is calculated based on the analysis result, and by the commerce element recommendation unit, Among the plurality of related products, the related product having the highest product reliability may be recommended.

A server according to an embodiment includes a transceiver; data collection unit; Creator feature extraction unit; Creation resource recommendation department; Subscriber information analysis unit; and a commerce element recommendation unit, wherein the transceiver receives personal data of a creator account from a terminal, and the data collection unit pre-processes the personal data to convert personal data of the creator account into creator preference information, creator ability information, and contents. information, scenario information, and guest information, and the creator feature extraction unit extracts a feature vector from the creator propensity information and the creator capability information using a creating resource recommendation model composed of a neural network, The creating resource recommendation unit obtains an output vector from the feature vector using the creating resource recommendation model, and recommends a creating resource suitable for the output vector by comparing the output vector with a pre-learned reference vector; The subscriber information analysis unit collects personal data of the subscriber account and account information of the creator account, analyzes the personal data of the subscriber account and account information of the creator account, and the commerce element recommendation unit, based on the analysis result, Product reliability for each of a plurality of related products related to the uploaded video of the creator account may be calculated, and the commerce element recommendation unit may recommend a related product having the highest product reliability among the plurality of related products.

A terminal according to an embodiment includes a processor; transceiver; Memory; and a display, wherein the transceiver transmits personal data of a creator account stored in a memory to a server, the transceiver receives information about a related product recommended from the server, and the display displays information on the related product. The personal data is pre-processed by the server, and the personal data of the creator account is classified as one of creator tendency information, creator ability information, content information, scenario information, and guest information, and created by a neural network. A feature vector is extracted from the creator propensity information and the creator capability information using the Eating resource recommendation model, an output vector is obtained from the feature vector using the creating resource recommendation model, and the output vector is pre-learned. Creating resources suitable for the output vector are recommended by comparing the reference vector, personal data of the subscriber account and account information of the creator account are collected, personal data of the subscriber account and account information of the creator account are analyzed, , Product reliability for each of a plurality of related products related to the uploaded video of the creator account is calculated based on the analysis result, and the commerce element recommendation unit has the highest product reliability among the plurality of related products. Products can be recommended.

According to embodiments, the creation assistance method may improve the creator's content quality and commerce efficiency.

Effects obtainable from the embodiments are not limited to the effects mentioned above, and other effects not mentioned are clearly derived and understood by those skilled in the art based on the detailed description below. It can be.

BRIEF DESCRIPTION OF THE DRAWINGS Included as part of the detailed description to aid understanding of the embodiments, the accompanying drawings provide various embodiments and, together with the detailed description, describe technical features of the various embodiments.
1 is a diagram illustrating a configuration of an electronic device according to an exemplary embodiment.
2 is a diagram showing the configuration of a program according to an embodiment.
3 is a diagram showing the entire configuration of a creation assistance system according to an embodiment.
4 is a flowchart illustrating an operation of a creation assistance method according to an exemplary embodiment.
5 is a diagram illustrating an example of a display screen of a terminal in which a related product is recommended according to an exemplary embodiment.
6 is a diagram for explaining a learning process of a creating resource recommendation model according to an embodiment.
7 is a diagram showing the structure of a server according to an embodiment.
8 is a diagram illustrating a structure of a terminal according to an exemplary embodiment.

The following embodiments combine elements and features of the embodiments in a predetermined form. Each component or feature may be considered optional unless explicitly stated otherwise. Each component or feature may be implemented in a form not combined with other components or features. In addition, various embodiments may be configured by combining some components and/or features. The order of operations described in various embodiments may be changed. Some components or features of one embodiment may be included in another embodiment, or may be replaced with corresponding components or features of another embodiment.

In the description of the drawings, procedures or steps that may obscure the gist of various embodiments are not described, and procedures or steps that can be understood by those skilled in the art are not described. did

Throughout the specification, when a part is said to "comprising" or "including" a certain element, it means that it may further include other elements, not excluding other elements, unless otherwise stated. do. In addition, terms such as “… unit”, “… unit”, and “module” described in the specification mean a unit that processes at least one function or operation, which is hardware or software or a combination of hardware and software. can be implemented as Also, “a or an”, “one”, “the” and like terms are used herein in the context of describing various embodiments (particularly in the context of the claims below). Unless otherwise indicated or clearly contradicted by context, both the singular and the plural can be used.

Hereinafter, embodiments according to various embodiments will be described in detail with reference to the accompanying drawings. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The detailed description set forth below in conjunction with the accompanying drawings is intended to describe exemplary embodiments of various embodiments, and is not intended to represent a single embodiment.

In addition, specific terms used in various embodiments are provided to help understanding of various embodiments, and the use of these specific terms may be changed into other forms without departing from the technical spirit of various embodiments. .

1 is a diagram illustrating a configuration of an electronic device according to an exemplary embodiment.

1 is a block diagram of an electronic device 101 within a network environment 100, according to various embodiments. Referring to FIG. 1 , in a network environment 100, an electronic device 101 communicates with an electronic device 102 through a first network 198 (eg, a short-range wireless communication network) or through a second network 199. It may communicate with at least one of the electronic device 104 or the server 108 through (eg, a long-distance wireless communication network). According to one embodiment, the electronic device 101 may communicate with the electronic device 104 through the server 108 . According to an embodiment, the electronic device 101 includes a processor 120, a memory 130, an input module 150, an audio output module 155, a display module 160, an audio module 170, a sensor module ( 176), interface 177, connection terminal 178, haptic module 179, camera module 180, power management module 188, battery 189, communication module 190, subscriber identification module 196 , or the antenna module 197 may be included. In some embodiments, in the electronic device 101, at least one of these components (eg, the connection terminal 178) may be omitted or one or more other components may be added. In some embodiments, some of these components (eg, sensor module 176, camera module 180, or antenna module 197) are integrated into a single component (eg, display module 160). It can be. The electronic device 101 may also be referred to as a client, terminal, or peer.

The processor 120, for example, executes software (eg, the program 140) to cause at least one other component (eg, hardware or software component) of the electronic device 101 connected to the processor 120. It can control and perform various data processing or calculations. According to one embodiment, as at least part of data processing or operation, the processor 120 transfers instructions or data received from other components (e.g., sensor module 176 or communication module 190) to volatile memory 132. , processing commands or data stored in the volatile memory 132 , and storing resultant data in the non-volatile memory 134 . According to one embodiment, the processor 120 may include a main processor 121 (eg, a central processing unit or an application processor) or a secondary processor 123 (eg, a graphic processing unit, a neural network processing unit ( NPU: neural processing unit (NPU), image signal processor, sensor hub processor, or communication processor). For example, when the electronic device 101 includes the main processor 121 and the auxiliary processor 123, the auxiliary processor 123 may use less power than the main processor 121 or be set to be specialized for a designated function. can The secondary processor 123 may be implemented separately from or as part of the main processor 121 .

The secondary processor 123 may, for example, take the place of the main processor 121 while the main processor 121 is in an inactive (eg, sleep) state, or the main processor 121 is active (eg, running an application). ) state, together with the main processor 121, at least one of the components of the electronic device 101 (eg, the display module 160, the sensor module 176, or the communication module 190) It is possible to control at least some of the related functions or states. According to one embodiment, the auxiliary processor 123 (eg, image signal processor or communication processor) may be implemented as part of other functionally related components (eg, camera module 180 or communication module 190). have. According to an embodiment, the auxiliary processor 123 (eg, a neural network processing device) may include a hardware structure specialized for processing an artificial intelligence model. AI models can be created through machine learning. Such learning may be performed, for example, in the electronic device 101 itself where the artificial intelligence model is performed, or may be performed through a separate server (eg, the server 108). The learning algorithm may include, for example, supervised learning, unsupervised learning, semi-supervised learning or reinforcement learning, but in the above example Not limited. The artificial intelligence model may include a plurality of artificial neural network layers. Artificial neural networks include deep neural networks (DNNs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), restricted boltzmann machines (RBMs), deep belief networks (DBNs), bidirectional recurrent deep neural networks (BRDNNs), It may be one of deep Q-networks or a combination of two or more of the foregoing, but is not limited to the foregoing examples. The artificial intelligence model may include, in addition or alternatively, software structures in addition to hardware structures.

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

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

The input module 150 may receive a command or data to be used by a component (eg, the processor 120) of the electronic device 101 from the outside of the electronic device 101 (eg, a user). The input module 150 may include, for example, a microphone, a mouse, a keyboard, a key (eg, a button), or a digital pen (eg, a stylus pen).

The sound output module 155 may output sound signals to the outside of the electronic device 101 . The sound output module 155 may include, for example, a speaker or a receiver. The speaker can be used for general purposes such as multimedia playback or recording playback. A receiver may be used to receive an incoming call. According to one embodiment, the receiver may be implemented separately from the speaker or as part of it.

The display module 160 may visually provide information to the outside of the electronic device 101 (eg, a user). The display module 160 may include, for example, a display, a hologram device, or a projector and a control circuit for controlling the device. According to one embodiment, the display module 160 may include a touch sensor set to detect a touch or a pressure sensor set to measure the intensity of force generated by the touch.

The audio module 170 may convert sound into an electrical signal or vice versa. According to one embodiment, the audio module 170 acquires sound through the input module 150, the sound output module 155, or an external electronic device connected directly or wirelessly to the electronic device 101 (eg: Sound may be output through the electronic device 102 (eg, a speaker or a headphone).

The sensor module 176 detects an operating state (eg, power or temperature) of the electronic device 101 or an external environmental state (eg, a user state), and generates an electrical signal or data value corresponding to the detected state. can do. According to one embodiment, the sensor module 176 may include, for example, a gesture sensor, a gyro sensor, an air pressure sensor, a magnetic sensor, an acceleration sensor, a grip sensor, a proximity sensor, a color sensor, an IR (infrared) sensor, a bio sensor, It may include a temperature sensor, humidity sensor, or light sensor.

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

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

The haptic module 179 may convert electrical signals into mechanical stimuli (eg, vibration or motion) or electrical stimuli that a user may perceive through tactile or kinesthetic senses. According to one embodiment, the haptic module 179 may include, for example, a motor, a piezoelectric element, or an electrical stimulation device.

The camera module 180 may capture still images and moving images. According to one embodiment, the camera module 180 may include one or more lenses, image sensors, image signal processors, or flashes.

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

The battery 189 may supply power to at least one component of the electronic device 101 . According to one embodiment, the battery 189 may include, for example, a non-rechargeable primary cell, a rechargeable secondary cell, or a fuel cell.

The communication module 190 is a direct (eg, wired) communication channel or a wireless communication channel between the electronic device 101 and an external electronic device (eg, the electronic device 102, the electronic device 104, or the server 108). Establishment and communication through the established communication channel may be supported. The communication module 190 may include one or more communication processors that operate independently of the processor 120 (eg, an application processor) and support direct (eg, wired) communication or wireless communication. According to one embodiment, the communication module 190 is a wireless communication module 192 (eg, a cellular communication module, a short-range wireless communication module, or a global navigation satellite system (GNSS) communication module) or a wired communication module 194 (eg, : a local area network (LAN) communication module or a power line communication module). Among these communication modules, a corresponding communication module is a first network 198 (eg, a short-range communication network such as Bluetooth, wireless fidelity (WiFi) direct, or infrared data association (IrDA)) or a second network 199 (eg, legacy It may communicate with the external electronic device 104 through a cellular network, a 5G network, a next-generation communication network, the Internet, or a telecommunications network such as a computer network (eg, a LAN or a WAN). These various types of communication modules may be integrated as one component (eg, a single chip) or implemented as a plurality of separate components (eg, multiple chips). The wireless communication module 192 uses subscriber information (eg, International Mobile Subscriber Identifier (IMSI)) stored in the subscriber identification module 196 within a communication network such as the first network 198 or the second network 199. The electronic device 101 may be identified or authenticated.

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

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

According to various embodiments, the antenna module 197 may form a mmWave antenna module. According to one embodiment, the mmWave antenna module includes a printed circuit board, an RFIC disposed on or adjacent to a first surface (eg, a lower surface) of the printed circuit board and capable of supporting a designated high frequency band (eg, mmWave band); and a plurality of antennas (eg, array antennas) disposed on or adjacent to a second surface (eg, a top surface or a side surface) of the printed circuit board and capable of transmitting or receiving signals of the designated high frequency band. can do.

At least some of the components are connected to each other through a communication method between peripheral devices (eg, a bus, general purpose input and output (GPIO), serial peripheral interface (SPI), or mobile industry processor interface (MIPI)) and signal ( e.g. commands or data) can be exchanged with each other.

According to an embodiment, commands or data may be transmitted or received between the electronic device 101 and the external electronic device 104 through the server 108 connected to the second network 199 . Each of the external electronic devices 102 or 104 may be the same as or different from the electronic device 101 . According to an embodiment, all or part of operations executed in the electronic device 101 may be executed in one or more external electronic devices among the external electronic devices 102 , 104 , or 108 . For example, when the electronic device 101 needs to perform a certain function or service automatically or in response to a request from a user or another device, the electronic device 101 instead of executing the function or service by itself. Alternatively or additionally, one or more external electronic devices may be requested to perform the function or at least part of the service. One or more external electronic devices receiving the request may execute at least a part of the requested function or service or an additional function or service related to the request, and deliver the execution result to the electronic device 101 . The electronic device 101 may provide the result as at least part of a response to the request as it is or additionally processed. To this end, for example, cloud computing, distributed computing, mobile edge computing (MEC), or client-server computing technology may be used. The electronic device 101 may provide an ultra-low latency service using, for example, distributed computing or mobile edge computing. In another embodiment, the external electronic device 104 may include an internet of things (IoT) device. Server 108 may be an intelligent server using machine learning and/or neural networks. According to one embodiment, the external electronic device 104 or server 108 may be included in the second network 199 . The electronic device 101 may be applied to intelligent services (eg, smart home, smart city, smart car, or health care) based on 5G communication technology and IoT-related technology.

The server 108 is connected to the electronic device 101 and can provide a service to the connected electronic device 101 . In addition, the server 108 may proceed with a membership sign-up procedure, store and manage various types of information of users subscribed as members, and provide various purchase and payment functions related to services. In addition, the server 108 may share execution data of service applications executed in each of the plurality of electronic devices 101 in real time so that the service can be shared among users. This server 108 may have the same configuration as a conventional web server or WAP server in terms of hardware. However, in terms of software, it may include a program module that is implemented through any language such as C, C++, Java, Visual Basic, or Visual C and performs various functions. In addition, the server 108 is generally connected to an unspecified number of clients and/or other servers through an open computer network such as the Internet, and receives requests from clients or other servers to perform tasks and derives and provides work results. It means a computer system and the computer software (server program) installed for it. In addition, the server 108, in addition to the above-described server program, a series of application programs operating on the server 108 and various databases (DB: Database, hereinafter referred to as It should be understood as a broad concept including DB"). Accordingly, the server 108 classifies member registration information and various information and data about games, stores them in a DB, and manages them. This DB may be implemented inside or outside the server 108 . In addition, the server 108 uses a server program provided in various ways according to operating systems such as DOS, Windows, Linux, UNIX, and Macintosh in general server hardware. As a representative example, a website used in a Windows environment, Internet Information Server (IIS), and CERN, NCSA, APPACH, etc. used in a Unix environment can be used. In addition, the server 108 may interoperate with an authentication system and a payment system for user authentication of services or payment for purchases related to services.

The first network 198 and the second network 199 are a connection structure capable of exchanging information between nodes such as terminals and servers, or a network connecting the server 108 and the electronic devices 101 and 104. means (Network). The first network 198 and the second network 199 include Internet, Local Area Network (LAN), Wireless Local Area Network (Wireless Local Area Network), Wide Area Network (WAN), Personal Area Network (PAN), and 3G , 4G, LTE, 5G, Wi-Fi, etc., but are not limited thereto. The first network 198 and the second network 199 may be closed first networks 198 and second networks 199 such as LAN and WAN, but are preferably open such as the Internet. The Internet is based on the TCP/IP protocol and several services that exist on its upper layer, such as HTTP (HyperText Transfer Protocol), Telnet, FTP (File Transfer Protocol), DNS (Domain Name System), SMTP (Simple Mail Transfer Protocol), SNMP ( Simple Network Management Protocol), Network File Service (NFS), and Network Information Service (NIS).

The database may have a general data structure implemented in a storage space (hard disk or memory) of a computer system using a database management program (DBMS). The database may have a data storage form in which data can be freely searched for (extracted), deleted, edited, added, and the like. Databases are relational database management systems (RDBMS) such as Oracle, Informix, Sybase, and DB2, or object-oriented database management such as Gemston, Orion, and O2. It can be implemented according to the purpose of an embodiment of the present disclosure using a system (OODBMS) and XML Native Databases such as Excelon, Tamino, and Sekaiju, and its functions may have appropriate fields or elements to achieve.

2 is a diagram showing the configuration of a program according to an embodiment.

2 is a block diagram 200 illustrating a program 140 according to various embodiments. According to one embodiment, the program 140 includes an operating system 142, middleware 144, or an application 146 executable in the operating system 142 for controlling one or more resources of the electronic device 101. can include The operating system 142 may include, for example, Android™, iOS™, Windows™, Symbian™, Tizen™, or Bada™. At least some of the programs 140 are, for example, preloaded in the electronic device 101 at the time of manufacture, or when used by a user, an external electronic device (eg, the electronic device 102 or 104), or a server ( 108)) can be downloaded or updated. All or part of program 140 may include a neural network.

The operating system 142 may control management (eg, allocation or reclamation) of one or more system resources (eg, process, memory, or power) of the electronic device 101 . Operating system 142 may additionally or alternatively include other hardware devices of electronic device 101 , such as input module 150 , sound output module 155 , display module 160 , audio module 170 . , sensor module 176, interface 177, haptic module 179, camera module 180, power management module 188, battery 189, communication module 190, subscriber identification module 196, or It may include one or more driver programs for driving the antenna module 197.

The middleware 144 may provide various functions to the application 146 so that the function or information provided from one or more resources of the electronic device 101 may be used by the application 146 . The middleware 144 includes, for example, the application manager 201, the window manager 203, the multimedia manager 205, the resource manager 207, the power manager 209, the database manager 211, and the package manager 213. ), connectivity manager 215, notification manager 217, location manager 219, graphics manager 221, security manager 223, call manager 225, or voice recognition manager 227. can

The application manager 201 may manage the life cycle of the application 146 , for example. The window manager 203 may manage one or more GUI resources used in a screen, for example. The multimedia manager 205 identifies, for example, one or more formats necessary for reproducing media files, and encodes or decodes a corresponding media file among the media files using a codec suitable for the selected format. can be done The resource manager 207 may manage a source code of the application 146 or a memory space of the memory 130 . The power manager 209 may manage, for example, the capacity, temperature, or power of the battery 189, and determine or provide related information necessary for the operation of the electronic device 101 by using corresponding information among them. . According to an embodiment, the power manager 209 may interoperate with a basic input/output system (BIOS) (not shown) of the electronic device 101 .

The database manager 211 may create, search, or change a database to be used by the application 146, for example. The package manager 213 may manage installation or update of applications distributed in the form of package files, for example. The connectivity manager 215 may manage, for example, a wireless connection or a direct connection between the electronic device 101 and an external electronic device. The notification manager 217 may provide a function for notifying a user of occurrence of a designated event (eg, an incoming call, message, or alarm), for example. The location manager 219 may manage location information of the electronic device 101, for example. The graphic manager 221 may manage, for example, one or more graphic effects to be provided to a user or a user interface related thereto.

Security manager 223 may provide system security or user authentication, for example. The telephony manager 225 may manage, for example, a voice call function or a video call function provided by the electronic device 101 . The voice recognition manager 227 transmits, for example, the user's voice data to the server 108, and at least partially based on the voice data, a command corresponding to a function to be performed in the electronic device 101; Alternatively, text data converted at least partially based on the voice data may be received from the server 108 . According to one embodiment, the middleware 244 may dynamically delete some existing components or add new components. According to one embodiment, at least part of the middleware 144 may be included as part of the operating system 142 or may be implemented as separate software different from the operating system 142 .

The application 146 includes, for example, a home 251, a dialer 253, an SMS/MMS 255, an instant message (IM) 257, a browser 259, a camera 261, and an alarm 263. , Contacts (265), Voice Recognition (267), Email (269), Calendar (271), Media Player (273), Albums (275), Watch (277), Health (279) (e.g. exercise or blood sugar) measurement of biometric information) or environmental information 281 (eg, measurement of atmospheric pressure, humidity, or temperature information). According to an embodiment, the application 146 may further include an information exchange application (not shown) capable of supporting information exchange between the electronic device 101 and an external electronic device. The information exchange application may include, for example, a notification relay application configured to transmit designated information (eg, a call, message, or alarm) to an external electronic device, or a device management application configured to manage an external electronic device. have. The notification relay application, for example, transmits notification information corresponding to a designated event (eg, mail reception) generated in another application (eg, the email application 269) of the electronic device 101 to an external electronic device. can Additionally or alternatively, the notification relay application may receive notification information from an external electronic device and provide the notification information to the user of the electronic device 101 .

The device management application is, for example, a power source (eg, turn-on or turn-off) of an external electronic device that communicates with the electronic device 101 or some component thereof (eg, a display module or a camera module of the external electronic device). ) or functions (eg brightness, resolution, or focus). The device management application may additionally or alternatively support installation, deletion, or update of an application operating in an external electronic device.

Throughout this specification, a neural network, a neural network, and a network function may be used interchangeably. A neural network may consist of a set of interconnected computational units, which may be generally referred to as “nodes”. These “nodes” may also be referred to as “neurons”. A neural network includes at least two or more nodes. Nodes (or neurons) constituting neural networks may be interconnected by one or more “links”.

In a neural network, two or more nodes connected through a link may form a relative relationship of an input node and an output node. The concept of an input node and an output node is relative, and any node in an output node relationship with one node may have an input node relationship with another node, and vice versa. As described above, the input node to output node relationship can be created around the link. More than one output node can be connected to one input node through a link, and vice versa.

In a relationship between an input node and an output node connected through one link, the value of the output node may be determined based on data input to the input node. Here, a node interconnecting the input node and the output node may have a weight. The weight may be variable, and may be changed by a user or an algorithm in order for the neural network to perform a desired function. For example, when one or more input nodes are interconnected by respective links to one output node, the output node is set to a link corresponding to values input to input nodes connected to the output node and respective input nodes. An output node value may be determined based on the weight.

As described above, in a neural network, two or more nodes are interconnected through one or more links to form an input node and output node relationship in the neural network. Characteristics of the neural network may be determined according to the number of nodes and links in the neural network, an association between the nodes and links, and a weight value assigned to each link. For example, when there are two neural networks having the same number of nodes and links and different weight values between the links, the two neural networks may be recognized as different from each other.

3 is a diagram showing the entire configuration of a creation assistance system according to an embodiment.

According to an embodiment, the creation assistance system 300 may optimize a combination of various creation resources in consideration of a creator's inclination or characteristics.

According to an embodiment, the creation assistance system 300 may generate content suitable for a creator's situation. For example, the creation assistance system 300 may generate a narrative about the celebrity by analyzing related search words and posts related to the celebrity with respect to a channel dealing with the celebrity.

According to one embodiment, the creation assistance system 300 may analyze the creator's subscribers and reinforce the commerce element by using the analysis result. For example, if there are many subscribers but low commerce, the creation assistance system 300 may recommend a product highly related to the subscribers to the creator based on purchase information of the subscribers. Through this, the creation assistance system 300 can maximize the creator's profitability in the creator's current condition.

According to one embodiment, the creation assistance system 300 may analyze personal data of subscribers of a channel. For example, the creation assistance system 300 may analyze keyword search information, viewing time zone or age group of a video. For example, the subscriber's personal data of the channel may include the subscriber's keyword search information, viewing time of the video, age, gender, asset, job, education, interest, purchase history, and the like.

According to one embodiment, the creation assistance system 300 may reinforce commerce elements. The creation assistance system 300 may recommend a related product most suitable for the creator's subscribers to the creator based on the analysis result. For example, the commerce element may include product information, marketing information, and price information. Marketing information may include market information, media information, and message information.

To this end, the creation assistance system 300 may include a server 310, a terminal 321, a network 340, and a database 311. A terminal may include a memory, a processor, an input unit, a transceiver, and a display. The input unit of the terminal 321 may receive login information for a user account. The server may include a transceiver, a model learning unit, a data collection unit, a creator feature extraction unit, a creating resource recommendation unit, a subscriber information analysis unit, and a commerce element recommendation unit.

A creator refers to Internet broadcasters, and means a person who produces Internet broadcasts or videos and sound sources with a certain theme. For example, creators may include video-oriented Internet broadcasters active on Afreeca TV, Twitch, YouTube, Naver TV, Kakao TV, etc., or audio-oriented Internet broadcasters active on Apple Podcast, Podbbang, and Patty.

A creative resource may refer to an element required by a creator to create content. The creating resource may include creator's disposition information, creator ability information, shooting information, editing information, subtitle information, graphic information, planning information, and the like.

Creator propensity information is information related to the creator's propensity, including the creator's interests and creator type information. Creator type information may be classified according to the progress method of the content, and may include, for example, an explanation type, a progress type, an interactive type, and a content type.

The creator ability information means the ability of the creator directly related to content, and may exclude shooting information, editing information, subtitle information, graphic information, and planning information. For example, creator capability information may include speech information, song information, facial expression information, progress information, appearance information, and content proficiency information.

Speech information refers to information indicating speech ability. The song information refers to information representing an ability to sing. The facial expression information refers to information representing an ability to create an effective facial expression for creation. The progress information refers to information representing the ability to progress. Appearance information refers to information indicating the degree to which a creator's appearance appeals to subscribers or is effective in creation. Content proficiency information refers to information indicating proficiency in a specific content, and for example, in the case of game content, may indicate information indicating proficiency in a corresponding game.

The shooting information refers to information related to shooting for creation. For example, the shooting information may include information about shooting equipment, a shooting director, and the like.

Editing information refers to information related to editing for creation. For example, the editing information may include information about an editor, editor's editing ability, and the like.

Subtitle information refers to information related to subtitles for creation. For example, the caption information may include information about a caption worker and caption producing ability of the caption worker.

Graphic information refers to information related to graphics for creation. For example, the graphic information may include information about a graphic designer and the graphic designer's ability.

Planning information refers to planning elements for creation and may include content information, scenario information, location information, guest information, and budget information. The content information refers to information about content that is a subject of creation. Scenario information refers to information about a context in which a content, which is a material, is planned. Place information means a place where creation is performed. Guest information refers to the person who participates with the creator. Budget information refers to the cost required for creation.

According to one embodiment, the transceiver of the server 310 may receive personal data of a creator account from a terminal.

The data collection unit of the server 310 may pre-process personal data and classify the personal data of the creator account into one of creator tendency information, creator ability information, content information, scenario information, and guest information.

The data collection unit of the server 310 may preprocess personal data for each type and output a preprocessed vector for each type. The data collection unit of the server 310 may obtain a normalized vector by normalizing the preprocessing vector for each type. The data collection unit of the server 310 may classify the normalization vector into one of creator propensity information, creator capability information, content information, scenario information, and guest information.

The server 310 may convert creator tendency information and creator capability information into a feature vector by using a creating resource recommendation model. A feature vector can be compared to a reference vector. The reference vector may be a vector reflecting a combination of content information, scenario information, and guest information. Based on the comparison result, a reference vector closest to the feature vector may be selected. The selected reference vector may be converted into creating resource recommendation information.

Specifically, the creator feature extractor of the server 310 may extract a feature vector from creator propensity information and creator capability information using a creation resource recommendation model composed of a neural network.

The creator feature extraction unit of the server 310 may obtain a creator propensity vector from creator propensity information by using a creating resource recommendation model. The creator feature extraction unit of the server 310 may obtain a creator capability vector from the creator capability information by using the creating resource recommendation model. The creator feature extraction unit of the server 310 may obtain a feature vector by performing concatenation on the creator propensity vector and the creator capability vector.

The creative resource recommendation unit of the server 310 obtains an output vector from a feature vector using a creative resource recommendation model, and recommends a creative resource suitable for the output vector by comparing the output vector with a pre-learned reference vector. can do.

The creating resource recommendation unit of the server 310 may calculate a distance between the output vector and each of the plurality of pre-learned reference vectors. The creating resource recommendation unit of the server 310 may output a reference vector corresponding to a reference vector having the smallest distance based on a calculation result. The creation resource recommendation unit of the server 310 may recommend a combination of content information, scenario information, and guest information corresponding to the reference vector.

The creating resource recommendation model may be configured as a neural network. The creating resource recommendation model may include an input layer, one or more hidden layers, and an output layer.

The creating resource recommendation model may output a loss value by comparing a reference vector with a feature vector corresponding to creator tendency information and creator capability information. The creating resource recommendation model may update parameters including weights in a direction in which a loss value decreases. Thereafter, the creating resource recommendation model may update the reference vector in a direction in which the loss value becomes smaller.

Specifically, each of the plurality of learning data consists of learning creator tendency information, learning creator ability information, and a correct answer criterion vector, the correct answer label corresponds to a combination of content information, scenario information, and guest information, and each learning data is It is input to the input layer of the targeting resource recommendation model, passes through one or more hidden layers and output layers to output an output vector, the output vector is input to the loss function layer connected to the output layer, and the loss function layer is the output vector and the correct answer reference vector A loss value is output using a loss function that compares , and parameters of the creating resource recommendation model may be learned in a direction in which the loss value becomes smaller.

[mathematical expression]

The loss function may follow the equation In the equation, n is the number of training data for each criterion, y and j are identifiers representing the criterion, C is a constant value, M is the number of criterion, x_y is a probability value that the training data belongs to criterion y, x_j is the probability that the training data belongs to criterion j A probability value belonging to , L may mean a loss value.

A learning feature vector is obtained from learning data belonging to a first criterion among a plurality of learning data by using a consulting creating resource recommendation model, and each reference vector of a criterion different from the first criterion is obtained by using the creating resource recommendation model. Residuals between the learning feature vectors are obtained, and the high-intensity learning parameter is based on the reference residuals between the first reference vector of the first reference and the second reference vector of the second reference and the feature residuals between the first reference vector and the learning feature vector. Is calculated, and a loss value is calculated through a loss function including residuals between feature vectors and each reference vector of a residual and a criterion different from the first criterion to which a high-intensity learning parameter is applied to the feature residual, and based on the loss value Parameters or one or more reference vectors of the creating resource recommendation model may be learned.

A matching model may be learned in advance based on a plurality of learning data. A learning feature vector may be obtained from learning data belonging to a first criterion among a plurality of learning data by using a matching model. Here, the criterion means a classification target, and may include, for example, a specific trade name or menu. Residuals between each reference vector of a criterion different from the first criterion and the learning feature vector may be obtained by using the matching model.

A high-intensity learning parameter may be calculated based on a reference residual between a first reference vector of the first reference and a second reference vector of the second reference and a feature residual between the first reference vector and the learning feature vector. For example, the first reference vector may be a vector representing the user's business name or menu, and the second reference vector may be a vector representing a business name or menu unrelated to the user's business.

A loss value may be calculated through a loss function including residuals between feature vectors and reference vectors of residuals to which high-intensity learning parameters are applied to feature residuals and criteria different from the first criterion.

Based on the loss value, parameters of the matching model or one or more reference vectors may be learned. The learning device may learn parameters or one or more reference vectors of the matching model in a direction in which the loss value decreases or increases.

와 제2 기준의 제2 기준 벡터

간의 기준 레지듀얼

및 제1 기준 벡터

와 특징 벡터

간의 특징 레지듀얼

를 기초로 고강도 학습 파라미터

을 계산할 수 있다. 여기서, 제2 기준 벡터

는 하나 이상의 기준 벡터 중에서 제1 기준 벡터

와 레지듀얼의 차이가 가장 작은 기준 벡터일 수 있다.The learning device may further improve the ability of the neural network to answer correctly and the discrimination ability by applying the high-intensity learning parameter. For example, the learning device may use the loss function of Equation 2. The learning device is a first reference vector of a first reference

and the second criterion vector of the second criterion

Baseline residual of the liver

and the first reference vector

and feature vector

Liver Characteristics Residual

High-intensity learning parameters based on

can be calculated. Here, the second reference vector

Is a first reference vector among one or more reference vectors

It may be a reference vector having the smallest difference between the residual and the residual.

는 수학식 4에 의해 계산될 수 있다. 제1 기준 벡터

와 제j 기준의 제j 기준 벡터 wj의 내적을 코사인 값으로 하는 레지듀얼이 기준 레지듀얼

일 수 있다. 특징 레지듀얼

는 수학식 3에 의해 계산될 수 있다. 제1 기준 벡터

와 특징 벡터

의 내적을 코사인 값으로 하는 레지듀얼이 특징 레지듀얼

일 수 있다.Criteria Residual

can be calculated by Equation 4. First reference vector

and the residual whose cosine is the dot product of the j criterion vector wj of the j criterion is the criterion residual

can be Features Residual

can be calculated by Equation 3. First reference vector

and feature vector

Residual characterized by a residual whose cosine is the dot product of

can be

에 대한 특징 레지듀얼

및 기준 레지듀얼

간의 차이의 비율을 계산할 수 있다. 수학식 4을 참조하면, 특징 레지듀얼이 작을수록 고강도 학습 파라미터는 커지고, 기준 레지듀얼이 클수록 고강도 학습 파라미터가 커진다. 특징 레지듀얼이 작다는 것은 입력 이미지의 특징 벡터와 정답 기준을 의미하는 제1 기준의 제1 기준 벡터의 유사도가 높다는 것을 의미하며, 입력 이미지가 속한 기준을 잘 맞추는 것을 의미한다. 기준 레지듀얼이 크다는 것은 서로 다른 기준 간의 구별 능력이 크다는 것을 의미한다. The learning device may calculate a ratio of a difference between a feature residual to a reference residual and a reference residual as a high-intensity learning parameter. The learning device is the reference residual according to Equation 4

Features for Residual

and reference residual

The ratio of the difference between them can be calculated. Referring to Equation 4, the higher the feature residual is, the higher the high-intensity learning parameter is, and the larger the reference residual is, the higher the high-intensity learning parameter is. A small feature residual means that the similarity between the feature vector of the input image and the first reference vector of the first criterion indicating the correct answer criterion is high, and it means that the criterion to which the input image belongs is well matched. A large criterion residual means that the ability to discriminate between different criteria is large.

The performance of the neural network is recognized as degraded due to the high-intensity learning parameter, and the learning device learns the neural network more strongly. Due to the high-intensity learning parameter, as the residual inside the trigonometric function of the correct answer term in Equation 2 increases, the proportion of the correct term decreases, and the loss value of the loss function increases. As a result, the learning device learns the neural network more excessively, and the learning effect may be higher than when there is no high-intensity learning parameter. As the learning progresses and the performance of the neural network improves, the high-intensity learning parameter increases, and the learning result of the neural network can be further improved.

The subscriber information analysis unit of the server 310 may collect personal data of the subscriber account and account information of the creator account, and analyze the personal data of the subscriber account and the account information of the creator account.

The subscriber information analysis unit of the server 310 obtains an average of classification identifiers of videos watched by subscribers at the current time and an average of identifiers of products related to videos watched by subscribers from personal data of subscriber accounts. can

The subscriber information analysis unit of the server 310 determines the number of videos uploaded by the creator, the number of views of the uploaded video by the creator, the number of comments on the uploaded video by the creator, the length of the video, and the number of advertisements included in the uploaded video from the creator's account information. number, the total number of times the creator's uploaded video was linked, a category identifier of the video uploaded by the creator, and an identifier of a product related to the uploaded video.

The commerce element recommendation unit of the server 310 may calculate product reliability for each of a plurality of related products related to the uploaded video of the creator account based on the analysis result.

The commerce element recommendation unit of the server 310 indicates the number of uploaded videos by the creator, k is a variable for identifying the uploaded videos by the creator, γ_k is a normalized value of the number of views of the uploaded videos by the creator, and f_k is the uploaded video by the creator. Represents the normalized value of the sum of the number of comments on the video, the length of the video, the number of ads included in the uploaded video, and the number of times the creator's uploaded video was linked, T is the classification identifier of the video uploaded by the creator, t is The average of classification identifiers of videos watched by subscribers at the current time, R is the identifier of related products of uploaded videos, r is the average of identifiers of related products of videos watched by subscribers, and the classification of videos watched by subscribers at the current time Product reliability may be calculated based on the average of the identifiers and the average of the identifiers of related products of the videos watched by the subscriber.

The commerce element recommendation unit of the server 310 may calculate product reliability according to the following equation.

[mathematical expression]

In the equation, n is the number of uploaded videos by the creator, k is a variable that identifies the uploaded videos by the creator, γ_k represents the normalized value of the number of views of the creator's uploaded videos, and f_k is the comments of the creator's uploaded videos. Represents the normalized value of the sum of the number of videos, the length of the video, the number of ads included, and the number of times the video was linked, T is the classification identifier of the video uploaded by the creator, and t is the classification identifier of the videos watched by subscribers at the current time R is the average of identifiers of products related to uploaded videos, and r is the average of identifiers of products related to videos watched by subscribers.

The commerce element recommendation unit of the server 310 may recommend a related product having the highest product reliability among a plurality of related products.

According to one embodiment, the transceiver of the terminal 321 may transmit personal data of the creator account stored in the memory to the server. The transceiver of the terminal 321 may receive information about recommended related products from the server. The display of the terminal 321 may display information about related products.

Here, the personal data is pre-processed by the server 310, and the personal data of the creator account is classified as one of creator tendency information, creator ability information, content information, scenario information, and guest information, and recommended creation resources configured with a neural network. A feature vector is extracted from creator propensity information and creator ability information using the model, an output vector is obtained from the feature vector using a creating resource recommendation model, and an output vector is obtained by comparing the output vector with a pre-learned reference vector. Appropriate creative resources are recommended, personal data of subscriber accounts and account information of creator accounts are collected, personal data of subscriber accounts and account information of creator accounts are analyzed, and based on the analysis results, uploaded videos and Product reliability for each of a plurality of related products is calculated, and a related product having the highest product reliability among the plurality of related products may be recommended by the commerce element recommendation unit.

4 is a flowchart illustrating an operation of a creation assistance method according to an exemplary embodiment.

According to one embodiment, in operation 401, the transceiver may operate to receive personal data of a creator account from the terminal.

According to an embodiment, in operation 403, the personal data may be preprocessed by the data collection unit to classify the personal data of the creator account into one of creator preference information, creator ability information, content information, scenario information, and guest information. have.

According to an embodiment, in operation 405, a feature vector may be extracted from creator propensity information and creator capability information by using a creating resource recommendation model configured with a neural network by the creator feature extractor.

According to an embodiment, in operation 407, the creating resource recommendation unit obtains an output vector from the feature vector using the creating resource recommendation model, and compares the output vector with a pre-learned reference vector. Creation resources suitable for the output vector can be recommended.

According to an embodiment, in operation 409, the subscriber information analysis unit collects personal data of the subscriber account and account information of the creator account, and analyzes the personal data of the subscriber account and the account information of the creator account.

According to an embodiment, in operation 411, the commerce element recommender may calculate product reliability for each of a plurality of related products related to the uploaded video of the creator account based on the analysis result.

According to an embodiment, in operation 413, a related product having the highest product reliability among a plurality of related products may be recommended by the commerce element recommendation unit.

5 is a diagram illustrating an example of a display screen of a terminal in which a related product is recommended according to an exemplary embodiment.

6 is a diagram for explaining a learning process of a creating resource recommendation model according to an embodiment.

According to an embodiment, the learning apparatus is a matching model including a preceding layer 603, a feature extraction unit 605, a classification unit 607, a high-intensity learning parameter calculation unit 609, and a loss value calculation unit 611 ( 600) may be included. Each of the feature extractor 605, classifier 607, high-intensity learning parameter calculator 609, and loss value calculator 611 may be composed of one or more layers. The preceding layer 603, feature extraction unit 605, classification unit 607, high-intensity learning parameter calculation unit 609, and loss value calculation unit 611 may be included in one neural network or configured as separate objects. may be

The classification unit 607 may include a plurality of layers corresponding to a plurality of criteria. Parameters of the layer corresponding to each criterion may form a vector. Here, a vector of a layer corresponding to each criterion may be referred to as a criterion vector.

The learning device may receive learning data 601 as an input. Learning data 601 may be input to a preceding layer 603 . The preceding layer 603 may output a vector from the training data 601 . A vector may be input to the feature extractor 605 . The feature extractor 605 may output feature vectors. The feature vectors 613 and 319 may be transmitted to the high intensity learning parameter calculator 609 and the loss value calculator 611 .

According to one embodiment, the feature vector may be passed to the classification unit 607 . According to another embodiment, the vector output from the preceding layer 603 may be transferred to the classification unit 607 . The classification unit 607 may output a classification result of the feature vector. The classification results 617 and 319 may be transmitted to the high-intensity learning parameter calculation unit 609 and the loss value calculation unit 611 .

The high-intensity learning parameter calculation unit 609 may calculate the high-intensity learning parameter based on the feature vector 613 and the classification result 617 . The learning device calculates the high-intensity learning parameter based on a reference residual between a first reference vector of the first reference and a second reference vector of the second reference and a feature residual between the first reference vector and the feature vector. The learning device may calculate a ratio of a difference between a feature residual to a reference residual and a reference residual as a high-intensity learning parameter.

The loss value calculator 611 may calculate a loss value based on the feature vector 615 , the classification result 619 , the reference label 623 , and the high-intensity learning parameter 621 . The reference label 623 means a correct answer. The learning apparatus may calculate a loss value through a loss function including a residual to which a high-intensity learning parameter is applied to the feature residual. The loss function is the ratio of the exponent values for the feature residuals of the first criterion to the sum of the exponent values for the feature residuals of all criteria except the first criterion and the sum of the exponent values for the feature residuals of the first criterion. can include As the learning progresses, the high-intensity learning parameter is gradually increased and learning becomes easier. The learning device can further enhance the effect of learning by determining and applying a suitable high-intensity learning parameter for each criterion and learning data.

7 is a diagram showing the structure of a server according to an embodiment.

According to an embodiment, the server 710 (eg, the server 310 of FIG. 3) includes a transceiver 701, a data collection unit 703, a creator feature extraction unit 705, and a creation resource recommendation unit 707. ), a subscriber information analysis unit 709 and a commerce element recommendation unit 711.

According to an embodiment, the transceiver 701 may receive personal data of a creator account from a terminal.

According to an embodiment, the data collection unit 703 may pre-process the personal data and classify the personal data of the creator account into one of creator tendency information, creator ability information, content information, scenario information, and guest information.

According to an embodiment, the creator feature extractor 705 may extract a feature vector from creator propensity information and creator capability information by using a creation resource recommendation model composed of a neural network.

According to an embodiment, the creating resource recommendation unit 707 obtains an output vector from a feature vector using a creating resource recommendation model, and compares the output vector with a pre-learned reference vector to suit the output vector. Can recommend creative resources.

According to an embodiment, the subscriber information analysis unit 709 may collect personal data of a subscriber account and account information of a creator account, and analyze the personal data of a subscriber account and the account information of a creator account.

According to an embodiment, the commerce element recommendation unit 711 may calculate product reliability for each of a plurality of related products related to the uploaded video of the creator account based on the analysis result. The commerce element recommendation unit 711 may recommend a related product having the highest product reliability among a plurality of related products.

8 is a diagram illustrating a structure of a terminal according to an exemplary embodiment.

According to an embodiment, the terminal 810 (eg, the terminal 321 of FIG. 3 ) may include a processor 801 , a transceiver 803 , a memory 805 , and a display 807 .

The transceiver 803 may transmit personal data of the creator account stored in the memory to the server. The transceiver 803 may receive information about recommended related products from the server. The display 807 may display information about related products.

Here, personal data is preprocessed by the server, and the personal data of the creator account is classified as one of creator propensity information, creator ability information, content information, scenario information, and guest information, and a creation resource recommendation model composed of a neural network is used. feature vectors are extracted from creator propensity information and creator ability information, output vectors are obtained from feature vectors using a creative resource recommendation model, and creation suitable for output vectors is obtained by comparing output vectors with pre-learned reference vectors. dating resources are recommended, the personal data of the subscriber account and the account information of the creator account are collected, the personal data of the subscriber account and the account information of the creator account are analyzed, and based on the analysis result, a plurality of videos related to the uploaded video of the creator account are collected. Product reliability for each related product is calculated, and a related product having the highest product reliability among a plurality of related products may be recommended by the commerce element recommendation unit.

The embodiments described above may be implemented as hardware components, software components, and/or a combination of hardware components and software components. For example, the devices, methods and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate (FPGA). array), programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. A processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will understand that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it can include. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. You can command the device. Software and/or data may be any tangible machine, component, physical device, virtual equipment, computer storage medium or device, intended to be interpreted by or provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

Claims (10)

receiving, by the transceiver, personal data of the creator account from the terminal;
pre-processing the personal data and classifying the personal data of the creator account into one of creator tendency information, creator ability information, content information, scenario information, and guest information, by a data collection unit;
extracting, by a creator feature extractor, a feature vector from the creator propensity information and the creator capability information by using a creating resource recommendation model constructed of a neural network;
A creating resource recommender obtains an output vector from the feature vector using the creating resource recommendation model, and compares the output vector with a pre-learned reference vector to create a resource suitable for the output vector. action to recommend;
collecting personal data of the subscriber account and account information of the creator account and analyzing the personal data of the subscriber account and the account information of the creator account, by the subscriber information analyzer;
calculating, by a commerce element recommender, product reliability for each of a plurality of related products related to the uploaded video of the creator account based on the analysis result; and
An operation of recommending, by the commerce element recommendation unit, a related product having the highest product reliability among the plurality of related products.
Including, a creation assist method. According to claim 1,
The classification operation,
pre-processing the personal data for each type and outputting a pre-processed vector for each type, by the data collecting unit;
obtaining a normalized vector by normalizing the preprocessing vector for each type by the data collection unit; and
An operation of classifying the normalization vector into one of creator propensity information, creator ability information, content information, scenario information, and guest information.
Including, a creation assist method. According to claim 1,
The operation of extracting the feature vector,
obtaining a creator propensity vector from the creator propensity information by using a creating resource recommendation model;
obtaining a creator capability vector from the creator capability information by using a creating resource recommendation model; and
Obtaining the feature vector by performing concatenation on the creator propensity vector and the creator capability vector.
Including, a creation assist method. According to claim 1,
The operation of recommending the creation resource,
calculating a distance between the output vector and each of a plurality of pre-learned reference vectors;
outputting a reference vector corresponding to a reference vector having the smallest distance based on the calculation result; and
An operation of recommending a combination of content information, scenario information, and guest information corresponding to the reference vector
Including, a creation assist method. According to claim 1,
The creating resource recommendation model includes an input layer, one or more hidden layers, and an output layer,
Each of the plurality of learning data consists of learning creator tendency information, learning creator ability information, and a correct answer criterion vector, the correct answer label corresponds to a combination of content information, scenario information, and guest information, and each learning data corresponds to the creating It is input to the input layer of the resource recommendation model, passes through the one or more hidden layers and an output layer, and outputs an output vector, the output vector is input to a loss function layer connected to the output layer, and the loss function layer outputs the output vector. A loss value is output using a loss function that compares a vector with a correct answer reference vector, and parameters of the creating resource recommendation model are learned in a direction in which the loss value becomes smaller,
[mathematical expression]

The loss function follows the above equation,
In the above equation, n is the number of training data for each criterion, y and j are identifiers representing the criterion, C is a constant value, M is the number of criterion, x_y is a probability value that the training data belongs to criterion y, x_j is the training data is the criterion The probability value belonging to j, L is the loss value,
A creative aid method. According to claim 1,
The operation of analyzing the personal data of the subscriber account and the account information of the creator account,
obtaining an average of classification identifiers of videos watched by the subscriber and an average of identifiers of products related to the videos watched by the subscriber at a current time from personal data of the subscriber account; and
From the account information of the creator, the number of videos uploaded by the creator, the number of views of the uploaded video by the creator, the number of comments on the uploaded video by the creator, the video length, the number of advertisements included in the uploaded video, Acquisition of the total number of times the creator's uploaded video has been linked, a classification identifier of the video uploaded by the creator, and an identifier of a product related to the uploaded video
Including, a creation assist method. According to claim 6,
The operation of calculating the product reliability,
The number of videos uploaded by the creator, the number of views of the uploaded video by the creator, the number of comments on the uploaded video by the creator, the length of the video, the number of advertisements included in the uploaded video, and the number of ads linked to the uploaded video by the creator. number of times, classification identifier of videos uploaded by the creator, average of classification identifiers of videos watched by the subscriber at the current time, identifier of related products of the uploaded videos, and average of identifiers of related products of videos watched by the subscriber , Calculating the product reliability based on the average of classification identifiers of the videos watched by the subscriber and the average of identifiers of related products of the videos watched by the subscriber at the current time
Including, a creation assist method. transmitting, by the transceiver, the personal data of the creator account stored in the memory to the server;
receiving, by the transceiver, information about a related product recommended from the server; and
Including an operation of displaying information about the related product by a display,
By the server, the personal data is preprocessed, and the personal data of the creator account is classified as one of creator tendency information, creator ability information, content information, scenario information, and guest information, and a creation resource recommendation model composed of a neural network is generated. feature vectors are extracted from the creator propensity information and the creator capability information, and an output vector is obtained from the feature vector using the creating resource recommendation model, and the output vector is compared with a pre-learned reference vector. Creative resources suitable for the output vector are recommended, personal data of the subscriber account and account information of the creator account are collected, personal data of the subscriber account and account information of the creator account are analyzed, and based on the analysis result The product reliability for each of a plurality of related products related to the uploaded video of the creator account is calculated, and the related product having the highest product reliability among the plurality of related products is recommended by the commerce element recommendation unit.
A creative aid method. transceiver;
data collection unit;
Creator feature extraction unit;
Creation resource recommendation department;
Subscriber information analysis unit; and
Includes a commerce element recommendation unit,
The transceiver receives personal data of a creator account from a terminal,
The data collection unit pre-processes the personal data and classifies the personal data of the creator account into one of creator tendency information, creator ability information, content information, scenario information, and guest information;
The creator feature extractor extracts a feature vector from the creator propensity information and the creator capability information using a creative resource recommendation model composed of a neural network;
The creating resource recommendation unit obtains an output vector from the feature vector using the creating resource recommendation model, and compares the output vector with a pre-learned reference vector to select a creating resource suitable for the output vector. recommended,
The subscriber information analysis unit collects personal data of the subscriber account and account information of the creator account, analyzes the personal data of the subscriber account and account information of the creator account,
The commerce element recommendation unit calculates product reliability for each of a plurality of related products related to the uploaded video of the creator account based on the analysis result,
The commerce element recommendation unit recommends a related product having the highest product reliability among the plurality of related products.
server. processor;
transceiver;
Memory; and
including a display;
The transceiver transmits the personal data of the creator account stored in the memory to the server,
The transceiver receives information about related products recommended from the server,
The display displays information about the related product,
By the server, the personal data is preprocessed, and the personal data of the creator account is classified as one of creator tendency information, creator ability information, content information, scenario information, and guest information, and a creation resource recommendation model composed of a neural network is generated. feature vectors are extracted from the creator propensity information and the creator capability information, and an output vector is obtained from the feature vector using the creating resource recommendation model, and the output vector is compared with a pre-learned reference vector. Creative resources suitable for the output vector are recommended, personal data of the subscriber account and account information of the creator account are collected, personal data of the subscriber account and account information of the creator account are analyzed, and based on the analysis result The product reliability for each of a plurality of related products related to the uploaded video of the creator account is calculated, and the related product having the highest product reliability among the plurality of related products is recommended by the commerce element recommendation unit.
terminal.

Images