KR20230019554A - Electronic apparatus and method for personal contents recommendation, and program stored in computer readable medium performing the same - Google Patents

Abstract

Provided are an electronic apparatus for personal content recommendation, a method thereof, and a program stored in a computer-readable recording medium to execute the same. The electronic apparatus for personal content recommendation may: collect information regarding a first interest area of a user; generate first recommended content information according to the first interest area; collect information regarding a second interest area of the user; generate second recommended content information according to the second interest area; generate user-customized content information by mixing the first recommended content information with the second recommended content information at a predetermined ratio; and transmit the user-customized content information to the user terminal. Therefore, the recommendation mode can be changed and provided according to the user request in real time.

Description

Electronic device and method for recommending individual content, and a program stored in a computer-readable recording medium to perform the same

The present invention relates to a content recommendation algorithm for each individual, and more specifically, an electronic device and method for recommending individual contents that can be recommended to a user by adjusting the ratio of contents related to a subject of interest to each individual and contents not related to the subject of interest to each individual. And, it relates to a program stored on a computer readable recording medium to perform it.

An artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike existing rule-based smart systems, machines learn, judge, and become smarter on their own. The more AI systems are used, the higher the recognition rate and the more accurate understanding of user preferences. Existing rule-based smart systems are gradually being replaced by Deep Learning-based AI systems.

Artificial intelligence technology consists of machine learning (deep learning) and element technologies using machine learning. Machine learning is an algorithm technology that classifies/learns the characteristics of input data by itself, and element technology is a technology that uses machine learning algorithms such as deep learning to mimic functions such as recognition and judgment of the human brain. It consists of technical fields such as understanding, inference/prediction, knowledge expression, and motion control.

The various fields where artificial intelligence technology is applied are as follows. Linguistic understanding is a technology for recognizing and applying/processing human language/characters, including natural language processing, machine translation, dialogue systems, question and answering, voice recognition/synthesis, and the like. Visual understanding is a technology for recognizing and processing objects like human vision, and includes object recognition, object tracking, image search, person recognition, scene understanding, space understanding, image improvement, and the like. Inference prediction is a technique of reasoning and predicting logically by judging information, and includes knowledge/probability-based reasoning, optimization prediction, preference-based planning, and recommendation. Knowledge expression is a technology that automatically processes human experience information into knowledge data, and includes knowledge construction (data creation/classification) and knowledge management (data utilization). Motion control is a technology for controlling the autonomous driving of a vehicle and the movement of a robot, and includes motion control (navigation, collision, driving), manipulation control (action control), and the like.

In general, in order to apply machine learning algorithms to real life, learning is performed in a trial and error method due to the nature of the basic methodology of machine learning. In particular, deep learning requires hundreds of thousands of iterations. It is impossible to execute this in an actual physical external environment, so instead, the actual physical external environment is virtually implemented on a computer and learning is performed through simulation.

Existing portal sites, Google advertisements, YouTube channels, Amazon bookstores, Netflix, Watcha, Apple Music, Spotify, etc. continuously expose users to content or products preferred by customers according to their personal activities and interests. Artificial intelligence (AI) machine learning recommendation algorithms have been used as much as possible. However, this recommendation algorithm has a problem in that it makes users more simple and narrow-minded, and can lead them to be buried as people trapped in narrow-minded knowledge.

In other words, it is changing into an environment where it is difficult to experience various fields or contents, and there is a problem in that it is difficult to continue to have colorful and creative open-mindedness by staying only in what one likes or likes.

A technical problem to be solved by the present invention is an electronic device and method for recommending individual contents that can be recommended to a user by adjusting the ratio of contents related to individual subjects of interest and contents not related to individual subjects of interest, and performing the same. It is to provide a program stored in a computer readable recording medium to do so.

Another technical problem to be solved by the present invention is to provide an electronic device and method for recommending content for individuals, which can change and provide a recommendation mode in real time at the request of a user, and a program stored in a computer readable recording medium to perform the same. are doing

The technical problem to be solved by the present invention is not limited to the above.

An electronic device for recommending content for each individual according to an embodiment of the present invention includes one or more processors; and one or more memories storing instructions that, when executed by the one or more processors, cause the one or more processors to perform calculations, wherein the one or more processors collect information about a first region of interest of a user; Forming first recommended content information according to a first area of interest, collecting information on a second area of interest of the user, forming second recommended content information according to the second area of interest, and forming the first recommended content information and mixing the second recommended content information at a preset ratio to form user-customized content information, and transmit the user-customized content information to a user terminal.

As an embodiment, the information on the first region of interest may include information on a subject searched by the user a preset number of times or more.

As an example, the information on the second region of interest may include information on the user's call record, text record, and search record.

As an example, the information on the user's call record may include information on the age, residence area, and area of interest of the call counterpart.

As an example, the information on the text record of the user may include information on the age, residence area, and area of interest of the text counterpart.

As an embodiment, the information on the text record of the user is obtained by extracting a text message including a hyper-link among text messages received by a user terminal, and visiting a website linked by a hyperlink in the extracted text message. It can contain information retrieved from .

As an embodiment, the one or more processors may form user-customized content information by adjusting a mixing ratio of the first recommended content information and the second recommended content information based on a user request.

one or more processors according to an embodiment of the present invention; and one or more memories storing instructions for causing the one or more processors to perform calculations when executed by the one or more processors. Collecting information about; forming first recommended content information according to the first region of interest; collecting information on a second region of interest of a user; forming second recommended content information according to the second region of interest; forming user-customized content information by mixing the first recommended content information and the second recommended content information at a preset ratio; and transmitting the user-customized content information to a user terminal.

As one embodiment, receiving a user request; and forming user-customized content information by adjusting a mixing ratio of the first recommended content information and the second recommended content information based on the user request.

one or more processors according to an embodiment of the present invention; And a computer program stored in a computer-readable recording medium so as to be executable by a computer including one or more memories in which instructions for causing the one or more processors to perform operations when executed by the one or more processors are stored, the user's first interest Collecting information about the area; forming first recommended content information according to the first region of interest; collecting information on a second region of interest of a user; forming second recommended content information according to the second region of interest; forming user-customized content information by mixing the first recommended content information and the second recommended content information at a preset ratio; and transmitting the user-customized content information to a user terminal.

An electronic device and method for recommending content for each individual according to an embodiment of the present invention and a program stored in a computer-readable recording medium to perform the same, according to an embodiment of the present invention, a ratio of content related to a topic of interest to each individual and content not related to a topic of interest to each individual. It can be adjusted and recommended to the user. As a result, it is possible to form a balanced mindset by receiving various contents that are not biased to a specific field.

In addition, an electronic device and method for recommending contents for each individual according to an embodiment of the present invention, and a program stored in a computer-readable recording medium to perform the same, content related to a subject of interest for each individual and a subject of interest for each individual in real time at the request of a user You can adjust the ratio of content that is not related to. In this way, content can be recommended in consideration of the user's taste, and user convenience can also be improved.

1 is an exemplary view showing the configuration of a content recommendation environment for each individual according to an embodiment of the present invention.
2 is an exemplary view showing the configuration of an electronic device for recommending content for each individual according to an embodiment of the present invention.
4 is a diagram for explaining learning of a neural network according to an embodiment of the present invention.
5 is a flowchart showing a procedure of a content recommendation method for each individual according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the technical idea of the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided so that the disclosed content will be thorough and complete, and the spirit of the present invention will be sufficiently conveyed to those skilled in the art.

In this specification, when an element is referred to as being on another element, it means that it may be directly formed on the other element or a third element may be interposed therebetween.

In addition, although terms such as first, second, and third are used to describe various elements in various embodiments of the present specification, these elements should not be limited by these terms. These terms are only used to distinguish one component from another. Therefore, what is referred to as a first element in one embodiment may be referred to as a second element in another embodiment. Each embodiment described and illustrated herein also includes its complementary embodiments. In addition, in this specification, 'and/or' is used to mean including at least one of the elements listed before and after.

In the specification, expressions in the singular number include plural expressions unless the context clearly dictates otherwise. In addition, the terms "comprise" or "having" are intended to designate that the features, numbers, steps, components, or combinations thereof described in the specification exist, but one or more other features, numbers, steps, or components. It should not be construed as excluding the possibility of the presence or addition of elements or combinations thereof.

In addition, in this specification, "connection" is used to mean both indirectly and directly connecting a plurality of components. In addition, "connection" is a concept including physical connection as well as electrical connection.

In addition, in the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

1 is an exemplary view showing the configuration of a content recommendation environment for each individual according to an embodiment of the present invention.

As shown in FIG. 1, the individual contents recommendation environment 100 includes a plurality of user terminals 110-1, ..., 110-n, an electronic device 120 for individual contents recommendation, a database 130, and a network ( N) may be included. According to an embodiment, the database 130 may be implemented in a cloud environment separately from the electronic device 120 for recommending content for each individual, but is not limited thereto, and the database 130 is an electronic device for recommending content for each individual. It may be provided in (120).

The plurality of user terminals 110-1, ..., 110-n, the electronic device 120 for recommending content for each individual, and the database 130 may be communicatively connected to each other through a network N. The plurality of user terminals 110-1, ..., 110-n store information on the user's first region of interest and information on the user's second region of interest, and store first and second information in real time or non-real time. Information on the region of interest may be transmitted to the electronic device 120 and/or the database 130 for recommending content for each individual through the network N. In addition, the plurality of user terminals 110-1, ..., 110-n receive a user request for adjusting the mixing ratio of the first recommended content information and the second recommended content information when the user-customized content information is formed, and the network (N ) through which it can be transmitted to the electronic device 120 for individual content recommendation. According to an embodiment, the user terminals 110-1, ..., 110-n include a smartphone, a tablet personal computer (PC), a mobile phone, a video phone, and an e-book reader (e -book reader), desktop personal computer (laptop personal computer), netbook computer (netbook computer), personal digital assistant (PDA), portable multimedia player (PMP), MP3 player, or wearable device device) (e.g. head-mounted-device (HMD) such as electronic glasses, electronic clothing, electronic bracelet, electronic necklace, electronic appcessory, electronic tattoo, smart car or smart watch) It may include at least one of, but is not limited thereto.

The network N may perform wireless or wired communication between the plurality of user terminals 110-1, ..., 110-n, the electronic device 120 for individual content recommendation, and the database 130. For example, the network N may include long-term evolution (LTE), LTE Advanced (LTE-A), code division multiple access (CDMA), wideband CDMA (WCDMA), wireless broadband (WiBro), wireless fidelity (WiFi) , Bluetooth, near field communication (NFC), global positioning system (GPS), or global navigation satellite system (GNSS). For example, the network N may perform wired communication according to a method such as universal serial bus (USB), high definition multimedia interface (HDMI), recommended standard 232 (RS-232), or plain old telephone service (POTS). may be

The database 130 may store various data. Data stored in the database 130 is acquired, processed, or obtained by at least one component of the electronic device 120 for recommending content for each user terminal (110-1, ..., 110-n) or individual. As the data used, it may include software (eg: program). Database 130 may include volatile and/or non-volatile memory. As an example, the database 130 may store information on the first and second regions of interest received from the plurality of user terminals 110-1 to 110-n.

In the present invention, the program is software stored in the database 130, and an operating system, an application, and/or an application for controlling the resources of the electronic device 120 for recommending content for each individual is used to recommend the electronic device 120 for recommending content for each individual. It can include middleware that provides various functions to applications so that they can utilize the resources of the system.

The electronic device 120 for individual content recommendation receives information on a first region of interest from a plurality of user terminals 110-1, ..., 110-n connected through a network N, and receives information on a first region of interest. First recommendation content information may be formed using the information on the region of interest. According to one embodiment, the information on the first region of interest may include information on a subject searched for by the user a preset number of times (eg, 10 times, 50 times, 100 times, etc.) or more, but is not limited thereto. don't For example, the information on the first area of interest may be related to topics (eg, real estate, games, travel, restaurants, etc.) that the user frequently searches using the user terminal (eg, 110-1). information may be included.

The electronic device 120 for recommending content for each individual receives information on a second region of interest from a plurality of user terminals 110-1, ..., 110-n connected through a network N, and receives information on a second region of interest. Second recommended content information may be formed using the information on the region of interest. According to an embodiment, the information on the second region of interest may include information on the user's call record, text record, and search record, but is not limited thereto. For example, the information on the second region of interest may include information on topics that the user does not usually search for frequently using the user terminal (eg, 110 - 1 ).

The electronic device 120 for recommending individual content allows the user to frequently search using a user terminal (eg, 110-1) using at least one information of the user's call record, text record, and search record. You can form information on topics that are not covered. According to an embodiment, the electronic device 120 for recommending content for each person extracts information about the age, residence area, and area of interest of the other party using the user's call record, and uses the extracted information to provide information to the user. Information on subjects that are not frequently searched for may be formed using the user terminal (eg, 110-1). For example, the electronic device 120 for recommending content for each person extracts that the other party is in his 40s, resides in Daegu, and his area of interest is real estate, using the user's call record, and displays the extracted information. Since the area of interest of the other party on the call is real estate, real estate can be extracted as information on the second area of interest of the user.

According to another embodiment, the electronic device 120 for recommending content for each person extracts information about the age, residence area, and area of interest of the text counterpart by using the user's text record, and uses the extracted information to provide information to the user. Information on topics that are not frequently searched for may be formed using the user terminal (eg, 110 - 1 ). For example, the electronic device 120 for recommending content for each individual uses the user's text record to extract that the text counterpart's age is in their 30s, their area of residence is Busan, and their area of interest is travel, and the extracted information is displayed. Since the residential area of the text counterpart is Busan, the game fair can be extracted as information on the user's second area of interest.

The electronic device 120 for recommending content for each individual extracts a text message including a hyper-link from text messages received by the user terminal 110-1, and links the extracted text message to the hyperlink. Information retrieved from a website can be formed as information about a text record. According to an embodiment, the electronic device 120 for recommending content for each individual extracts a hyperlink (eg, https://store.pinkfong.com/) from a user's text message, and converts the extracted hyperlink into the extracted hyperlink. Since the linked website is related to infants, early childhood education may be extracted as information on the user's second area of interest.

The electronic device 120 for recommending content for each individual is of first interest through a deep learning algorithm using artificial intelligence (AI) or a machine learning algorithm using big data. Although the first recommended content information may be formed using the information on the area and the second recommended content information may be formed using the information on the second area of interest, the first and second content information forming methods are not limited thereto. don't

The electronic device 120 for recommending content for each individual may form user-customized content information by mixing the formed first recommended content information and second recommended content information at a preset ratio. According to an embodiment, the electronic device 120 for recommending content for each individual sets the first recommended content information and the second recommended content information at 50:50, 40:60, 30:70, 20:80, 10:90, User-customized content information may be formed by mixing at preset ratios such as 60:40, 70:30, 80:20, and 90:10.

The electronic device 120 for recommending content for each individual may transmit the formed user-customized content information to the user terminal 110 - 1 having received the information on the first and second regions of interest through the network N. According to an embodiment, the electronic device 120 for recommending content for each individual may transmit user-customized content information to the user terminal 110-1 in the form of a digital packet.

The electronic device 120 for recommending content for each individual receives information on a first region of interest from any one user terminal (eg, 110-1) among a plurality of user terminals 110-1, ..., 110-n. may be received, and a topic of interest of the user indicated in the information on the first region of interest may be extracted by learning the received information on the first region of interest.

The electronic device 120 for recommending content for each individual receives information on a second region of interest from one of the plurality of user terminals 110-1, ..., 110-n, and The received information on the second region of interest may be learned to extract information on subjects that are not frequently searched for by the user indicated in the information on the second region of interest.

When receiving information on the new first and second areas of interest, the electronic device 120 for recommending content for each individual provides information about the user's interest topic and the new second area of interest indicated in the information on the new first area of interest. Information on topics that are not frequently searched by the user shown in the information is extracted and stored in the database 130, and topics of interest to the user previously stored in the database 130 and topics that the user does not usually search for frequently The database 130 may be updated based on newly stored information on topics of interest of the user and information on topics that the user does not usually search for frequently.

In addition, the electronic device 120 for recommending content for each individual includes first information indicating a first area of interest of the user, second information indicating first recommended content information according to the first area of interest, and second area of interest of the user. Obtaining one or more training data each including third information indicating the third information and fourth information indicating the second recommended content information according to the second region of interest, and using the machine learning algorithm of artificial intelligence technology to obtain one or more training data You can train a neural network model.

In the present invention, artificial intelligence (AI) means a technology that imitates human learning ability, reasoning ability, perception ability, etc., and implements it with a computer, and may include concepts such as machine learning and symbolic logic. there is. Machine learning (ML) is an algorithm technology that classifies or learns the characteristics of input data by itself. Artificial intelligence technology is a machine learning algorithm that analyzes input data, learns the result of the analysis, and can make judgments or predictions based on the result of the learning. In addition, technologies that mimic the functions of the human brain, such as recognition and judgment, using machine learning algorithms, can also be understood as artificial intelligence. For example, technical fields such as linguistic understanding, visual understanding, inference/prediction, knowledge expression, and motion control may be included.

Machine learning may refer to processing that trains a neural network model using experience of processing data. Through machine learning, computer software could mean improving its own data processing capabilities. A neural network model is constructed by modeling a correlation between data, and the correlation may be expressed by a plurality of parameters. The neural network model derives a correlation between data by extracting and analyzing features from given data, and optimizing the parameters of the neural network model by repeating this process can be referred to as machine learning. For example, a neural network model may learn a mapping (correlation) between an input and an output with respect to data given as an input/output pair. Alternatively, even when only input data is given, the neural network model may learn the relationship by deriving a regularity between given data.

An artificial intelligence learning model or a neural network model may be designed to implement a human brain structure on a computer, and may include a plurality of network nodes having weights while simulating neurons of a human neural network. A plurality of network nodes may have a connection relationship between them by simulating synaptic activities of neurons that transmit and receive signals through synapses. In the artificial intelligence learning model, a plurality of network nodes can send and receive data according to a convolutional connection relationship while being located in layers of different depths. The artificial intelligence learning model may be, for example, an artificial neural network model, a convolutional neural network model (CNN), and the like. As an embodiment, the artificial intelligence learning model may be machine-learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms for performing machine learning include Decision Tree, Bayesian Network, Support Vector Machine, Artificial Neural Network, Adaboost, and Perceptron. (Perceptron), genetic programming, clustering, etc. may be used.

Of these, CNNs are a type of multilayer perceptrons designed to use minimal preprocessing. A CNN consists of one or several convolution layers and general artificial neural network layers on top, and additionally utilizes weights and pooling layers. Thanks to this structure, CNN can fully utilize the input data with a two-dimensional structure. Compared to other deep learning structures, CNN shows good performance in both video and audio fields. CNNs can also be trained via standard back-propagation. CNNs are easier to train than other feedforward artificial neural network techniques and have the advantage of using fewer parameters.

Convolutional networks are neural networks that contain sets of nodes with bound parameters. The increasing size of available training data and the availability of computational power, combined with algorithmic advances such as piecewise linear unit and dropout training, have greatly improved many computer vision tasks. With huge data sets, such as those available for many tasks today, overfitting is not as important, and increasing the size of the network improves test accuracy. Optimal use of computing resources becomes a limiting factor. To this end, a distributed, scalable implementation of deep neural networks can be used.

2 is an exemplary diagram showing the configuration of an electronic device for recommending content for each individual according to an embodiment of the present invention.

As shown in FIG. 2, the electronic device 120 for recommending individual contents may include a receiver 121, a processor 122, a transmitter 123, a system bus 124, and a database 126. According to one embodiment, the receiver 121, the processor 122, the transmitter 123, and the database 126 may be connected to each other to be communicable through the system bus 124, and these components of the electronic device 120 At least one of them may be omitted or another element may be added to the electronic device 120 . In addition, "additionally" or alternatively, "some components may be integrated and implemented," or implemented as singular or plural entities.

The receiving unit 121 receives the first and second user terminals from any one user terminal (eg, 110-1) among a plurality of user terminals 110-1, ..., 110-n through the network (N). Information on the second region of interest may be received in the form of a digital packet 125 in real time or non-real time and transmitted to the processor 122 . In addition, when the user-customized content information is formed from the user terminal 110-1 through the network N, the receiver 121 digitally transmits a user request for adjusting the mixing ratio of the first recommended content information and the second recommended content information. It may be received in real time or non-real time in the form of a packet 120 and transmitted to the processor 122 .

The processor 122 may form first recommended content information according to information about the first region of interest received in the form of a digital packet 125 from the user terminal 110 - 1 through the receiver 121 . According to an embodiment, the processor 122 may use the information on the first area of interest to provide information on topics for which the user searches a preset number of times or more through the user terminal 110-1 as first recommended content. As information, it can be formed in the form of a digital packet 125.

The processor 122 may form second recommended content information according to information about a second region of interest received in the form of a digital packet 125 from the user terminal 110 - 1 through the receiver 121 . According to an embodiment, the processor 122 may use information on the second area of interest to allow the user to search by using information on call records, text records, and search records performed by the user through the user terminal 110-1. Information on topics that are not frequently searched for can be formed in the form of a digital packet 125 as second recommended content information.

The processor 122 may form user-customized content information in the form of a digital packet 125 by mixing the first and second recommended content information formed in the form of "digital packet 125" at a preset ratio. According to an embodiment, the processor 112 adjusts the mixing ratio of the first recommended content information and the second recommended content information based on a user request received in the form of a digital packet 125 from the user terminal 110-1. Thus, user-customized content information can be formed.

The processor 122 may transmit user-customized content information formed in the form of a digital packet 125 to the transmitter 123 through the system bus 124 .

The transmitter 123 may transmit the user-customized content information received from the processor 122 in the form of a digital packet 125 to the user terminal 110-1 that has received information on the first and second regions of interest.

The database 130 may store various data. The data stored in the database 130 is acquired by, or processed by, at least one component of the electronic device 120 for recommending content for each individual, a plurality of user terminals 110-1, ??, or 110-n. , 　 As data to be used, 　 may include software (eg: 　 program). The database 130 may include volatile and/or non-volatile memory. As an embodiment, the database 130 receives and stores information on the first and second regions of interest from a plurality of user terminals 110-1, ??, and 110-n through the network N, First and second recommended content information and user-customized content information may be received and stored from the electronic device 120 for individual content recommendation through the network N.

3 is a conceptual diagram illustrating real-time recommendation mode change according to an embodiment of the present invention.

As shown in FIG. 3, a plurality of user terminals 110-1, ??, and 110-n have real-time recommendations capable of changing the reflection ratio of first and second recommended content information in real time when user-customized content information is formed. A mode change screen may be displayed.

30:70

20:80 ??). 또한, 사용자 요청에 따라서 원모양(ROI-1)의 크기가 증가하면(ROI-2) 그에 비례하여 사용자 맞춤형 컨텐츠 정보 형성 시 제1 추천 컨텐츠 정보의 반영 비율을 증가시키고, 제2 추천 컨텐츠 정보의 반영 비율을 감소시킬 수 있다(예를 들어, 60:40

70:30

80:20 ??).In the center of the real-time recommendation mode change screen, a circle (ROI-1) indicating the user's first region of interest is displayed. When the user-customized content information is formed, the reflection ratio of the first and second recommended content information may be adjusted in real time by reducing (ROI-2) or inward of the circular shape (ROI-1) (ROI-3). According to an embodiment, when the electronic device 120 for recommending content for each individual and user-customized content information are formed, the reflection ratio of the first and second recommended content information may be set to 50:50, and a circular shape ( When the size of ROI-1) decreases (ROI-3), the reflection ratio of the first recommended content information may be reduced and the reflection ratio of the second recommended content information may be increased in proportion to the user-customized content information formation (eg For example, 40:60

30:70

20:80 ??). In addition, when the size of the circular shape (ROI-1) increases (ROI-2) according to the user's request, the reflection ratio of the first recommended content information is increased in proportion to the user-customized content information formation, and the second recommended content information's You can reduce the reflection ratio (e.g. 60:40

70:30

80:20 ??).

4 is a diagram for explaining learning of a neural network according to an embodiment of the present invention.

As shown in FIG. 4, the learning device uses the information on the first region of interest to form first recommended content information, and uses the information on the second region of interest to form second recommended content information. The network 127 can be trained. According to an embodiment, the learning device may be a separate subject different from the electronic device 120 for recommending individual contents, but is not limited thereto.

The neural network 127 includes an input layer 126 into which training samples are input and an output layer 128 into which training outputs are output, and can be learned based on differences between the training outputs and labels. Here, the labels may be defined based on the user's response level corresponding to the audio output signal. The neural network 127 is connected as a group of a plurality of nodes, and is defined by weights between the connected nodes and an activation function that activates the nodes.

The learning device may train the neural network 127 using a “Gradient Descent (GD)” technique or a “Stochastic Gradient Descent (SGD)” technique. The learning device may use a loss function designed by outputs and labels of the neural network.

The learning device may calculate a training error using a predefined loss function. The loss function may be predefined with labels, outputs, and parameters as input variables, where the parameters may be set by weights in the neural network 127 . For example, the loss function may be designed in the form of “Mean Square Error (MSE)”, the form of “entropy”, etc., and various techniques or methods may be employed in an embodiment in which the loss function is designed.

The learning device may find weights affecting the training error using a backpropagation technique. Here, "weights" are relationships between nodes in the neural network 127. The learning device may use the "SGD" technique using labels and outputs to optimize the weights found through the backpropagation technique. For example, the learning device may update weights of a loss function defined based on labels, outputs, and weights using the SGD technique.

According to an embodiment, the learning device may obtain information on the training first and second regions of interest, and extract training first and second recommended content information from the training information on the first and second training regions of interest. . The learning device may acquire pre-labeled information (first labels) for the information on the first and second training regions of interest, respectively, and the information on the training first and second regions of interest may be predefined. First labels indicating areas of interest (eg, games, real estate, travel, restaurants, early childhood education, etc.) may be obtained.

According to an embodiment, the learning device may generate first training feature vectors based on number sequence features, pattern features, and repetition features of information on first and second training regions of interest. Various methods may be employed to extract features of information on the training first and second regions of interest.

According to an embodiment, the learning device may obtain training outputs by applying the first training feature vectors to the neural network 127 . The learning device may train the neural network 127 based on the training outputs and the first labels. The learning device may train the neural network 127 by calculating training errors corresponding to training outputs and optimizing a connection relationship of nodes in the neural network 127 to minimize the training errors. The electronic device 120 for recommending content for each individual may extract first and second recommended content information from information on the first and second regions of interest using the neural network 127 that has been learned.

5 is a flowchart showing a procedure of a content recommendation method for each individual according to an embodiment of the present invention. Although process steps, method steps, algorithms, etc. are described in a sequential order in the flowchart of FIG. 5, such processes, methods, and algorithms may be configured to operate in any suitable order. In other words, the steps of the processes, methods, and algorithms described in various embodiments of the present invention need not be performed in the order described herein. Additionally, although some steps are described as being performed asynchronously, in other embodiments some of these steps may be performed concurrently. Further, the illustration of a process by depiction in the drawings does not mean that the illustrated process is exclusive of other changes and modifications thereto, and that any of the illustrated process or steps thereof may be one of various embodiments of the present invention. It does not imply that it is essential to one or more, and does not imply that the illustrated process is desirable.

As shown in FIG. 5, in step S510, information on the first region of interest is collected. For example, referring to FIGS. 1 to 4 , the processor 122 included in the electronic device 120 for recommending content for each individual, from a plurality of user terminals 110-1, ..., 110-n, first Information on an area of interest can be collected. According to one embodiment, the information on the first region of interest may include information on a subject searched for by the user a preset number of times (eg, 10 times, 50 times, 100 times, etc.) or more, but is not limited thereto. don't

In step S520, first recommended content information is formed. For example, referring to Figures 1 to 4, the processor 122 included in the electronic device 120 for recommending content for each person makes a first recommendation using the information on the first area of interest collected in step S510. Content information can be formed. According to an embodiment, the first recommendation content information includes topics (eg, real estate, games, travel, restaurants, etc.) that the user frequently searches using the user terminal (eg, 110-1). may contain information about

In step S530, information on the second region of interest is collected. For example, referring to FIGS. 1 to 4 , the processor 122 included in the electronic device 120 for recommending content for each individual, from a plurality of user terminals 110-1, ..., 110-n, second Information on an area of interest can be collected. According to an embodiment, the information on the second region of interest may include information on the user's call record, text record, and search record. For example, the information on the second region of interest may include information on topics that the user does not usually search for frequently using the user terminal 110 - 1 .

In step S540, second recommended content information is formed. For example, referring to Figures 1 through 4, the processor 122 included in the electronic device 120 for recommending content for each individual uses the information on the second area of interest collected in step S530 to make the second recommendation. Content information can be formed. According to an embodiment, the second recommended content information may include information on subjects that the user does not usually search for frequently using the user terminal (eg, 110-1). For example, the processor 122 uses the user's call records to extract that the other party is in their 40s, the area of residence is Daegu, and the area of interest is real estate, and uses the extracted information to extract the interest of the other party. Since the area is real estate, real estate may be extracted as information on the user's second area of interest, and content related to real estate may be formed as second recommended content information.

In step S550, user-customized content information is formed. For example, referring to Figures 1 through 4, the processor 122 included in the electronic device 120 for recommending content for each individual includes the first recommended content information formed in step S520 and the second recommended content formed in step S540. User-customized content information may be formed by mixing information at a preset ratio. For example, the processor 122 may form user-customized content information such that the ratio of the first recommended content information and the second recommended content information is 50:50.

In step S560, user-customized content information is transmitted to the user terminal. For example, referring to Figures 1 to 4, the processor 122 included in the electronic device 120 for recommending personalized content transmits the user-customized content information formed in step S550 to the user terminal through the network N ( 110-1).

Although the method has been described through specific embodiments, it is also possible to implement the method as computer readable code on a computer readable recording medium. A computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage devices. In addition, the computer-readable recording medium is distributed in computer systems connected through a network, so that computer-readable codes can be stored and executed in a distributed manner. In addition, "functional" programs, "codes, and code segments for implementing the embodiments can be easily inferred by programmers in the technical field to which the present invention belongs.

In the above, the present invention has been described in detail using preferred embodiments, but the scope of the present invention is not limited to specific embodiments, and should be interpreted according to the appended claims. In addition, those skilled in the art should understand that many modifications and variations are possible without departing from the scope of the present invention.

100: Personal content recommendation environment 110-1,...,110-n: User terminal
120: Electronic device for recommending personalized content 130: Database
121: receiver 122: processor
123: transmission unit 124: system bus
125: digital packet 126: input layer
127: neural network 128: output layer

Claims (10)

In an electronic device for recommending content for each individual,
one or more processors; and
one or more memories storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations;
The one or more processors,
Collect information about a user's first area of interest;
Forming first recommended content information according to the first region of interest;
Collect information about the user's second area of interest;
Forming second recommended content information according to the second region of interest;
Mixing the first recommended content information and the second recommended content information at a preset ratio to form user-customized content information;
Transmitting the user-customized content information to a user terminal,
An electronic device for personalized content recommendation.
According to claim 1,
The information on the first region of interest includes information on a subject searched by the user a preset number of times or more.
An electronic device for personalized content recommendation.
According to claim 1,
The information on the second region of interest includes information on the user's call record, text record, and search record.
An electronic device for personalized content recommendation.
According to claim 3,
Information about the user's call records,
including information about the age of the person on the other end of the call, where they live, and areas of interest;
An electronic device for personalized content recommendation.
According to claim 3,
Information about the user's text record,
including information about the person's age, place of residence, and areas of interest;
An electronic device for personalized content recommendation.
According to claim 1,
Information about the user's text record,
Extracting a text message containing a hyper-link among text messages received by the user terminal, and including information retrieved from a website linked by a hyperlink in the extracted text message,
An electronic device for personalized content recommendation.
According to claim 1,
The one or more processors,
Forming user-customized content information by adjusting a mixing ratio of the first recommended content information and the second recommended content information based on a user request;
An electronic device for personalized content recommendation.
one or more processors; and
A method for recommending content for individuals using an electronic device for recommending content for individuals including one or more memories storing instructions for causing the one or more processors to perform calculations when executed by the one or more processors,
collecting information about a user's first region of interest;
forming first recommended content information according to the first region of interest;
collecting information about a second region of interest of a user;
forming second recommended content information according to the second region of interest;
forming user-customized content information by mixing the first recommended content information and the second recommended content information at a preset ratio; and
Transmitting the user-customized content information to a user terminal
How to recommend personalized content.
According to claim 8,
receiving a user request; and
Further comprising forming user-customized content information by adjusting a mixing ratio of the first recommended content information and the second recommended content information based on the user request,
How to recommend personalized content.
one or more processors; And a computer program stored in a computer-readable recording medium so as to be executable by a computer including one or more memories in which instructions for causing the one or more processors to perform operations when executed by the one or more processors are stored,
collecting information about a user's first region of interest;
forming first recommended content information according to the first region of interest;
collecting information on a second region of interest of a user;
forming second recommended content information according to the second region of interest;
forming user-customized content information by mixing the first recommended content information and the second recommended content information at a preset ratio; and
A computer program stored in a computer-readable recording medium to perform the step of transmitting the user-customized content information to a user terminal.

Images