KR20220054050A - Apparatus and method for providing information recommendation service - Google Patents

Abstract

The present invention relates to an apparatus for providing an information recommendation service to provide personalized advertising services effectively. According to the present invention, the apparatus comprises a communication unit, a storage unit, and a processor operatively connected to the communication unit and the storage unit. The processor is configured to receive a service request signal activating a content-related application from a set-top box; receive data related to the content-related application currently being provided through the set-top box according to the service request signal; collect content analysis data related to a user from stored content analysis data on the basis of the received data; predict the current user's propensity by using a user analysis model learned to predict the user's propensity on the basis of the received data and the related content analysis data; and provide a service related to the predicted propensity to a display device connected to the set-top box.

Description

Device and method for providing information recommendation service {APPARATUS AND METHOD FOR PROVIDING INFORMATION RECOMMENDATION SERVICE}

The present invention relates to an apparatus and method for providing an information recommendation service.

In general, when watching a broadcast through TV or cable, advertisements are provided before and after a program and between programs, and these advertisements are uniformly provided to all viewers watching the same channel at the same time.

For example, advertisements are organized in the time zone between the first program (news) and the second program (drama) transmitted by broadcasting station A, and uniform and unilateral advertisements are transmitted according to the programming of the broadcasting station, regardless of which viewers watch it. this is done

In other words, all viewers watch the same advertisement as if they were watching a broadcast program. Considering the interest of various viewers, advertisement time passes useless without the effect of advertisement, and it is difficult to check the effect of the advertisement. .

The description underlying the invention has been prepared to facilitate understanding of the invention. It should not be construed as an admission that the matters described in the background technology of the invention exist as prior art.

Conventionally, a method for producing a customized advertisement has been proposed, but specifically, the method has not been proposed as to when and how to selectively provide the customized advertisement, so there is still a problem in that the personalized advertisement service cannot be effectively provided. are doing

Accordingly, there is a need for a technology capable of effectively providing a personalized advertisement service.

Accordingly, the inventors of the present invention have attempted to develop a method and apparatus capable of providing various services suitable for users by predicting a user's tendency to currently view content through a set-top box.

As a result, the inventors of the present invention activate a content-related application, collect data obtained by a user's manipulation on the application and content analysis data related thereto, to develop a method and apparatus capable of accurately predicting a user's tendency. reached

Moreover, the inventors of the present invention sought to maximize the effect of providing a service by acquiring visual and auditory elements that can be identified from the currently provided content, and matching the acquired data with data collected by a user's behavior.

The problems of the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

In order to solve the above problems, an apparatus for providing an information recommendation service according to an embodiment of the present invention is provided. The device includes a communication unit, a storage unit, and a processor operatively connected to the communication unit and the storage unit, wherein the processor receives a service request signal for activating a content-related application from a set-top box, and responds to the service request signal. Accordingly, data associated with a content-related application currently provided through the set-top box is received, and content analysis data associated with a user among pre-stored content analysis data is collected based on the received data, and the received data and the associated content are analyzed. It is configured to predict the current user's disposition using a user analysis model trained to predict the user's disposition based on data, and to provide a service related to the predicted disposition to a display device connected to the set-top box.

According to a feature of the present invention, the data is received through a controller linked to the set-top box, and the processor is configured to receive at least one of navigation data on the content-related application or output data of the display device. can be configured.

According to another feature of the present invention, the navigation data may include at least one of left and right, vertical movement, residence time, and selection data.

According to another feature of the present invention, the processor may be configured to acquire a frame or an image at a time point at which the data is received within the content currently being provided through the set-top box.

According to another feature of the present invention, the processor may be configured to collect at least one of a person, an object, a place, and a character in the frame as the content analysis data.

According to another feature of the present invention, when acquiring an image, the processor converts the dialogue or narration of a person in the image into text, extracts a word to be defined as an object of interest from the converted text, and extracts the may be configured to collect words into the content analysis data.

According to another feature of the present invention, the processor may be configured to receive selection information about one person through the communication unit when there are two or more persons in the currently provided content.

According to another feature of the present invention, the processor may be configured to provide any one service among an additional information providing service, a product information providing service, and an advertisement providing service on the content-related application.

According to another feature of the present invention, in the content-related application, a graphic screen representing the one service may be provided in the form of a pop-up window.

According to another feature of the present invention, the processor may be configured to identify any one of the users matched with the set-top box based on the data received through the one of the services.

In order to solve the above problems, there is provided a method for providing an information recommendation service according to another embodiment of the present invention. The method includes the steps of: receiving a service request signal for activating a content-related application from a set-top box; receiving data related to a content-related application currently being provided through the set-top box according to the service request signal; Collecting content analysis data associated with the user from among previously stored content analysis data based on the content analysis data, the current user's tendency using a user analysis model learned to predict the user's tendency based on the received data and the related content analysis data and providing a service associated with the predicted tendency to a display device connected to the set-top box.

According to a feature of the present invention, the data is received through a controller linked to the set-top box, and the receiving of the data includes at least one of navigation data on the content-related application or output data of the display device. It may be a step of receiving data.

According to another feature of the present invention, the navigation data may include at least one of left and right, vertical movement, residence time, and selection data.

According to another feature of the present invention, the collecting of the content analysis data may further include acquiring a frame or an image at a time point at which the data is received within the content currently being provided through the set-top box. .

According to another feature of the present invention, in the collecting of the content analysis data, when a frame is obtained in the obtaining step, at least one target of a person, an object, a place, and a character in the frame is analyzed for the content It may be a step of collecting data.

According to another feature of the present invention, in the collecting of the content analysis data, when an image is obtained in the obtaining step, a dialogue or narration of a person in the image is converted into text, and the converted text is converted into a text It may be a step of collecting content analysis data.

According to another feature of the present invention, the step of extracting the word may further include receiving selection information about any one person when there are two or more persons in the currently provided content.

According to another feature of the present invention, the step of providing the service related to the predicted tendency is a step of providing any one service among an additional information providing service, a product information providing service, and an advertisement providing service on the content-related application. can

According to another feature of the present invention, in the content-related application, a graphic screen representing the one service may be provided in the form of a pop-up window.

According to another feature of the present invention, after providing the service related to the predicted tendency, any one of the users matched with the set-top box is identified based on the data received through the one of the services It may further include the step of

Specific details of other embodiments are included in the detailed description and drawings.

According to the present invention, by collecting data related to a user watching content and content analysis data in a condition in which a content-related application is activated, it is possible to predict a user's tendency. Accordingly, the user who is watching the content may be provided with a service suitable for his/her preference, and the service provider may maximize the effect of providing the service.

For example, the present invention sequentially recognizes data acquired by a user's manipulation on a content-related application, and grasps the user's flow of thought, preference for a specific item, etc. based on the finally selected data, so that the user's tendency can be accurately predicted.

The effect according to the present invention is not limited by the contents exemplified above, and more various effects are included in the present invention.

1 is a schematic diagram of an information recommendation service system according to an embodiment of the present invention.
2 is a schematic diagram illustrating a configuration for collecting data in an information recommendation service system according to an embodiment of the present invention.
3 is a schematic diagram for explaining an outline of a method for providing an information recommendation service according to an embodiment of the present invention.
4 is a block diagram showing the configuration of an information recommendation service providing apparatus according to an embodiment of the present invention.
5 is a schematic flowchart of a method for providing an information recommendation service according to an embodiment of the present invention.
6 to 9 are schematic diagrams for explaining an interface through which an information recommendation service is provided according to an embodiment of the present invention.
10A to 10C are schematic diagrams for exemplarily explaining an interface through which an information recommendation service is activated according to an embodiment of the present invention.

Advantages and features of the present invention, and a method for achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in a variety of different forms, only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the technical field to which the present invention pertains It is provided to fully inform the possessor of the scope of the invention, and the present invention is only defined by the scope of the claims. In connection with the description of the drawings, like reference numerals may be used for like components.

In this document, expressions such as "have," "may have," "includes," or "may include" refer to the presence of a corresponding characteristic (eg, a numerical value, function, operation, or component such as a part). and does not exclude the presence of additional features.

In this document, expressions such as "A or B," "at least one of A or/and B," or "one or more of A or/and B" may include all possible combinations of the items listed together. . For example, "A or B," "at least one of A and B," or "at least one of A or B" means (1) includes at least one A, (2) includes at least one B; Or (3) it may refer to all cases including both at least one A and at least one B.

As used herein, expressions such as "first," "second," "first," or "second," may modify various elements, regardless of order and/or importance, and refer to one element. It is used only to distinguish it from other components, and does not limit the components. For example, the first user equipment and the second user equipment may represent different user equipment regardless of order or importance. For example, without departing from the scope of the rights described in this document, the first component may be named as the second component, and similarly, the second component may also be renamed as the first component.

A component (eg, a first component) is "coupled with/to (operatively or communicatively)" to another component (eg, a second component); When referring to "connected to", it will be understood that the certain element may be directly connected to the other element or may be connected through another element (eg, a third element). On the other hand, when it is said that a component (eg, a first component) is "directly connected" or "directly connected" to another component (eg, a second component), the component and the It may be understood that other components (eg, a third component) do not exist between other components.

The expression "configured to (or configured to)" as used in this document, depending on the context, for example, "suitable for," "having the capacity to ," "designed to," "adapted to," "made to," or "capable of." The term “configured (or configured to)” may not necessarily mean only “specifically designed to” in hardware. Instead, in some circumstances, the expression “a device configured to” may mean that the device is “capable of” with other devices or parts. For example, the phrase "a processor configured (or configured to perform) A, B, and C" refers to a dedicated processor (eg, an embedded processor) for performing the corresponding operations, or by executing one or more software programs stored in a memory device. , may mean a generic-purpose processor (eg, a CPU or an application processor) capable of performing corresponding operations.

Terms used in this document are only used to describe specific embodiments, and may not be intended to limit the scope of other embodiments. The singular expression may include the plural expression unless the context clearly dictates otherwise. Terms used herein, including technical or scientific terms, may have the same meanings as commonly understood by one of ordinary skill in the art described in this document. Among terms used in this document, terms defined in a general dictionary may be interpreted with the same or similar meaning as the meaning in the context of the related art, and unless explicitly defined in this document, ideal or excessively formal meanings is not interpreted as In some cases, even terms defined in this document cannot be construed to exclude embodiments of this document.

Each feature of the various embodiments of the present invention may be partially or wholly combined or combined with each other, and as those skilled in the art will fully understand, technically various interlocking and driving are possible, and each embodiment may be implemented independently of each other. It may be possible to implement together in a related relationship.

For clarity of interpretation of the present specification, terms used herein will be defined below.

As used herein, the term “content-related application” may refer to an application that is installed and executable on a device that enables content to be reproduced on a user's display device. Through the execution of such an application, an image may be reproduced or an image may be output in the form of a separate pop-up window on the content, and new content may be reproduced according to a user's manipulation.

As used herein, the term “user analysis model” may be a model trained to predict a user's propensity based on data received by a user's manipulation and content analysis data.

Specifically, the user analysis model can derive one output data based on two types of input data. More specifically, data received by a user's manipulation, which is one of the input data (user data), is displayed on the content-related application. It means data on the top and bottom, left and right, search and selection, and the other input data, the content analysis data, is the data about the content being played at the same time as the user data was acquired (e.g. content title, outline (synopsis) ), genre, character, background, place, lines, fingerprints, subtitles, etc.). In addition, the user's tendency defined by the output data may mean an item that the user is interested in for each content type. For example, in the case of a sports broadcasting channel, scores, player performance, rankings, commentators, etc. may mean user tendencies, and in the case of dramas and movies, the person's filmography, clothes, products possessed or equipped by the person, and shooting location. and the like may mean user tendencies. That is, items that can be visually and aurally within one content and a specific search term that the user searches for in relation to the content may mean the user's tendency. In addition, the user's tendency may include two or more according to a time point at which user data is received. That is, since the user may repeat movement and selection and then movement and selection while viewing one or more contents, the user analysis model may predict one or more user propensities for each content based on this.

ve Bayes), LNB (locally weighted na

ve Bay), 및 SVN (support vector machine) 중 적어도 하나일 수 있으며, 이는 기계학습 알고리즘, 신경망 알고리즘(DNN, CNN, DCNN, RNN, RBM, DBN, SSD)), 유전자 알고리즘, 딥러닝 알고리즘 중 적어도 하나의 인공지능 알고리즘을 이용하여 학습될 수 있다. According to an embodiment, the user analysis model is a random forest (random forest), GNB (Gaussian na)

ve Bayes), LNB (locally weighted na

ve Bay), and at least one of a support vector machine (SVN), which is at least one of a machine learning algorithm, a neural network algorithm (DNN, CNN, DCNN, RNN, RBM, DBN, SSD)), a genetic algorithm, and a deep learning algorithm. It can be learned using a single artificial intelligence algorithm.

In addition, the user tendency prediction result output by the user analysis model may be used to provide information that the user may need or may be interested in on the content the user is viewing.

That is, as the user propensity is predicted based on the user analysis model, the present invention can provide a suitable service to the current user, thereby increasing the service provision satisfaction and enhancing the advertisement according to the service provision and the promotional effect thereof. .

1 is a schematic diagram of an information recommendation service system according to an embodiment of the present invention, FIG. 2 is a schematic diagram for explaining a configuration for collecting data in an information recommendation service system according to an embodiment of the present invention, FIG. 3 is a schematic diagram for explaining an outline of a method for providing an information recommendation service according to an embodiment of the present invention.

Referring to FIG. 1 , the information recommendation service system 1000 includes a plurality of set-top boxes 10 and an information recommendation service providing device 100 that provides an information recommendation service to the set-top boxes 10 (hereinafter, “service providing device ( 100)”) and a content providing server 200 that provides information on content that can be provided to the set-top box 10 .

The set-top box 10 may deliver an information recommendation service to a user. Specifically, the set-top box 10 may receive data (user data) obtained by a user's operation, and transmit it to the service providing apparatus 100 , and the current set-top box 10 from the service providing apparatus 100 . ) can be provided to the user by receiving an information recommendation service suitable for the user.

Referring to FIG. 2 , the set-top box 10 is connected to the controller 20 and the display device 30 , and receives data (user data) by a user's manipulation from the controller 20 , and the display device 30 . information recommendation service can be delivered.

According to an embodiment, the controller 20 may be an electronic device equipped with at least one physical button and a touch button, or an electronic device equipped with an output device (not shown). Here, a virtual button selectable by the user may be output through the output device. That is, the controller 20 may be a portable electronic device such as a smart phone or a tablet PC in addition to the commonly known electronic device equipped with a plurality of buttons.

According to an embodiment, the display device 30 may be connected to the set-top box 10 to output an interface for an information recommendation service provided by the set-top box 10 . Specifically, the display device 30 may output a screen in which a service suitable for the user's preference is overlaid on the content, and an additional service screen may be overlaid on the content or new content according to the user's selection within the output screen. can be output.

In addition, the interface screen provided through the display device 30 may be provided when a content-related application is activated according to a user's request, and the set-top box 10 may have a content-related application installed and executed.

The service providing apparatus 100 is a server that provides various services, such as a product purchase service, an information provision service, and an advertisement provision service, to a user, and views content for each of a plurality of set-top boxes 10 of the user through the set-top box 10 . It is possible to predict the propensity of a current user and provide a service suitable for the current user's propensity. To this end, the service providing apparatus 100 may generate and distribute a content-related application that can be installed in the set-top box 10 , and may learn a user analysis model for predicting a user's tendency.

The content providing server 200 is a server that provides various contents to the set-top box 10 , and may include a broadcasting station server 210 , a content distribution server 220 , and an OTT service providing server 230 . For example, the broadcasting station server 210 may be a server operated by a terrestrial, airwave, or cable broadcasting station, and the content distribution server 220 produces content such as a mobile network operator (MNO) and a mobile virtual network operator (MVNO). And it may be a server operated by a mobile communication company that sells it, and the OTT service providing server 230 may be a server operated by a video sharing company such as YouTube or a server operated by a content service producer such as Netflix. there is.

According to an embodiment, the content providing server 200 may provide the content according to the user's selection to the set-top box 10 , and in response to the request of the service providing device 100 , the user may view it through the set-top box 10 . You can provide data about the content being played. For example, data on content includes the title, outline (synopsis), genre, character, background, place, It may include lines, fingerprints, subtitles, etc., which may be stored in the service providing device 100 .

Meanwhile, the content providing server 200 may further include an advertisement content providing server (not shown). The advertisement content providing server may be a server operated by an advertiser that produces/sells advertisements, and the service providing device 100 receives advertisements according to the user's disposition from the advertisement contents providing server, and receives an advertisement according to the user's disposition from the advertisement contents providing server, and according to the predicted disposition, a set-top box ( 10) can provide advertisements to users.

In addition, the broadcasting station server 210 , the content distribution server 220 , and the OTT service providing server 230 may provide recommendable advertisements and content information to the service providing apparatus 100 according to a user's preference. For example, when it is predicted that the user is interested in the filmography of actor A, the service providing device 100 provides the user with advertisement information/content information among various channels in which the actor's work is being played. may recommend moving.

In general, a user who watches content through TV or IPTV may view advertisements through a separate home shopping channel or directly search for desired information using the display device 30 . Recently, for user convenience, a technology for recommending information to a user by learning all information collected through a set-top box 10 connected to a display device 30 such as a TV or IPTV has been disclosed. Unless you specifically enter your personal information, there is a limit to recommending suitable information to the user.

For example, in order to provide customized information according to user information and content viewing history, the user's personal information and further personal information of family members must be input at the initial stage of installing the set-top box 10, It can be inferred whether a family member is watching content, but depending on the situation, it cannot be said to be an accurate inference.

Accordingly, in an embodiment of the present invention, the service providing apparatus 100 can predict the user's tendency through the data (user data) obtained by the current user's manipulation, and accordingly provide a customized service. there is. In other words, instead of inferring the current user based on the content viewing history collected over a day or a week through the set-top box 10, the user's propensity is predicted based on the data collected in real time, and a service that matches the predicted propensity is provided. can provide

Referring to FIG. 3 , the service providing apparatus 100 receives data (user data) obtained by a user's manipulation from the set-top box 10 , and receives content analysis data from the content providing server 200 . It is possible to predict user propensity and provide information recommendation service through the user analysis model.

According to an embodiment, the content analysis data includes the title, outline (synopsis), genre, person, background, place, etc. of the content, such as persons, objects, places, characters, lines, narrations, etc. in the screen/video of the content currently being viewed by the user. may be included, and according to an embodiment, the data (user data) obtained by the user's manipulation may include navigation data and output environment change data collected through the controller 20 . For example, navigation data means data on how a user's operation is sequentially changed, such as left, right, up and down, selection, and stay while content is being played, and output environment change data is data about audio volume or channel adjustment means

The service providing apparatus 100 puts two pieces of information into the user analysis model, can predict the current user's propensity, and provides various information recommendation services according to the predicted propensity and data (user data) obtained by subsequent user manipulation. may be provided to the corresponding set-top box 10 . For example, the information recommendation service may include an additional information providing service, a product information providing service, an advertisement providing service, etc. based on a user's tendency.

So far, the information recommendation service system 1000 according to an embodiment of the present invention has been described. According to the present invention, the service providing apparatus 100 can provide a service suitable for predicting a user's tendency in real time even if it does not store a broadcast viewing history for each user, user personal information, and the like.

Hereinafter, the service providing apparatus 100 for predicting a user's tendency and providing a service will be described in detail.

4 is a block diagram showing the configuration of an information recommendation service providing apparatus according to an embodiment of the present invention.

Referring to FIG. 4 , the service providing apparatus 100 may include a communication unit 110 , a storage unit 120 , and a processor 130 .

The communication unit 110 may exchange data with the set-top box 10 and the content providing server 200 through a wired/wireless network. For example, the communication unit 110 may receive data (user data) obtained by a user's manipulation from the set-top box 10 , and may receive content data/content analysis data from the content providing server 200 . For example, the communication unit 110 receives daily, weekly, monthly channel schedules and content (program) data from the broadcasting station server 210 or periodically from the content distribution server 220 and the OTT service providing server 230 . to receive newly provided content data.

The storage unit 120 may store various data to be used in the content user analysis model. Specifically, the storage unit 120 may store a channel schedule in units of days, weeks, and months, and may store content by genre. For example, genres include Action/Adventure, Animation, Children/Family, Classic Movie, Comedy, Documentary, Drama, Foreign Movie, Horror Movie, Music Movie, Romance Movie, Sci-Fi Fantasy Movie, Sports, Thriller, TV Show, etc. may be included, and may be classified into more specific items within a genre. For example, within music films, Latin music, children's music, country/folk, rock/pop concerts, world music concerts, jazz/easy listening, urban/dance concerts, musicals, classic musicals, Disney musicals, stage musicals, show business musicals, etc. can be classified and stored as

According to an embodiment, the storage unit 120 may store the content data analyzed by the processor 130 . Specifically, the storage unit 120 may store image data, audio data, and text data of an object (person, animal, thing) in an image/screen extracted by the processor 130 , and is divided by the processor 130 . It is also possible to store the location data in the contents. Here, the image/screen may be an image or a screen split based on a time preset by the processor 130 in content having audiovisual elements, and the storage unit 120 periodically stores the content that is newly organized or supplied. Data extracted by the processor 130 may be stored.

That is, the storage unit 120 stores genres, titles, performers, outlines, characters, places, sales product information, issue headlines, genres by VOD, titles, outlines, performers, places, backgrounds, Content title, outline, person, text, place, and background information in the OTT service can be stored.

Also, the storage unit 120 may store data related to the set-top box 10 that will provide the information recommendation service. For example, the storage unit 120 may store advertisement data to be provided to the set-top box 10 and data (user data) obtained by a user's manipulation according to the advertisement provision.

In addition, the storage unit 120 may store the set-top box 10 data, the user personal information collected for each set-top box 10, the user viewing history, etc. It can be used when data (user data) is not collected by

In various embodiments, the storage unit 120 may include a volatile or nonvolatile recording medium capable of storing various data, commands, and information. For example, the storage unit 120 is a flash memory type, hard disk type, multimedia card micro type, card type memory (SD, XD memory, etc.), RAM, SRAM, ROM, EEPROM, PROM, network storage storage, cloud , may include at least one type of storage medium among the block chain database.

Also, the storage unit 120 may record commands for the operation of the service providing apparatus 100 . In various embodiments, an application (not shown) for predicting a user's tendency and providing an information recommendation service according to the predicted tendency may be recorded in the storage unit 120 .

The processor 130 communicates It is operably connected to the unit 110 and the storage unit 120 to control the overall operation of the service providing apparatus 100 , and to drive an application or program stored in the storage unit 120 to provide an information recommendation service. Various commands to provide can be executed.

The processor 130 may correspond to a computing device such as a central processing unit (CPU) or an application processor (AP). In addition, the processor 130 may be implemented in the form of an integrated chip (IC), such as a system on chip (SoC) in which various computing devices are integrated.

According to an embodiment, the processor 130 receives a service request signal for activating a content-related application from the set-top box 10 through the communication unit 110 , and is currently provided through the corresponding set-top box 10 according to the service request signal. It is possible to receive data related to a content-related application in progress.

More specifically, in order for the processor 130 to provide an information recommendation service suitable for a user in real time, a content-related application should be basically activated. Accordingly, the processor 130 may switch the mode of the set-top box 10 to the data collection mode for providing information recommendation service so as to collect data (user data) by the user's manipulation as the content-related application is activated. there is.

According to an embodiment, the processor 130 may add a guide line or a layout for a unit to the content provided to the display device 30 as a content-related application is activated. That is, the processor 130 may set a guide line or unit for collecting navigation data based on the structural data of different images according to the properties of each content, and the guide line or unit is displayed on the display device 30 according to the user's selection. ) can be output. Here, the user's selection may be collected through the controller 20 connected to the set-top box 10 .

For example, in the case of a sports relay channel or program, a graphic object representing a game situation, such as a normal score, is output on the upper left side of the display device 30 , and a panel promoting sponsors is output on the border area of the display device 30 . As a result, the processor 130 may add a guide line or a unit for an item to be selected by the user on the content.

For another example, in the case of a TV show program, typically, a subject or performer information for content is output to a corner area of the display device 30 , and the processor 130 provides guidelines for an item to be selected by the user on the content. Or you can add units.

After the content-related application is activated, the data associated with the content-related application received by the service providing device 100 through the set-top box 10 includes navigation data on the content-related application or output data of the display device 30 . may include Such data may be utilized as data for predicting a user's propensity for content currently being reproduced.

More specifically, the navigation data may mean a user's manipulation sequence on a guide line or unit designated for each content. For example, the navigation data may be a mixture of data moving left and right, moving up and down on a guide line or unit, data regarding a time of moving or staying for a preset time, and data selecting a specific unit. In other words, the navigation data may include information on which flow the user moves and which object is selected in the guide line or unit provided by the processor 130 .

In addition, if additional information related to the content currently being reproduced is provided by the processor 130 , the processor 130 may receive navigation data on the graphic object indicating the additional information through the communication unit 110 .

After receiving the data (user data) by the user's manipulation, the processor 130 may collect analysis data related to the user from among pre-stored content analysis data based on the received data. More specifically, in order to provide the information recommendation service, the processor 130 should check what kind of content was being reproduced when the user's reaction appeared, that is, when the data was received. Accordingly, the processor 130 may acquire a frame or an image at a time when data in the currently provided content is received through the set-top box 10 . For example, if data is received through the set-top box 10 at 8:32 pm, the processor 130 displays an image frame at 8:32 pm of the content being reproduced through the corresponding set-top box 10 . Alternatively, an image between 8:31 and 33 pm may be acquired.

According to an embodiment, when acquiring a frame of content, the processor 130 may collect at least one of a person, an object, a place, and a character in the corresponding frame as content analysis data. That is, the processor 130 primarily collects data related to the content being viewed in the corresponding set-top box 10 from among the content analysis data pre-stored in the storage unit 120 , and among them, data related to a specific frame as content analysis data can be collected

According to another embodiment, when acquiring an image of content, the processor 130 may convert a person's lines or narration in the image into text, and collect the converted text as content analysis data. In addition, the processor 130 may extract a word to be defined as an object of interest from the converted text, and may collect the extracted word as content analysis data. In addition, the content analysis data defined as the object of interest may be given a weight when the user analysis model is subsequently applied.

Meanwhile, the processor 130 may convert the lines of all persons in the image into text, but may also convert only lines of a specific person in which the user is interested into text. Accordingly, when there are two or more persons in the currently reproduced content, the processor 130 may receive selection information about one person through the communication unit 110 before converting the audio into text.

The processor 130 may predict the tendency of the current user by using the user analysis model learned to predict the tendency of the user based on the data received through the set-top box 10 and the content analysis data associated with the user. Here, the user's tendency may mean an item that the user is interested in for each content type. For example, in the case of a sports broadcasting channel, interest in scores, player performance, rankings, commentators, etc. can be predicted by the user's tendency, and in the case of a drama, the person's filmography, clothes, and goods possessed/equipped by the person , a photographing location, etc. may be predicted by the user's tendency.

In this way, the processor 130 may predict the current user's propensity by using the learned user analysis model, and may provide a service related to the predicted propensity to the display device 30 connected to the set-top box 10 . According to an embodiment, the processor 130 may provide an additional information providing service, a product information providing service, and an advertisement providing service on a content-related application. Here, the additional information means more specific information related to the content analysis data (eg, person, text), and the product information is information on the URL or purchase place to purchase the product related to the content analysis data (eg, the person's clothes) can mean For example, the processor 130 may additionally provide quarterly results of the player if the user's tendency is a sports relay channel - player performance, and if the drama - a product possessed/equipped by a person, the lowest price purchase url of the product can provide

In addition, a graphic screen representing such a service may be provided in the form of a pop-up window on the screen of the display device 30 on which content is being reproduced. In addition, the service screen may be floated on the edge or corner of the display screen 30 according to the user's selection, and accordingly, the user's content appreciation may not be disturbed.

According to an embodiment, after the processor 130 provides one service based on data divided by one operation of the user, data associated with a graphic screen indicating the service may be collected, and based on the collected data Any one user matched with the set-top box 10 may be identified.

That is, as the content-related application is activated, the processor 130 may check the user information previously stored in the storage unit 120 for each set-top box 10, and the current user through interaction with the graphic screen It is possible to identify which user among the users stored in (10). Through this identification process, the processor 130 may collect data for learning the user analysis model, and may more quickly and accurately predict the user's tendency.

So far, the service providing apparatus 100 according to an embodiment of the present invention has been described. According to the present invention, even if the service providing apparatus 100 does not have the user's personal information in advance, it is possible to predict the user's tendency and provide a service that matches the predicted tendency.

Hereinafter, a method of providing a service suitable for a user's preference through the service providing apparatus 100 will be described.

5 is a schematic flowchart of a method for providing an information recommendation service according to an embodiment of the present invention.

Referring to FIG. 5 , the service providing apparatus 100 receives a service request signal for activating a content-related application from the set-top box 10 ( S110 ). That is, the service providing apparatus 100 may provide a service only when the set-top box 10 receives a user's request without indiscriminately collecting all information received through the controller 20 . When the service providing device 100 receives such a signal, the set-top box 10 sets the mode of the set-top box 10 to collect data (user data) by the user's manipulation of the data collection for information recommendation service. mode can be switched.

After step S110, the service providing apparatus 100 receives data related to a content-related application currently being provided through the set-top box 10 according to the service request signal (S120).

According to an embodiment, the service providing apparatus 100 may add a guide line or a layout for a unit to the content provided to the display apparatus 30 according to the service request signal. That is, the service providing apparatus 100 may set a guide line or unit for collecting navigation data based on structural data of different images according to the properties of each content, and the guide line or unit is displayed according to the user's selection. (30) can be output. Here, the user's selection may be collected through the controller 20 connected to the set-top box 10 .

For example, in the case of a movie/drama provided in the VOD method, the characters of the work are usually output to the middle area and the background is output to the remaining area, and the service providing apparatus 100 provides the content within the outputable frame. A guideline or unit for an item to be selected by the user on the content can be added in the form of “▣”.

According to an embodiment, the data associated with the content-related application may include navigation data on the content-related application or output data of the display device 30 . Such data may be utilized as data for predicting a user's propensity for content currently being reproduced.

More specifically, the navigation data may mean a user's manipulation sequence on a guide line or unit designated for each content. For example, the navigation data may be a mixture of data moving left and right, moving up and down on a guide line or a unit, data about a moving or staying time for a preset time, and data selecting a specific unit. In other words, the navigation data may include information on which flow the user moves and which object is selected in the guide line or unit provided by the processor 130 .

In addition, if additional information related to the content currently being reproduced is provided by the processor 130 , the service providing device 100 may receive navigation data on a graphic object indicating additional information through the set-top box 10 . there is.

After the step S120, the service providing apparatus 100 receives the content analysis data associated with the user among the previously stored content analysis data for the received data (S130). Specifically, it is necessary to check what kind of content was being played when the user's reaction appeared, that is, when the data was received. Accordingly, the service providing apparatus 100 may acquire a frame or an image at a time when data in the currently provided content is received through the set-top box 10 . For example, if the time at which data is received through the set-top box 10 is 8:32 pm, the service providing apparatus 100 sets the time at 8:32 pm of the content being reproduced through the corresponding set-top box 10 . You can acquire an image frame or an image between 8:31 and 33 pm.

According to an embodiment, when acquiring a frame of content, the service providing apparatus 100 may collect at least one of a person, an object, a place, and a character in the corresponding frame as content analysis data. That is, the service providing apparatus 100 may first collect data related to content being viewed in the corresponding set-top box 10 from among pre-stored content analysis data, and among them, data related to a specific frame may be collected as content analysis data. .

According to another embodiment, when acquiring an image of content, the service providing apparatus 100 may convert a person's lines or narration in the image into text, and collect the converted text as content analysis data. In addition, the service providing apparatus 100 may extract a word to be defined as an object of interest from the converted text, and may collect the extracted word as content analysis data. In addition, the content analysis data defined as the object of interest may be given a weight when the user analysis model is subsequently applied.

Meanwhile, the service providing apparatus 100 may convert the lines of all persons in the image into text, but may also convert only lines of a specific person in which the user is interested into text. Accordingly, when there are two or more persons in the currently reproduced content, the service providing apparatus 100 may receive selection information about any one person through the set-top box 10 before converting audio into text. .

After step S130, the service providing apparatus 100 predicts the tendency of the current user by using the user analysis model learned to predict the tendency of the user based on the received data and content analysis data associated with the user (S140). Specifically, the user analysis model may be a model trained to predict a user's tendency based on data (user data) and content analysis data received by the user's manipulation.

Specifically, the user analysis model can derive one output data based on two types of input data. More specifically, data received by a user's manipulation, which is one of the input data (user data), is displayed on the content-related application. It means data on the top and bottom, left and right, search and selection, and the other input data, the content analysis data, is the data about the content being played at the same time as the user data was acquired (e.g. content title, outline (synopsis) ), genre, character, background, place, lines, fingerprints, subtitles, etc.). In addition, the user's tendency defined by the output data may mean an item that the user is interested in for each content type. For example, in the case of a sports broadcasting channel, scores, player performance, ranking, commentators, etc. may mean user tendencies, and in the case of a movie, a person's filmography, clothes, goods possessed/equipped by the person, shooting location, etc. It may mean user preference. That is, items that can be visually or aurally identified within one content and a specific search term that the user searches for in relation to the content may mean the user's disposition.

ve Bayes), LNB (locally weighted na

ve Bay), 및 SVN (support vector machine) 중 적어도 하나일 수 있으며, 이는 기계학습 알고리즘, 신경망 알고리즘(DNN, CNN, DCNN, RNN, RBM, DBN, SSD)), 유전자 알고리즘, 딥러닝 알고리즘 중 적어도 하나의 인공지능 알고리즘을 이용하여 학습될 수 있다.예를 들어, 스포츠 중계 채널의 경우, 스코어, 선수 경기력, 순위 등에 대한 관심 표시가 사용자의 성향으로 예측될 수 있으며, 드라마의 경우, 인물의 필모그래피, 의상, 인물이 소지/장착한 상품, 촬영 장소 등에 대한 관심 표시가 사용자의 성향으로 예측될 수 있다. According to an embodiment, the user analysis model is a random forest (random forest), GNB (Gaussian na)

ve Bayes), LNB (locally weighted na

ve Bay), and at least one of a support vector machine (SVN), which is at least one of a machine learning algorithm, a neural network algorithm (DNN, CNN, DCNN, RNN, RBM, DBN, SSD)), a genetic algorithm, and a deep learning algorithm. It can be learned using a single artificial intelligence algorithm. For example, in the case of a sports relay channel, the display of interest in scores, player performance, ranking, etc. can be predicted by the user's tendency, and in the case of a drama, the filmography of a person , clothes, goods possessed/equipped by a person, and interest in a shooting location may be predicted by the user's tendency.

After step S140 , the service providing apparatus 100 provides a service associated with the predicted tendency to the display device 30 connected to the set-top box 10 ( S150 ). According to an embodiment, the service providing apparatus 100 may provide an additional information providing service, a product information providing service, and an advertisement providing service on a content related application. Here, the additional information means more specific information related to the content analysis data (eg, person, text), and the product information is information on the URL or purchase place to purchase the product related to the content analysis data (eg, the person's clothes) can mean For example, if the user's tendency is a sports relay channel - player performance, the service providing device 100 may additionally provide quarterly results of the player, and if the drama - a product possessed/equipped by a person, the lowest price of the product You can provide a purchase url.

In addition, a graphic screen representing such a service may be provided in the form of a pop-up window on the screen of the display device 30 on which content is being reproduced. In addition, the service screen may be floated on the edge or corner of the display screen 30 according to the user's selection, thereby eliminating the inconvenience of using the service without interfering with the user's content appreciation.

According to an embodiment, after the service providing apparatus 100 provides one service based on data divided by one operation of the user, data collected through a graphic screen indicating the service through the set-top box 10 may be received, and any one user matched with the set-top box 10 may be identified based on the received data.

The method for providing a service according to an embodiment of the present invention has been described so far, and an interface screen for providing a service will be described with reference to FIGS. 6 to 9 below.

6 to 9 are schematic diagrams for explaining an interface through which an information recommendation service is provided according to an embodiment of the present invention.

Referring to FIG. 6 , when the data collection mode for providing the information recommendation service is activated while the sports relay screen is played on the display device 30 connected to the set-top box 10, the service providing device 100 is the user's Depending on the operation, a guide line or a layout for a unit may be displayed with a dotted line on the content. The guideline or unit herein may be utilized to collect navigation data on a content-related application or output data of the display device 30 .

For example, in the case of a baseball relay channel screen, the graphic object 61 representing the score is set as one unit in the upper left, and the panel image 62 promoting the sponsor is set as one unit in the central area, Areas in which the pitcher, batter, catcher, and referee are located may be set as one unit, and the aforementioned unit may be navigation data.

In addition, the graphic object 65 representing the audio volume or channel switching of the display device 30 may be set as one unit in one area, and this may be output data.

According to an embodiment, when the user moves left, right, or up and down to select any one object, the dotted line shown in FIG. 6 may be displayed to inform that data has been collected.

Referring to FIG. 7 , when a user selects any one designated button designated through the controller 20 within a screen being viewed, the mode is switched to a data collection mode for providing an information recommendation service, and the service providing device 100 provides navigation data can be collected. If the graphic object representing the score is selected (①), the service providing apparatus 100 may provide the additional information 71 related to the score on the content.

Thereafter, the service providing apparatus 10 may collect navigation data on the graphic object. Specifically, the service providing device 10 controls the operation sequence on the graphic object, moving from the Doosan vs Hanwha score unit to the Lotte vs LG score (②), selecting the Lotte vs LG score (③), and from the match result unit to the game video unit. The movement to (④) and the movement from the game video unit to the cheering SNS unit (⑤) can be collected as navigation data.

In the case of the drama channel screen shown in FIG. 8 , when the user selects one person, the selected unit 81 may be temporarily displayed as shown by a dotted line. The service providing device 100 provides information 82 about a work in which a person appears based on data (eg, a character) collected by the user's manipulation, and specific information 83 about any one of them. It can be provided on the content.

Next, in the case of the OTT service screen as shown in FIG. 9 , the service providing device 100 based on the object information 91 reproduced in real time according to the search keyword 92 and the previous viewing history in the OTT service server (not shown). and then the video information 93 may be collected as content analysis data.

Meanwhile, when two types of input data input to the user analysis model are obtained, the service providing apparatus 100 may provide various types of services, which will be described below with reference to FIG. 10 .

10A to 10C are schematic diagrams for exemplarily explaining an interface through which an information recommendation service is activated according to an embodiment of the present invention.

Referring to FIG. 10A , when it is predicted that the user's tendency is a product possessed or worn by the person, the service providing device 100 displays "Shoes worn by actor A - home shopping broadcast" in one area of the display device 30 . A service window 101 for channel switching may be provided, saying that it will start after 1 minute--If you want to move to the corresponding channel, press the OK button.”

In addition, referring to 10b, the service providing device 100 may provide the accommodation reservation site connection window 102 for visiting the place, such as “Try to book Gyeongwonjae Ambassador Incheon,” in relation to the place in the drama. there is. Here, in the case of a lodging reservation site, the service providing apparatus 100 may be information previously provided from an advertisement content providing server (not shown).

Finally, referring to 10c, the service providing device 100 is a display device 30 using the OTT service, saying, “The Doosan vs Kiwoom JunPO match will start in a while - If you want to move to the broadcast broadcast by the commentator, Please press the move button”, a service window 103 for moving to a broadcasting channel of a club in which the user is interested may be provided. However, when there are multiple channels relaying the same game, the service providing server 100 may provide the user with the channel of the specific broadcasting station server 210 requested for the advertisement.

So far, the information recommendation service providing system 1000 and the service providing method according to an embodiment of the present invention have been described. According to the present invention, the service providing server 100 may predict the user's tendency by collecting data and content analysis data associated with the user who is watching the content under the condition that the content-related application is activated. Accordingly, the user who is watching the content may be provided with a service suitable for his/her preference, and the service provider may maximize the effect of providing the service.

Although one embodiment of the present invention has been described in more detail with reference to the accompanying drawings, the present invention is not necessarily limited to these embodiments, and various modifications may be made within the scope without departing from the technical spirit of the present invention. . Therefore, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the technical spirit of the present invention is not limited by these embodiments. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The protection scope of the present invention should be construed by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

1000: information recommendation service system
10: set-top box
20: controller
30: display device
100: information recommendation service providing device
110: communication department
120: storage
130: processor
200: content providing server
210: broadcasting station server
220: content distribution server
230: OTT service providing server

Claims (20)

communication department;
storage; and
a processor operatively connected to the communication unit and the storage unit; including,
The processor is
receiving a service request signal for activating a content-related application from the set-top box;
Receives data related to a content-related application currently being provided through the set-top box according to the service request signal,
Collects content analysis data related to the user from among the pre-stored content analysis data based on the received data,
Predicting the tendency of the current user by using a user analysis model trained to predict the tendency of the user based on the received data and the associated content analysis data,
An information recommendation service providing device, configured to provide a service associated with the predicted tendency to a display device connected to the set-top box. According to claim 1,
The data is received through a controller linked to the set-top box,
The processor is
and receiving at least one of navigation data on the content-related application and output data of the display device. 3. The method of claim 2,
The navigation data is
An apparatus for providing an information recommendation service including data of at least one of left and right, vertical movement, residence time, and selection. According to claim 1,
The processor is
The apparatus for providing an information recommendation service, configured to acquire a frame or an image at a time point when the data is received in the content currently being provided through the set-top box. 5. The method of claim 4,
The processor is
The apparatus for providing an information recommendation service, configured to collect at least one of a person, an object, a place, and a character in the frame as the content analysis data. 5. The method of claim 4,
The processor is
information, configured to convert a person's lines or narration in the image into text when acquiring an image, extract a word to be defined as an object of interest from the converted text, and collect the extracted word as the content analysis data Recommendation service provider. 7. The method of claim 6,
The processor is
and receiving selection information about one person through the communication unit when there are two or more persons in the currently provided content. According to claim 1,
The processor is
An information recommendation service providing apparatus, configured to provide any one of an additional information providing service, a product information providing service, and an advertisement providing service on the content-related application. 9. The method of claim 8,
The content-related application,
An apparatus for providing an information recommendation service, wherein a graphic screen indicating the one service is provided in the form of a pop-up window. 10. The method of claim 9,
The processor is
The apparatus for providing an information recommendation service, configured to identify one of the users matched with the set-top box based on data received through the one service. Receiving a service request signal for activating a content-related application from the set-top box;
receiving data related to a content-related application currently being provided through the set-top box according to the service request signal;
collecting content analysis data associated with a user from among pre-stored content analysis data based on the received data;
predicting a tendency of a current user using a user analysis model trained to predict a tendency of a user based on the received data and the associated content analysis data; and
providing a service associated with the predicted tendency to a display device connected to the set-top box; A method of providing information recommendation service, including. 12. The method of claim 11,
The data is received through a controller linked to the set-top box,
Receiving the data includes:
Receiving at least one of navigation data on the content-related application and output data of the display device, the information recommendation service providing method. 13. The method of claim 12,
The navigation data is
A method for providing an information recommendation service, comprising at least one of left and right, vertical movement, residence time, and selection data. 12. The method of claim 11,
The step of collecting the content analysis data includes:
The method of providing information recommendation service further comprising: acquiring a frame or an image at the point in time when the data is received in the currently provided content through the set-top box. 15. The method of claim 14,
The step of collecting the content analysis data includes:
When a frame is acquired in the acquiring step, collecting at least one of a person, an object, a place, and a character in the frame as the content analysis data, the information recommendation service providing method. 15. The method of claim 14,
The step of collecting the content analysis data includes:
When acquiring an image in the acquiring step, converting the dialogue or narration of a person in the image into text, and collecting the converted text as the content analysis data, the information recommendation service providing method. 17. The method of claim 16,
The step of collecting the content analysis data includes:
When there are two or more persons in the currently provided content, receiving selection information for any one person, the information recommendation service providing method further comprising. 12. The method of claim 11,
The step of providing a service related to the predicted tendency,
The step of providing any one of an additional information providing service, a product information providing service, and an advertisement providing service on the content-related application, the information recommendation service providing method. 19. The method of claim 18,
The content-related application,
A method for providing an information recommendation service, wherein a graphic screen indicating the one service is provided in the form of a pop-up window. 20. The method of claim 19,
After providing a service related to the predicted tendency,
The method of providing an information recommendation service further comprising: identifying one of the users matched with the set-top box based on the data received through the one service.

Images