WO2019078387A1

WO2019078387A1 - Apparatus and method for recommending advertisement on basis of content

Info

Publication number: WO2019078387A1
Application number: PCT/KR2017/011701
Authority: WO
Inventors: 김수연; 이상훈
Original assignee: 대구대학교 산학협력단
Priority date: 2017-10-20
Filing date: 2017-10-20
Publication date: 2019-04-25

Abstract

An apparatus for recommending an advertisement on the basis of content comprises a memory and a processor electrically connected to the memory, wherein the processor includes: (a) a function for determining a content theme related to content; (b) a function for linking the content theme with an advertiser intention received through an advertiser terminal; (c) a function for providing, to a user terminal, recommended content in which a user preference is reflected; and (d) a function for providing, during watching of the selected content, an advertisement in which the user preference or the advertiser intention is preferentially reflected according to whether the recommended content is selected when the content is selected by the user terminal.

Description

Content-based ad recommendation device and method

More particularly, the present invention relates to a content-based advertisement recommendation apparatus and method capable of maximizing an advertisement effect between a user of a company that advertises a product and a user who views the advertisement together .

In general, the advertisement recommendation technology provides advertisements in a manner of providing a recommended advertisement for the advertiser when an advertisement environment desired by the advertiser is detected, or searches the advertisement based on the list of advertisement information corresponding to the user interest category, To provide advertisements in a manner that Such conventional technology has a disadvantage in that it is difficult to provide advertisement contents centered on either an advertiser or a user, and to provide advertisement contents suitable for both.

Korean Patent No. 10-1522379 (registered May 5, 2015) relates to a system and method for providing an advertisement centered on a user, in which advertisement information including user benefit information is received from an advertiser terminal and an advertisement category of the advertisement information is set Advertisement information management means for registering and managing advertisement information; User information management means for registering and managing a user's interest advertisement category from a user terminal; And extracting a list of advertisement information of the advertisement category corresponding to the user's interest advertisement category, scoring the exposure priority based on the user benefit information of the advertisement information, sorting the advertisement information list according to the exposure priority score, The user terminal is provided with a user-centered customized advertisement from the advertisement providing server using an application or a web browser.

Korean Registered Patent No. 10-0937474 (Registered on Jan. 11, 2010) stores a series advertisement classified by viewer condition or viewing environment based on an advertiser for a representative advertisement transmitted from terrestrial broadcasting, And then transmits the extracted series advertisement to the receiving terminal of the viewer so that advertiser-based targeted advertisements can be obtained.

An embodiment of the present invention is to provide a content-based advertisement recommendation apparatus and method capable of maximizing an advertisement effect between a company that needs to advertise a product and a user who views an advertisement together.

An embodiment of the present invention is to provide a content-based advertisement recommendation apparatus and method capable of performing a conditional multi-dimensional analysis by causing a result of an advertisement recommendation to be different according to a user's selection of the recommended content.

An embodiment of the present invention is to provide a content-based advertisement recommendation apparatus and method that can reflect user's preference of a user on an advertisement to ultimately obtain user preference information suitable for a user.

In embodiments, a content-based ad recommendation device includes a memory and a processor electrically coupled to the memory. (B) a function of associating the content theme with the advertiser intention received through the advertiser terminal; (c) providing a recommendation content reflecting user preference to the user terminal; (D) a function of providing an advertisement that reflects one of the user preference and the advertiser's intention according to whether or not the content is selected by the user terminal in the course of watching the selected content .

The function (a) includes a sub-function for analyzing the metadata of the content to determine a viewing grade and a genre, and a sub-function for determining a target advertisement target determined based on the viewing grade and genre as the content theme .

(A) a sub-function for analyzing a moving image of the content to determine an identical image subject section that satisfies a specific criterion, and a sub-function for determining a content sub-theme based on at least the object in the same image subject section, Function.

The function (a) may further include a sub-function of adding the content sub-theme and time information on the same video theme section to the content theme.

The function (b) may include a function of receiving a target advertisement object as the advertiser intention and associating the target advertisement object with the content theme.

The function (b) may further include a function of receiving a specific image topic that can be associated with the content sub-theme according to the selection of the advertiser terminal.

The function (c) may include a function of reflecting the content type classified on the basis of at least one of the title, the actor, and the genre related to the content selected by the user in the process of providing the recommended content to the user preference.

The function (c) may further include a function of determining the user preference according to the following equation.

[Equation 1]

User_Pref = {(w1, Maker), (w2, Actor), (w3, Genre)}

The User_Pref may be expressed as a preference vector reflecting the content type constituting the most recently selected content as the user preference. W1, w2 and w3 are the probabilities of selecting the recommended content based on the title of the content type (represented by the Title), the actor (expressed by the Actor) and the genre (expressed by the Genre) The sum of w3 is equal to one.

&Quot; (2) "

Tn {(w1_tn, Maker_tn), (w_tn, Actor_tn), (w3_tn, Genre_tn)}], ..., Tn {(w1_t1, Maker_t1), (w_t2, Actor_t2)

The above equation (2) reflects contents of n (n is a natural number of 2 or more) most recently selected with respect to the above equation (1). The User_Pref_n may be expressed as a set of preference vectors reflecting the content type of the n contents as the user preference, and the recommended content is determined based on at least a part of arbitrary contents selected at the first to n-th points . The T1 {} is determined on the basis of the user preference regarding the content selected at the first point among the n contents, and the Tn {} determines the user preference with respect to the content selected at the n-th point of the n contents. W1_tn, w2_tn and w3_tn are the probability of selecting the recommended content based on the title of the content type (expressed by the Title), the actor (expressed by the Actor), and the genre (expressed by the Genre) , The sum of w1_tn, w2_tn and w3_tn corresponds to one.

The function (d) may include the function of providing the advertisement in the same scene regardless of the selection of the recommended content when the advertiser's intention includes a content sub-theme.

The function (d) may include a function of providing an advertisement that reflects the user preference first when the recommended content is selected.

The function (d) may further include a function of reflecting, in the user preference, a product type related to an advertisement that has been displayed for a predetermined period of time or longer by the user in the process of providing the advertisement.

The function (d) may further include a function of reflecting, in the user preference, a correlated product type obtained through an SNS (Social Network Service) analysis and having a correlation with the product that is higher than a specific reference.

The function (d) may include a function of providing an advertisement reflecting the advertiser's intention first if the recommended content is not selected.

If the advertisement that reflects the advertiser's intention in the process of viewing the selected content is displayed for a predetermined time or longer, the function (d) reflects the product type related to the advertisement to the temporary user preference, which is a copy of the user preference .

The function (d) may further include a function of increasing the weight of the temporary user preference by adjusting the priority of the temporary user and the advertiser intention.

The function (d) may further include reflecting the user preference to be reflected when the content other than the selected content is also selected when the advertisement reflecting the advertiser's intention first is selected by the user.

The function (d) may include a function of determining a priority of reflection according to a setting of an administrator and determining the advertisement.

Among the embodiments, a content-based advertisement recommendation method is performed in a content-based advertisement recommendation apparatus including a memory and a processor electrically connected to the memory. The content-based advertisement recommendation method includes the steps of (a) determining a content theme with respect to content, (b) associating the content theme with an advertiser intention received through an advertiser terminal, (c) To the user terminal, and (d) if the content is selected by the user terminal, the advertisement that reflects one of the user preference and the advertiser's intention according to whether the recommended content is selected is provided during the viewing process of the selected content .

Among the embodiments, a recording medium executable by a computer records the method of the 19th aspect.

The disclosed technique may have the following effects. It is to be understood, however, that the scope of the disclosed technology is not to be construed as limited thereby, as it is not meant to imply that a particular embodiment should include all of the following effects or only the following effects.

The apparatus and method for recommending a content based advertisement according to an exemplary embodiment of the present invention can maximize the advertisement effect between both of the needs of a company that advertises a product and the needs of users who view an advertisement together.

The apparatus and method for recommending a content based advertisement according to an embodiment of the present invention can perform a conditional multi-dimensional analysis by causing a result of recommendation of an advertisement to be different according to user selection of a recommended content.

The apparatus and method for recommending a content-based advertisement according to an embodiment of the present invention may reflect the user's preference of the user on the advertisement to ultimately obtain user preference information suitable for the user.

1 is a view for explaining an advertisement recommendation system according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a content-based advertisement recommendation apparatus shown in FIG. 1. FIG.

Figure 3 is a block diagram illustrating the functional elements of the processor in Figure 2;

4 is a flowchart illustrating a content-based advertisement recommendation procedure performed by the processor shown in FIG.

5 is a flowchart showing a process of providing advertisements when the processor shown in FIG. 2 selects recommended content.

6 is a diagram showing a flow of providing an advertisement when the processor shown in FIG. 2 selects the recommended content as unselected.

7 is a flow diagram illustrating an embodiment of a content based advertisement recommendation procedure performed by the processor of FIG.

The description of the present invention is merely an example for structural or functional explanation, and the scope of the present invention should not be construed as being limited by the embodiments described in the text. That is, the embodiments are to be construed as being variously embodied and having various forms, so that the scope of the present invention should be understood to include equivalents capable of realizing technical ideas. Also, the purpose or effect of the present invention should not be construed as limiting the scope of the present invention, since it does not mean that a specific embodiment should include all or only such effect.

Meanwhile, the meaning of the terms described in the present application should be understood as follows.

The terms " first ", " second ", and the like are intended to distinguish one element from another, and the scope of the right should not be limited by these terms. For example, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

It is to be understood that when an element is referred to as being "connected" to another element, it may be directly connected to the other element, but there may be other elements in between. On the other hand, when an element is referred to as being "directly connected" to another element, it should be understood that there are no other elements in between. On the other hand, other expressions that describe the relationship between components, such as "between" and "between" or "neighboring to" and "directly adjacent to" should be interpreted as well.

It is to be understood that the singular " include " or " have " are to be construed as including the stated feature, number, step, operation, It is to be understood that the combination is intended to specify that it does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

In each step, the identification code (e.g., a, b, c, etc.) is used for convenience of explanation, the identification code does not describe the order of each step, Unless otherwise stated, it may occur differently from the stated order. That is, each step may occur in the same order as described, may be performed substantially concurrently, or may be performed in reverse order.

The present invention can be embodied as computer-readable code on a computer-readable recording medium, and the computer-readable recording medium includes all kinds of recording devices for storing data that can be read by a computer system . Examples of the computer-readable recording medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like. In addition, the computer-readable recording medium may be distributed over network-connected computer systems so that computer readable codes can be stored and executed in a distributed manner.

All terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, unless otherwise defined. Commonly used predefined terms should be interpreted to be consistent with the meanings in the context of the related art and can not be interpreted as having ideal or overly formal meaning unless explicitly defined in the present application.

Referring to FIG. 1, the advertisement recommendation system 100 includes a content-based advertisement recommendation apparatus 110, an advertiser terminal 120, and a user terminal 130.

The content-based advertisement recommendation device 110 corresponds to a computing device that can be connected to the advertiser terminal 120 and the user terminal 130 through a network. In one embodiment, the content-based advertisement recommendation apparatus 110 may be connected to the advertiser terminal 120 through the content agent installed in the advertiser terminal 120 to receive the advertiser's intention from the advertiser terminal 120, The user terminal 130 may be connected to the user terminal 130 through the content agent installed in the user terminal 130 to provide the recommended content to the user terminal 130. Here, if the content agent is installed in the advertiser terminal 120 or the user terminal 130, the content agent may be an agent, which is a software capable of interacting with the terminal and the content-based advertisement recommending apparatus 110 through a network, It corresponds to the program.

The advertiser terminal 120 may transmit the advertiser's intention to the content-based advertisement recommendation device 110 and the user terminal 130 may receive the recommended content from the content-based advertisement recommendation device 110. The advertiser terminal 120 and the user terminal 130 correspond to a computing device that can be connected to the content-based advertisement recommendation apparatus 110 through a network. For example, the advertiser terminal 120 and the user terminal 130 may be implemented as a desktop, a notebook, .

Referring to FIG. 2, the content-based advertisement recommendation apparatus 110 may include a processor 210, a memory 220, a user input / output unit 230, and a network input / output unit 240.

Processor 210 may execute the content based advertisement recommendation procedure shown in FIG. 4 and may manage memory 220 that is read or created throughout the process, and may include volatile memory in memory 220 and non- Synchronization time between memories can be scheduled. The processor 210 may control the entire operation of the content-based advertisement recommendation apparatus 110 and may be electrically connected to the memory 220, the user input / output unit 230, and the network input / output unit 240, Can be controlled. The processor 210 may be implemented as a CPU (Central Processing Unit) of the content-based advertisement recommendation apparatus 110.

The memory 220 may include an auxiliary memory device, which is implemented as a nonvolatile memory such as a solid state disk (SSD) or a hard disk drive (HDD) And may include main memory implemented with volatile memory such as RAM (Random Access Memory).

The user input and output unit 230 includes an environment for receiving user input and an environment for outputting specific information to a user and may be a mouse, a trackball, a touch pad, a graphic tablet, a scanner, a touch screen, An input device including an adapter such as a device, and an output device including an adapter, such as a monitor or touch screen. In one embodiment, the user input and output unit 230 may correspond to a computing device connected via a remote connection, in which case the content-based ad recommendation device 110 may be implemented as a server.

The network input / output unit 240 includes an environment for connecting with the advertiser terminal 120 and the user terminal 130 via a network, and may include an adapter for LAN (Local Area Network) communication, for example

Referring to FIG. 3, the processor 210 may include a content theme determining unit 310, an advertiser's intention associating unit 320, a recommended content providing unit 330, an advertisement providing unit 340 and a control unit 350 have.

The content-theme determining unit 310 determines a content theme regarding the content. More specifically, the content-theme determining unit 310 may analyze each content to detect at least one of a title, a rating, a genre, a producer, an actor, and a material, and determine a content theme based on the detected content characteristic have. In one embodiment, the content theme may correspond to a target advertisement target of the content, and in another embodiment, the content theme may correspond to a content characteristic matched with at least one target advertisement target.

The content-theme determining unit 310 can determine the viewing grade and genre by analyzing the metadata of the content, and can determine the target advertisement target determined based on the viewing grade and genre as the content theme. In one embodiment, the content-theme determining unit 310 can add and manage a viewing grade identifier for a content rating and a genre identifier for a genre to the metadata attribute of the content file, One of a plurality of target advertisement targets set in advance may be determined as a content theme of the content by collectively reflecting the viewing rating and genre of the content detected based on the viewing rating identifier and the genre identifier in the attribute. For example, the content-theme determining unit 310 may analyze the meta data of the content [Pororo] to detect the viewing rating [total viewing] and the genre [adventure] It can be derived as a theme.

The content-theme determining unit 310 may analyze a moving image of the content to determine a same image subject interval that satisfies a specific criterion, and determine a content sub-theme based on at least objects in the same image subject interval You can do more. In one embodiment, the content-theme determining unit 310 analyzes all the tagged tags in at least one specific time point or a specific region in the content, and determines whether the text information of the tag object is a specific reference (for example, The tagging time point or the tagging time point associated with the corresponding tag information can be determined as the same image subject section, and based on the text information of the tag objects in the determined same image subject section, (For example, consecutively matched text information is determined as corresponding text data).

In one embodiment, the content-theme determining unit 310 may analyze the tag object in the content in a time-series manner to give a higher weight to the tag object detected later. For example, the content-theme determining unit 310 determines whether or not the tag information [the start of the adventure], [the joy of the adventure], and the [adventure journey] are tagged in the same three- It is possible to determine the tag information [adventure itinerary] of the tag object which is the latest in the time series in the process of determining the content sub-theme as the content sub-theme.

In one embodiment, the content theme determining unit 310 may add the content sub-theme and time information about the same video theme section to the content theme. For example, when the content sub-theme is detected, the content theme determining unit 310 may further include the content sub-theme information in the content theme and the start time information at which the related same video theme section starts.

The advertiser intention associating unit 320 associates the content theme with the advertiser intent received through the advertiser terminal 120. In one embodiment, the advertiser intent associator 320 may receive a target ad target as an advertiser intent and associate it with a content theme. For example, the advertiser intention associating unit 320 may receive the target advertisement object preferred by the advertiser 120 from the advertiser terminal 120 and manage the advertiser intent according to the advertiser identifier, The advertiser identifier and the advertiser's intention may be associated with each other to form a content population for providing the content associated with the advertisement.

The advertiser intent associator 320 may receive a specific video topic that may be associated with the content sub-theme according to the advertiser terminal 120's selection. In one embodiment, the advertiser intention associator 320 may associate a particular video topic with a content sub-theme of the content when at least one specific video topic is received from the advertiser terminal 120, for example, It is possible to associate [action] received from the advertiser terminal 120 with the contents sub-theme [adventure] of the contents while associating the content theme [child] of [Pororo] with the advertiser intention [child].

The recommended content providing unit 330 provides the user terminal 130 with the recommended content reflecting the user's preference. In one embodiment, the user preferences may include an advertisement preference and a content preference, wherein the advertisement preference may be defined through at least one of a product type, a brand, an ad type, and an advertisement actor, May be defined through at least one of a rating, genre, crew, actor, and material. In one embodiment, the recommended content provider 330 may reflect the content type classified on the basis of at least one of the title, the actor, and the genre related to the content selected by the user in the process of providing the recommended content to the user preference .

In one embodiment, the recommended content provider 330 may determine the user preference according to Equation 1 below. For example, in the case of User_Pref = {(0.7, Winter Sonata), (0.2, BYJ), (0.1, Melodrama)}, The user can have the likelihood of having a probability of 0.2 and the melodrama having a probability of 0.1.

[Equation 1]

User_Pref = {(w1, Maker), (w2, Actor), (w3, Genre)}

(Where User_Pref can be expressed as a preference vector reflecting the content type constituting the most recently selected content as a user preference, w1, w2 and w3 can be represented by a title of the content type (represented by the Title), an actor The sum of w1, w2, and w3 corresponds to one, as a probability of selecting the recommended content based on the genre (expressed as Genre)

In one embodiment, the recommended content provider 330 may determine the user preference in accordance with Equation (2) below.

&Quot; (2) "

Here, the expression (2) reflects contents of n (n is a natural number of 2 or more) most recently selected with respect to the expression (1). Also, the User_Pref_n may be represented as a preference vector set reflecting the content type related to n contents as the user preference, and the recommended content is determined based on at least a part of arbitrary contents selected at the first to n-th points . Also, T1 {} is determined on the basis of the user's preference with respect to the content selected at the first point among the n contents, and Tn {} can determine the user's preference with respect to the content selected at the nth point among the n contents. W1_tn, w2_tn and w3_tn are the probabilities of selecting the recommended content based on the title (expressed as Title), the actor (expressed as Actor) and the genre (expressed as Genre) of the content type constituting the content selected at that point of time and w1_tn , the sum of w2_tn and w3_tn corresponds to one.

When the content is selected by the user terminal 130, the advertisement provider 340 provides an advertisement reflecting the user's preference and the advertiser's intention according to whether or not the recommended content is selected in the process of watching the selected content. For example, when the recommended content is selected by the user terminal 130, the advertisement providing unit 340 may search for the first advertisement reflecting the user preference first and provide the first advertisement during the viewing process of the corresponding content. If not, Can be searched for and provided.

In one embodiment, when the advertiser intention includes a content sub-theme, the advertisement providing unit 340 may provide the advertisement in the same scene regardless of the selection of the recommended content. For example, if the advertiser's intention includes the content sub-theme [adventure], the advertisement provider 340 may provide the content sub-theme of the content in the process of providing the specific content [Pororo] to the user terminal 130 Depending on whether the adventure matches the content sub-theme [adventure] of that advertiser's intent or matches more than a certain number of characters, you can choose whether to serve that ad at a certain point in the same video topic segment associated with the [adventure] You can decide.

When the recommended content is selected, the advertisement providing unit 340 may provide an advertisement reflecting the user preference first. This will be described with reference to FIG.

5 is a flowchart showing a process of providing advertisements when the processor shown in FIG. 2 selects recommended content. In FIG. 5, when the recommended content is selected by the user terminal 130, the advertisement provider 340 determines that the recommendation based on the user preference of the user is well applied, and determines the advertisement preference based on the user preference. For example, the recommended content providing unit 330 detects [Movie preference of a genre] from the content preference in the user preference of the user, searches for specific movie contents in the content population matching the movie of the corresponding A genre The advertisement providing unit 340 can provide the recommended content to the user terminal 130. If the provided recommended content is selected by the user, the advertisement provider 340 detects the shoe product from the advertisement preference in the user preference, A specific shoe advertisement matching the shoe product can be searched and provided as an advertisement.

In the above embodiment, the advertisement providing unit 340 may provide an advertisement reflecting the advertiser's intention preferentially regardless of the selection of the recommended content in the case of the content having the content theme consistent with the advertiser's intention. For example, if the content theme and the advertiser's intention match (e.g., the content theme [child] matches the corresponding advertiser's intention [child]), or if the content has a difference within a predetermined standard range (For example, if the difference between the set value of the content theme [child] and the intended advertiser [youth] is less than the reference range), the advertiser may be provided with an advertisement that reflects the advertiser's intent irrespective of the selection of the recommended content.

The advertisement providing unit 340 may reflect the product type related to the advertisement that has been displayed by the user for a specific time or longer in the user preference in the process of providing the advertisement. For example, the advertisement providing unit 340 can update the user preference by reflecting the product type related to the advertisement reproduced for 10 seconds or more to the advertisement preference.

The advertisement providing unit 340 may reflect a correlated product type, which is obtained through an SNS (Social Network Service) analysis and whose correlation with the product is higher than a specific standard, to the user preference. For example, when the product type related to the advertisement that has been displayed by the user for a specific time or more corresponds to the automobile, the advertisement providing unit 340 may detect the fragrance as a correlation product type correlated with the product type through SNS analysis And can be reflected in the user preference.

If the recommended content is not selected, the advertisement providing unit 340 may provide an advertisement reflecting the advertiser's intention first. This will be described further with reference to FIG.

6 is a diagram showing a flow of providing an advertisement when the processor shown in FIG. 2 selects the recommended content as unselected. 6, if the recommended content is not selected by the user terminal 130, the advertisement provider 340 determines that the recommendation based on the user preference of the user is not applied well, have. For example, if the recommended content provider 330 searches for specific movie content that matches the user preference and provides the recommended movie content to the user terminal 130 but the recommended content is not selected by the user, the advertisement provider 340 ) Can be provided as an advertisement by bringing a [Coke] advertisement in the advertisement population and matching with the advertiser's intention.

In the above embodiment, the advertisement providing unit 340 may decide to provide the advertisement reflecting the user's preference regardless of whether the entire section or the total number of the advertisement providing service is unselected, The user preference may be updated by feeding back the user selection according to the provision of the user preference. Accordingly, although the recommendation according to the user preference is not applied initially, the advertisement provider 340 may gradually reflect the feedback of the advertisement in the user preference so that the advertisement providing rate based on the user preference may be increased .

In one embodiment, if the recommended content is not selected, the advertisement provider 340 may provide an advertisement considering the intention of the advertiser and the content theme in a comprehensive manner. For example, when the recommended content is not selected by the user, the advertisement providing unit 340 may randomly select one of the [COOLER] advertisements matching the advertiser's intention and the [movie theater] And can be brought as an advertisement.

In one embodiment, if an advertisement reflecting the advertiser's intention in the course of viewing the selected content is displayed for a specific time or longer, the advertisement providing unit 340 reflects the product type related to the advertisement in the temporary user preference, which is a copy of the user's preference . For example, if an advertisement reflecting the intention of the advertiser associated with the content theme of the content is played for 10 seconds or more in the course of watching the selected recommended content, the advertisement providing unit 340 displays the product type of the advertisement in the user preference It can be judged to be in conformity and the corresponding product type can be reflected in the temporary user preference.

In one embodiment, the advertisement provider 340 may adjust the priority of the temporary user and the advertiser intention to increase the proportion of the reflectance of the temporary user preference. Accordingly, even if the content theme and the advertiser intention are related to each other, the advertisement providing unit 340 can ultimately increase the proportion of reflecting the preference of the temporary user.

In one embodiment, the advertisement provider 340 may reflect the user preference so that when an advertisement reflecting the advertiser's intention is selected by the user, content other than the selected content is also selected. For example, the advertisement providing unit 340 may reflect the product type of the selected advertisement in the advertisement preference of the user preference.

In one embodiment, the advertisement provider 340 can determine the advertisement priority by determining the reflection probability first according to the administrator's setting. For example, the advertisement provider 340 may determine a preference reflection of one of the user preference and the advertiser intention based on the priority preference probability designated by the administrator, and if the preference preference probability of the user preference is 3/4 If it is designated, the probability calculation of whether or not the user preference first reflects the probability value is processed, and if the user preference preference reflection is determined, the provision of the advertisement reflecting the user preference can be determined.

In the above embodiment, the advertisement providing unit 340 may adjust the priority reflecting probability according to whether the corresponding advertisement is selected. For example, when the user preference reflecting advertisement provided based on the user preference first reflecting probability designated by the administrator Is selected, it is possible to update the user preference first reflection probability up to a specific rate (for example, 10%).

The control unit 350 can control the entire operation of the processor 210 and can store data between the content theme determining unit 310, the advertiser's intention associating unit 320, the recommended content providing unit 330 and the advertisement providing unit 340 You can control the flow.

In Fig. 4, the processor 210 performs a function of determining a content theme regarding the content (step S410). The processor 210 performs a function of associating the content theme with the advertiser intention received through the advertiser terminal 120 (step S420). The processor 210 performs a function of providing the user terminal 130 with the recommended content reflecting the user's preference (step S430). When the content is selected by the user terminal 130, the processor 210 provides the advertisement that reflects one of the user preference and the advertiser's intention according to whether the recommended content is selected in the process of watching the selected content (step S440 ).

In FIG. 7, the processor 210 may match the content theme with the advertiser intention for each content. The processor 210 may analyze the content to determine the content theme, and may associate the advertiser's intent with the content theme.

The processor 210 may detect [Movie Preference] from the content preference in the user preference of the user, determine specific movie content matched with the content preference as the recommended content, and provide the determined movie content to the corresponding user terminal 130, The user may be provided with an advertisement based on the user's preference or an advertisement based on the advertiser's intent depending on whether the recommended content is selected by the user.

The processor 210 may determine that the recommendation based on the user preference is suitable for the user if the recommended content is selected by the user terminal 130 and determine the advertisement based on the user preference. For example, the processor 210 detects [shoes advertisement preference] from the advertisement preference in the user preference of the user, determines a specific shoe advertisement matching the advertisement preference, and provides the advertisement to the corresponding user terminal 130 .

The processor 210 may determine that the recommendation based on the user's preference is inappropriate for the user if the recommended content is not selected by the user terminal 130 to determine the advertiser's intention or the content theme-based advertisement selection. For example, the processor 210 fetches a specific popcorn or cola advertisement associated with the content and matches the advertiser's intention, or fetches a specific movie theater advertisement that matches the content theme of the content, and provides the advertisement to the user terminal 130 as an advertisement can do.

The processor 210 may update the user preferences based on user preferences provided to the user terminal 130, advertiser intent, or whether a content theme based advertisement is viewed by a user. For example, the processor 210 may update the user preference by removing the corresponding product type from the advertisement preference of the user preference or by subtracting the preference value of the corresponding product type when the user has not provided an advertisement reflecting the user preference And if the user provides the advertisement reflecting the intention of the advertiser, the user can update the user preference by adding the product type to the advertisement preference of the user preference or by adding the preference value of the product type.

In this way, the processor 210 can maximize the effect of the purchase inducement of the user in accordance with the advertisement by making the advertisement provided according to the user selection of the recommended content different, and it is possible to maximize the user's preference according to the selection And ultimately obtain user preference information suitable for the user.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as set forth in the following claims And changes may be made without departing from the spirit and scope of the invention.

Claims

A memory and a processor electrically coupled to the memory,

The processor

   (a) a function of determining a content theme regarding the content;

   (b) associating the content theme with an advertiser intention received through an advertiser terminal;

   (c) providing the user terminal with a recommendation content reflecting the user preference; And

   (d) a function of providing an advertisement, which reflects one of the user preference and the advertiser's intention, according to whether or not the content is selected by the user terminal in the course of watching the selected content, Ad recommendation device.
The method of claim 1, wherein (a)

A sub-function for analyzing the metadata of the content to determine a viewing grade and a genre; And

And a sub-function of determining a target advertisement target determined based on the viewing grade and genre as the content theme.
3. The method of claim 2, wherein (a)

A sub-function for analyzing the moving image of the content to determine an identical image subject section that meets a specific criterion; And

And a sub-function for determining a content sub-theme based on at least the object in the same video subject section.
4. The method of claim 3, wherein the function (a)

And a sub-function of adding the content sub-theme and the time information on the same video subject section to the content theme.
The method of claim 1, wherein the function (b)

And receiving the target advertisement target as the advertiser intention and associating the target advertisement target with the content theme.
6. The method of claim 5, wherein the function (b)

Further comprising a function of receiving a specific video topic that can be associated with the content sub-theme according to the selection of the advertiser terminal.
The method of claim 1, wherein the function (c)

And a function of reflecting the content type classified on the basis of at least one of a title, an actor, and a genre related to the content selected by the user in the process of providing the recommended content to the user preference. Device.
8. The method of claim 7, wherein the function (c)

Further comprising a function of determining the user preference according to the following equation.

[Equation 1]

User_Pref = {(w1, Maker), (w2, Actor), (w3, Genre)}

The User_Pref may be expressed as a preference vector reflecting the content type constituting the most recently selected content as the user preference

W1, w2 and w3 are the probabilities of selecting the recommended content based on the title of the content type (represented by the Title), the actor (expressed by the Actor) and the genre (expressed by the Genre) The sum of w3 equals 1
9. The method of claim 8, wherein the function (c)

Further comprising a function of determining the user preference according to the following equation.

&Quot; (2) "

Tn {(w1_tn, Maker_tn), (w_tn, Actor_tn), (w3_tn, Genre_tn)}], ..., Tn {(w1_t1, Maker_t1), (w_t2, Actor_t2)

The above equation (2) reflects the contents of the n most recently selected (n is a natural number of 2 or more) with respect to the above-mentioned formula (1)

The User_Pref_n may be expressed as a set of preference vectors reflecting the content type of the n contents as the user preference, and the recommended content is determined based on at least a part of arbitrary contents selected at the first to n-th points Means

The T1 {} is determined based on a user preference for a content selected at a first point among the n contents, and the Tn {} is determined based on a user preference for a content selected at a n-th point among the n contents

W1_tn, w2_tn and w3_tn are the probability of selecting the recommended content based on the title of the content type (expressed by the Title), the actor (expressed by the Actor), and the genre (expressed by the Genre) , The sum of w1_tn, w2_tn and w3_tn corresponds to 1
The method of claim 1, wherein the function (d)

And if the advertiser intention includes a content sub-theme, providing the advertisement in the same scene section regardless of the selection of the recommended content.
The method of claim 1, wherein the function (d)

And if the recommendation content is selected, providing an advertisement that reflects the user preference first.
12. The method of claim 11, wherein the function (d)

Further comprising a function of reflecting, in the user preference, a product type related to an advertisement that has been displayed for a predetermined time or longer by a user in the process of providing the advertisement.
13. The method of claim 12, wherein the function (d)

Further comprising a function of reflecting a correlated product type obtained through an SNS (Social Network Service) analysis and having a correlation with the product to a predetermined criterion or higher, in the user preference,
The method of claim 1, wherein the function (d)

And if the recommended content is not selected, providing an advertisement reflecting the advertiser's intention first.
15. The method of claim 14, wherein the function (d)

When the advertisement reflecting the advertiser's intention in the course of viewing the selected content is displayed for a specific time or longer, reflecting the product type related to the advertisement in the temporary user preference which is a copy of the user's preference Content based ad recommendation device.
16. The method of claim 15, wherein (d)

Further comprising a function of adjusting a priority between the temporary user preference and the advertiser intention to increase the proportion of the reflectance of the temporary user preference.
17. The method of claim 16, wherein (d)

Further comprising a function of reflecting the user preference to be reflected even when content other than the selected content is selected when an advertisement reflecting the advertiser's intention first is selected by the user.
The method of claim 1, wherein the function (d)

And a function of determining the advertisement priority by determining a priority reflection probability according to a setting of an administrator.
A content-based advertisement recommendation method performed in a content-based advertisement recommendation apparatus including a memory and a processor electrically connected to the memory,

(a) determining a content theme for the content;

(b) associating the content theme with an advertiser intention received through an advertiser terminal;

(c) providing the user terminal with a recommendation content reflecting the user preference; And

(d) providing, when the content is selected by the user terminal, an advertisement that reflects one of the user preference and the advertiser's intention according to whether the recommended content is selected in a viewing process of the selected content Ad recommendation method.
A recording medium that can be executed by a computer recording the method of claim 19.