KR100608645B1 - Apparatus and method for providing realtime of recommendation contents - Google Patents

Abstract

The present invention relates to an apparatus and method for providing recommended content in real time, and more particularly, to an apparatus and method for providing recommended content according to analysis of user's preference and feedback information on content in real time. .

According to an embodiment of the present invention, an apparatus for providing recommended content in real time includes an input unit for receiving a key signal generated from a user, a check unit for checking the input key signal, a preference of the user, and a reproduction of the user. The recommended content list manager reconstructs the recommended content list based on the feedback information about the recommended content, and a display unit for reproducing the recommended content according to the priority of the recommended content list.

Recommended content, recommended content request key, affinity value, feedback information

Description

Apparatus and method for providing realtime of recommendation contents}

1 is a view showing a screen and the operation of the conventional content recommendation system.

2 is an internal block diagram of an apparatus for providing recommended content in real time according to an embodiment of the present invention.

3 is a diagram illustrating a screen for providing recommended content in real time according to another exemplary embodiment of the present invention.

4 is a diagram illustrating an apparatus having a request for recommending content button for requesting recommendation content for providing recommendation content in real time according to another exemplary embodiment of the present invention.

5 is a diagram illustrating a process of reconstructing a recommended content list to provide recommended content in real time according to another exemplary embodiment of the present invention.

6 is a flowchart illustrating a method of providing recommended content in real time according to another embodiment of the present invention.

7 is a flowchart illustrating a process of reconstructing a list of recommended content in a method of providing recommended content in real time according to another embodiment of the present invention.

<Explanation of symbols on main parts of the drawings>

100: recommended content display device

110: input unit 120: check unit

130: recommended content list management unit 140: user information storage unit

140: user information storage unit 150: recommended content list storage unit

160: display unit 170: control unit

The present invention relates to an apparatus and method for providing recommended content in real time, and more particularly, to an apparatus and method for providing recommended content according to analysis of user's preference and feedback information on content in real time. .

In recent years, the content accessible to the user is increasing exponentially due to the development of various A / V media transmission and storage technologies. In particular, with the introduction of digital broadcasting and high-speed Internet infrastructure and increasing the capacity of various A / V devices, users can enjoy numerous contents regardless of time and place.

However, as the amount of content increases, there is a problem that it takes a lot of time and effort to find what the user wants among them, and various effective interface technologies have been attempted to solve this problem.

Among the alternatives, the content recommendation method based on the user's disposition is being tried as a new method to reduce the burden on the user, but it is not used yet.

1 is a diagram illustrating a screen and an operation process of a conventional content recommendation system.

As shown, the content A is displayed on the screen, and a screen composed of a list of recommended content and summary information of each content is displayed when the user requests a recommendation.

Then, after the user checks the list of recommended contents and the summary information of each content, the user selects a predetermined selection method (for example, navigation keys (up / down, prev / next, etc.), numeric keys, and pointing, etc.). Select the content you want.

Thus, the content selected by the user (for example, content B) is displayed on the screen.

That is, the user can view the desired content by performing a process of requesting content recommendation in order to obtain desired content from the recommendation system, checking a list of displayed recommended content, and selecting a desired content using an input device. Can be.

However, the summary information (keywords, titles, playback time, and screenshots) of the contents presented on the recommendation list screen is difficult to know whether the content is the content desired by the user, and the list update period of the recommended content is in real time according to the user's request. There is a problem that it takes a long time to recover if there is an error in the recommendation.

In addition, it is highly likely that users will be ignored because additional learning effort is required on how to use the recommended content list, and if the selected content is not what they want, the process of requesting the list, reconfirming the list, and selecting to finally find the desired content again. It is inconvenient to repeat the back.

In addition, access to the content recommendation system is inferior because an additional process is required to bring up the recommended content list. For example, in an interface-constrained environment, such as a TV or a handheld device, the need for complex input by the recommender system can significantly degrade usability.

Korean Patent Publication No. 2001-074382 (program guide display method according to channel preference) classifies genres by analyzing all channels, and calculates genre analysis and viewing time for previous channels every time the user changes channels. The genre preference value and channel preference value are recalculated and prioritized, and when the user presses the user faculty button, a method of providing programs by priority is disclosed. In providing the priority program guide, inconveniences remain because the user has to repeat the priority program guide re-checking and predetermined program selection process.

An object of the present invention is to play the content in accordance with the priority of the list of recommended content created according to the user's preferences when the recommended content request key is selected from the user.

Another object of the present invention is to provide a list of recommended content reflecting the user's preferences in real time by analyzing the feedback information on the content selected and reproduced from the user in real time, and reconstructing the recommended content list with reference to the analyzed result. will be.

It is still another object of the present invention to provide a button for requesting recommended content, so that when a user selects the corresponding button, content according to priority is provided.

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

In order to achieve the above object, an apparatus for providing recommended content in accordance with an embodiment of the present invention in real time, an input unit for receiving a key signal generated from a user, a check unit for checking the input key signal, and A recommended content list manager reconstructing the recommended content list based on the user's preference and feedback information on the content reproduced by the user, and a display unit playing the content according to the priority of the recommended content list.

In addition, the method for providing recommended content in real time according to another embodiment of the present invention includes receiving a recommendation content request signal generated from a user, and referring to a list of recommended content for the user, the highest priority content Playing the content immediately; reproducing the next priority content in the recommended content list when the recommended content request signal is input again from the user; and playing the content when the recommended content request signal is not input from the user. Checking the feedback information about the content, calculating a preference value for the corresponding content based on the feedback information on the checked content, and reconstructing the recommended content list by applying the preference value for the calculated content. It includes a step.

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

Advantages and features of the present invention and methods for achieving them will be apparent with reference to the embodiments described below in detail with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but can be implemented in various different forms, and only the embodiments make the disclosure of the present invention complete, and those of ordinary skill in the art to which the present invention belongs. It is provided to fully inform the person having the scope of the invention, which is defined only by the scope of the claims. Like reference numerals refer to like elements throughout.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

2 is an internal block diagram of an apparatus for providing recommended content in real time according to an embodiment of the present invention.

As illustrated, the recommended content display apparatus 100 may include an input unit 110, a checker 120, a recommended content list manager 130, a user information storage 140, a recommended content list storage 150, and a display. The unit 160 and the control unit 170 are configured.

The input unit 110 receives a key signal generated from a user, and the generated key signal may be understood as a signal such as a recommended content request key, a content play key, a content prev / next key, and a volume control key. In addition, the content includes both video content and audio content.

The checker 120 checks a key signal input by the input unit 110, that is, checks which command a key signal generated from the user requests and transmits the command signal to the control unit 170.

The recommended content list manager 130 manages the recommended content list according to the user's preference, and reconstructs the recommended content list in real time based on feedback information about the content reproduced by the user. Here, the feedback information may be understood as a reproduction time of the content, a reproduction interval of the content, and a change in the ranking of the content.

That is, the content list manager 130 creates a recommended content list based on information such as the gender and age of the user and information on the content that the user enjoys, and then feedbacks on the content played by the user in the recommended content list. Reconstruct the recommended content list using the information.

For example, the content list manager 130 creates a recommendation content list based on information on a gender, age, and favorite content of the user, and stores the created recommendation content list in the recommended content list storage unit 150. .

Next, when the user selects the recommended content request key, the content list manager 130 plays the content having the highest priority in the previously stored recommended content list. At this time, the content list manager 130 loads the content having the highest priority in the previously stored recommended content list through Equation 1 below.

는 초기 재생되는 컨텐츠

를 나타내며,

는 추천 컨텐츠 목록 저장부(150)에 기 저장된 컨텐츠

의 점수를 나타낸다. here,

Is the initial playing content

Indicates

The pre-stored content in the recommended content list storage unit 150

Indicates a score.

Then, when the recommended content request key is selected again from the user, the content that has been played is ended, and the next higher priority content among the previously stored recommended contents is played back to the user.

In addition, the content list manager 130 calculates a preference value for the corresponding content using the feedback information on the reproduced and completed recommended content, and reconstructs the recommended content list in real time using the calculated preference value.

Here, the feedback information may be understood as a reproduction time of the content, a reproduction interval of the content, and a rank change of the content, and the like, and the feedback score for the corresponding content is calculated based on the score of each item according to the feedback information. In addition, the items for calculating the feedback score (the playing time of the content, the playing interval of the content, the change in the rank of the content, etc.) are exemplary embodiments, and the feedback score may be calculated using other items.

Meanwhile, an equation for calculating a preference value of the corresponding recommended content using the feedback information on the recommended content is shown in Equation 2 below.

Score, Feedback

1 이다. Where 0

Score, Feedback

1

는 t+1 번째 목록에서 컨텐츠

의 점수를 나타내고,

는 t 번째 목록에서 추천된 컨텐츠

의 실시간 피드백 점수를 나타낸다. 그리고,

는 피드백 반영 정도(

)를 나타낸다. In other words,

Is the content from the t + 1 th list

Indicates a score of,

Is the recommended content in the t th list

Represents a real time feedback score. And,

Is the degree of feedback reflection (

).

를 구하는 식은 하기 [수학식3]과 같다. Meanwhile, in [Equation 2]

The equation to obtain is as shown in [Equation 3].

는 t 번째 목록에서 컨텐츠

의 재생 시간을 나타내며,

는 t 번째 목록에서 컨텐츠

의 재생 간격을 나타내며,

는 t 번째 목록에서 컨텐츠

의 순위 변동을 나타낸다. 그리고, a, b, 및 c, 상수이다. here,

Is the content in the t th list

Indicates the duration of

Is the content in the t th list

Playback interval of,

Is the content in the t th list

It indicates the rank change of. And a, b, and c are constants.

For example, the content playback time may be understood as a time ended from the time when the user played the recommended content, and the content playback interval may be understood as a time from the time when the user ends the recommended content to play again. The change in the ranking of the content may be understood as an amount of change in the rank of the score in the recommended content list, such as a rising or falling trend of the ranking. Here, the longer the playback time of the content, the higher the score, and the shorter the playback interval of the content, the higher the score.

Accordingly, the content list manager 130 obtains a preference value for the reproduced and finished content, and then creates a new recommended content list (that is, the t + 1 th list) by applying the corresponding preference value.

)를 구한다(1).For example, looking at the process of constructing a new recommended content list, first, the feedback score (ie,

(1).

)의 평균을 낸 후 전체 컨텐츠 개수(N)로 나눈 값(즉,

)을 구한다(2).Next, the entire contents (

) Is averaged and divided by the total number of content (N) (i.e.

(2).

Then, (1)-(2) are calculated to obtain a predetermined value (3).

)의 선호도 점수에 계산된 (3)을 더함으로써, 현재 재생된 컨텐츠에 대한 피드백 점수를 포함한 새로운 선호도 점수를 계산할 수 있다. Then, the pre-stored currently played content (that is,

By adding (3) calculated to the preference score of), a new preference score including a feedback score for the currently played content can be calculated.

의 점수(즉,

의 선호도 점수)는 추천 컨텐츠 목록 저장부(150)에 저장된 다른 추천 컨텐츠들의 선호도 점수와 비교되어 우선 순위가 정해진다. Then, the calculated

Score (i.e.

Preference score) is compared with preference scores of other recommended contents stored in the recommended content list storage 150 to determine the priority.

At this time, the priority of all the contents is stored in a higher group and a lower group, and only the priority contents included in the upper group are provided to the user.

For example, if 50 recommended contents exist in the recommended content list, the 50 contents have respective priority values.

In this case, the content is divided into upper groups and lower groups according to the priority value of each content, where the upper group may be understood as contents having a priority of 1 to contents having a priority of 10, and lower groups have priority. It may be understood as contents ranging from 11 people to contents having a priority of 50.

Meanwhile, each of the contents is updated with a new preference score reflecting a feedback score in real time, and the priority of each contents can be changed through the updated preference score. Thus, each of the contents can be moved to the parent / child group. Here, the definition of the parent / child group can be changed according to the user's arbitrary.

The user information storage unit 140 stores basic information such as user gender and age, and preference information such as a content preference genre of the user and a preferred actor (musician). Here, the content preference genre may be understood as melodies, comics, and actions in the case of movie contents, and ballads, rock, R & B, jazz, and classics in the case of music contents.

The recommended content list storage unit 150 stores the recommended content list generated by the recommended content list manager 130. Here, the recommended content list stores information updated in real time by the recommended content list manager 130 in real time.

In addition, the recommended content list is divided into upper and lower groups according to the priority of the contents.

The display 160 plays the corresponding contents in order of priority of the recommended contents list according to the user's selection of the recommended contents request key.

For example, if there are five recommended contents in the recommended content list, the user selects the recommended content request key, the content with priority 1 is played first, and then the user selects the recommended content request key again. The content with rank 2 is played.

That is, the content is played in order of high priority whenever the user selects the recommended content request key.

When the input unit 110 receives the key signal generated from the user, the controller 170 instructs the checker 120 to check the input key signal.

When the key signal generated by the user as a result of the check by the checker 120 is a recommended content request key, the controller 170 requests the recommended content list manager 130 to manage the recommended content list.

Accordingly, when the recommended content list manager 130 searches for the highest priority content in the recommended content list stored in the recommended content list storage 150, the controller 170 plays the retrieved content through the display 160. Let's do it.

3 is a diagram illustrating a screen for providing recommended content in real time according to another exemplary embodiment of the present invention. Here, it is assumed that the user is watching the content A through the TV, and the user has a remote controller having a recommendation content request key.

As illustrated, when the user selects the recommended content request key included in the remote controller while viewing the content A, the recommended content list manager 130 searches for the next highest content in the recommended content list next to the content A.

Next, the controller 170 plays the recommended content (eg, content B) found through the display unit 160 on the screen.

Therefore, the user can directly play and view the content according to the preference through the recommendation content request key included in the remote controller.

FIG. 4 is a diagram illustrating an apparatus having a request for recommendation content button for requesting recommendation content for providing recommendation content in real time according to another exemplary embodiment of the present invention.

4A illustrates a case in which the TV remote control is provided with a request for recommending content button (or key) 10 when the recommended content display apparatus is a TV. FIG. 4B illustrates recommended content on a PC monitor when the recommended content display apparatus is a PC. In this case, the request button (or key) 10 is provided. Here, the recommended content request key 10 is a key for receiving a request for playing the recommended content from the user. Whenever the user selects the recommended content request key 10, the contents of the recommended content list are sequentially listed in order of priority. Is played. In addition, the recommendation content request key 10 is provided as one key, so that the user can search and play desired content using only one recommendation content request key 10 to improve user convenience.

That is, the recommended content request button (or key) 10 is provided in close proximity to a user who uses the recommended content display device, so that the user can conveniently search for the desired content through the recommended content request button (or key) 10. Can be selected and played.

5 is a diagram illustrating a process of reconstructing a recommended content list to provide recommended content in real time according to another exemplary embodiment of the present invention.

As shown, the recommended content list storage unit 150 recommends the contents of the upper group to the user, and then the recommended content list management unit 130 plays back the feedback information of the respective contents (content A to content F). Check the time.

Then, the recommended content list management unit 130 calculates a new content score value using [Equation 2]. As a result, the contents are rearranged in the order of contents C, contents A, contents B, contents E, contents F, contents E, and the like. As a result of the rearrangement of the contents, the user's feedback is reflected so that the priority of the content C with the longest playback time goes up two levels, and then the priority of the content A with the long playback time goes down one level. In addition, the priority of content B goes down one level, and content E goes up one level.

Next, the priority is set according to the rearranged content order. At this time, the recommended content list management unit 130 selects the contents D and the content F whose score is lower than the contents of the lower group due to low user feedback. Delete and add new content (content G and content H).

Therefore, the recommended content list is reconstituted with the newly added content and the priority content rearranged according to the playing time.

In addition, although FIG. 5 illustrates an example of reconfiguring the recommended content list by calculating a user's feedback score based on the playing time of the content, the recommended content list may also be reconfigured through the playing interval of the content and the rank change of the content. In addition, the list of recommended content may be reconstructed including all of a content play time, a content play interval, and a rank change of the content.

In addition, the recommended content list may be reconstructed every predetermined period, or may be reconstructed in real time according to the reproduction and completion of the predetermined content.

6 is a flowchart illustrating a method of providing recommended content in real time according to another embodiment of the present invention.

First, when a user selects a recommended content request key to generate a recommended content request signal, the input unit 110 receives a recommended content request signal generated from the user (S100).

Next, the control unit 170 receives the input signal through the check unit 120 that the signal received by the input unit 110 is the recommended content request signal, and the control unit 170 has the highest priority to the recommended content list management unit 130. Request a search for high content.

Next, the recommended content list manager 130 searches for the content having the highest priority by referring to the recommended content list stored in the recommended content list storage 150. Next, the controller 170 plays the retrieved content on the screen through the display 160 (S110).

Next, when the recommended content request key signal is input again through the input unit 110 (S120), the recommended content list manager 130 may refer to the recommended content list stored in the recommended content list storage unit 150, and the priority may be changed. The next high content is retrieved and transmitted to the controller 170.

Accordingly, the controller 170 terminates the content currently being played on the screen and plays the retrieved content on the screen through the display 160 (S130).

Here, if the recommended content request key signal generated from the user is continuously input, the controller 170 continuously provides the content according to the priority recorded in the recommended content list to the user through the display unit 160.

Thus, when the user wants to play new content, the user only needs to press the recommended content request key 10 provided in the predetermined input device (or the recommended content display device), and thus, various contents can be conveniently provided.

On the other hand, when the user terminates the reproduced content or requests another content, the recommended content list manager 130 reconstructs the recommended content list based on the feedback information about the reproduced and finished content (S140).

Hereinafter, a process of reconfiguring the recommended content list will be described with reference to FIG. 7.

First, the recommended content list management unit 130 checks feedback information of content that has been played back (S142). Here, the feedback information may be understood as a reproduction time of the content, a reproduction interval of the content, and a change in the ranking of the content.

For example, the content playback time may be understood as the time when the recommended content is played from the user to the end time, and the content playback interval may be understood as the time from the time when the user ends the recommended content to play it again. The ranking change of the content may be understood as a rising trend or a falling trend of the ranking as a change amount of the score ranking on the recommended content list. Here, the longer the playback time of the content, the higher the score, and the shorter the playback interval of the content, the higher the score.

Next, the recommended content list manager 130 calculates a preference value for the corresponding content based on the checked feedback information (S144). Here, the preference value for the content is calculated using [Equation 2] and [Equation 3], and the calculation process is omitted since it has been described in FIG.

Next, the recommended content list manager 130 reconstructs the recommended content list according to the calculated preference value of the content (S146).

Therefore, by reconstructing the recommended content list by reflecting the user's preferences for the predetermined content in real time, the user can be provided with content that meets their own preferences more quickly.

Although embodiments of the present invention have been described above with reference to the accompanying drawings, those skilled in the art to which the present invention pertains may implement the present invention in other specific forms without changing the technical spirit or essential features thereof. I can understand that. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive.

According to the apparatus and method for providing the recommended content of the present invention as described above in real time, there are one or more of the following effects.

First, when the recommendation content request key is selected by the user, the content is reproduced according to the priority of the recommendation content list created according to the user's preference.

Second, by analyzing feedback information on the content selected and reproduced by the user in real time, and reconfiguring the recommended content list with reference to the analyzed result, there is an advantage of providing a recommendation content list reflecting the user's preference in real time.

Third, by providing a button of a specific function that allows the user to request the recommended content, there is an advantage that the recommended content according to the priority is immediately played according to the user's button selection.

Fourth, the list of recommended contents is not displayed when the user requests the recommended contents, and the contents are played according to the priority, so that the user does not have to perform the cumbersome process of requesting the contents again after checking the recommended list. There is an advantage that can play the content according to the preference.

Claims (5)

An input unit for receiving a key signal generated from a user; A check unit which checks the input key signal; A recommended content list manager configured to reconstruct a recommended content list based on the user's preference and feedback information on the content reproduced by the user; And And a display unit configured to play content according to the priority of the recommended content list. The method of claim 1, The input unit receives a recommended content request signal generated according to a user's recommendation content request key selection, and when the user selects the user's recommendation content request key, the corresponding content is immediately played in the order of priority of the recommendation content list. Device for providing the recommended content in real time. The method of claim 1, The recommended content list manager calculates a preference value of the corresponding content using the feedback information on the reproduced content. The equation for calculating the preference value for the content includes:

Device for providing the recommended content in real time, characterized in that. Receiving a recommendation content request signal generated from a user; Immediately playing the content having the highest priority by referring to the recommended content list for the user; Playing content of a next priority in the recommended content list when the recommended content request signal is input again from the user; Checking feedback information on the reproduced content when the recommended content request signal is not input from the user; Calculating a preference value for the corresponding content based on the feedback information on the checked content; And And reconstructing a recommended content list by applying a preference value for the calculated content. The method of claim 4, wherein Equation for calculating a preference value for the content,

Method for providing the recommended content in real time, characterized in that.

Images