WO2017057793A1

WO2017057793A1 - Method and apparatus for managing content

Info

Publication number: WO2017057793A1
Application number: PCT/KR2015/010970
Authority: WO
Inventors: Jae Hwan Park; Jae Won Lim
Original assignee: Melephant Inc.
Priority date: 2015-09-30
Filing date: 2015-10-16
Publication date: 2017-04-06

Abstract

Provided is a method of managing a content by a content managing apparatus. The content managing apparatus receives first information representing which replay section of the content a user prefers. The content managing apparatus evaluates a user preference for each replay section for the content based on the first information. The content managing apparatus transmits the user preference for each replay section to a user terminal so that the user preference for each replay section is displayed by the user terminal. In addition, the content managing apparatus increases or decreases a lifespan of the content which is a period in which a user can use the content based on the user preference for each replay section.

Description

METHOD AND APPARATUS FOR MANAGING CONTENT

The present invention relates to a method and an apparatus for managing a content.

Digital content means data or information which prepares, processes, and distributes a code, a text, a voice, a sound, an image, a video, and the like in a digital format in order to be used in a wired/wireless electrical communication network. Recently, as rapid information and communication society has arrived, there is a positive function in that many contents are created and distributed each day to enrich cultural life of people, but a negative function has reached a negligible level due to the spread of the indiscriminate contents.

For example, there are many contents which exist only on a server even though there is no value of existence of the contents socially because a valid term has expired or a user is not concerned with the content any longer, among the contents, and as a result, there are problems in that there is a limit in storing capacity of the server and a lot of time and efforts are required to find desired contents even in terms of the user who uses the contents.

Meanwhile, currently, there is no related art that reflects evaluation (preference) of the user for the contents to determine whether to grant a life to the contents, and the lifespan (an existence period) of the contents is also determined only by a lapse of an absolute time (a physical time generally used in real life).

The present invention has been made in an effort to provide a method and an apparatus of granting a life of a content and determining a lifespan (an existence period) of the content based on a user preference for the content other than a lapse of an absolute time (a physical time generally used in real life).

Further, the present invention has been made in an effort to provide a method and an apparatus for managing a content efficiently based on the preference for the content.

In addition, the present invention has been made in an effort to provide a method and an apparatus for predicting and recommending a content which may be preferred by the user with high accuracy.

An exemplary embodiment of the present invention provides a method for managing a content by a content managing apparatus. The method includes: receiving first information representing which replay section of the content a user prefers; evaluating a user preference for each replay section for the content based on the first information; transmitting the user preference for each replay section to a user terminal so that the user preference for each replay section is displayed by the user terminal; and increasing or decreasing a lifespan of the content which is a period in which a user can use the content based on the user preference for each replay section.

The method may further include: receiving second information representing which area of a video included in the content a user prefers; evaluating a user preference for each video area for the content based on the second information; and transmitting the user preference for each video area to the user terminal so that the user preference for each video area is displayed by the user terminal.

The increasing or decreasing the lifespan of the content may include increasing or decreasing the lifespan of the content based on the user preference for each replay section and the user preference for each video area.

The terminal may display replay sections of the content on a running time gauge representing a running time of the content by using a color corresponding to the user preference for each replay section.

The terminal may display the user preference for each video area on the video included in the content by using a color corresponding to the user preference for each video area.

The user terminal may determine how much a first user prefers the content based on at least one of intensity of pressure applied to a touched area and an area of the touched area when the area of the video included in the content is touched by the first user and generate at least one of first information and second information by using the determined result.

Another exemplary embodiment of the present invention provides an apparatus for managing a content. The apparatus may include: a calculation unit configured to calculate a speed representing the degree that a user preference for a content varies with time and an acceleration of the speed; a candidate group generation unit configured to include contents in the candidate group, in which a first value calculated by at least one of the speed and the acceleration is more than a first threshold value; a prediction unit configured to predict a content which may be preferred by a user among the contents included in the candidate group by using at least one of regression analysis and collaborative filtering, the speed, and the acceleration; and a processor configured to control the calculation unit, the candidate group generation unit, and the prediction unit.

The apparatus may further include a lifespan control unit controlled by the processor and configured to increase or decrease a lifespan of each predicted content based on the user preference for each predicted content.

The prediction unit may rank the predicted contents and select a predetermined number of contents among the ranked contents.

The selected contents may be displayed by a terminal.

The lifespan control unit may control the lifespan of a first content so as to not delete the first content over time when the user preference for the first content among the predicted contents is more than a second threshold value.

Yet another exemplary embodiment of the present invention provides an apparatus for managing a content. The apparatus includes: a determination unit configured to determine which group a user of a terminal belongs to among a plurality of user groups by using first data representing a history of using a service provided by the content managing apparatus for the user of the terminal when the terminal accesses the content managing apparatus; a prediction unit configured to predict a content which may be preferred by the user of the terminal by using an algorithm corresponding to the user group to which the user of the terminal belongs among a regression analysis algorithm, a collaborative filtering algorithm, and a market basket analysis algorithm; and a processor configured to control the determination unit and the prediction unit.

The apparatus may further include: a calculation unit controlled by the processor and configured to calculate a speed representing the degree that a user preference for a content varies with time and an acceleration of the speed with respect to each of a plurality of contents; and a candidate group generation unit controlled by the processor and configured to include contents of which a first value calculated by at least one of the speed and the acceleration is more than a first threshold value in the candidate group among the plurality of contents.

The prediction unit may predict a content which may be preferred by the user of the terminal among the contents included in the candidate group by using an algorithm corresponding to the user group to which the user of the terminal belongs, of the regression analysis algorithm and the collaborative filtering algorithm.

The apparatus may further include a detection unit configured to detect information on the terminal.

The candidate group generation unit may determine a first content which can not be replayed by the terminal among the contents included in the candidate group by using the information on the terminal, and exclude the first content from the candidate group.

The determination unit may calculate the degree that the user of the terminal uses the service based on the first data, determine that the user of the terminal belongs to a first user group of the plurality of user groups when the degree of using the service is more than a first reference value, and determine that the user of the terminal belongs to a second user group of the plurality of user groups when the degree of using the service is less than the first reference value.

The prediction unit may calculate a first value which is an average of the preference granted by the user of the terminal with respect to a plurality of first contents which is preferred by the user of the terminal when the user of the terminal belongs to the first user group. The prediction unit may determine a content preference tendency similarity between each of first users and the user of the terminal based on the preference granted to the plurality of first contents by the first users preferring at least one of the plurality of first contents. The prediction unit may calculate a second value which is a preference granted to the second content by the first user having the highest content preference tendency similarity among the first users. The prediction unit may calculate a third value which is an average of the preference granted to the second content by the first users of which the content preference tendency similarity is more than a second reference value among the first users. The prediction unit may predict how much the user of the terminal prefers the second content by applying the first value, the second value, and the third value to the collaborative filtering algorithm.

The collaborative filtering algorithm may apply the same or different weight values to the second value and the third value.

The prediction unit may predict how much the user of the terminal prefers the first content by applying user information provided for using the service by the user of the terminal, a kind of terminal, and a kind of operating system (OS) installed in the terminal to a logistic regression analysis algorithm which is the regression analysis algorithm, when the user of the terminal belongs to the second user group.

The logistic regression analysis algorithm may apply the same or different weight value to each of the user information, the kind of terminal, and the kind of OS.

The prediction unit may calculate a first value which is the number of users granting the preference to at least one of a plurality of contents for a first period. The prediction unit may calculate a second value which is the number of users granting the preference to both a first content and a second content among the plurality of contents for the first period. The prediction unit may calculate a third value which is the number of users granting the preference to the first content for the first period. The prediction unit may calculate a fourth value which is the number of users granting the preference to the second content for the first period. The prediction unit may calculate a fifth value by dividing the third value by the first value, and calculate a sixth value by dividing the fourth value by the first value. The prediction unit may calculate a sixth value by dividing the fourth value by the first value. The prediction unit may calculate a first index value by dividing the second value by the first value, a second index value by dividing the second value by the third value, a third index value by dividing the first index value by a multiple of the fifth value and the sixth value. The prediction unit may apply the first index value, the second index value, and the third index value to the basket analysis algorithm to predict how much the user of the terminal prefers the second content when the user of the terminal prefers the first content.

According to exemplary embodiments of the present invention, it is possible to grant a lifespan to the contents by a time corresponding to a user preference based not on a lapse of an absolute time but on the user preference for contents.

Further, it is possible to increase or decrease the lifespan of the contents based on the user preference for the contents even after the time when a life is granted to the contents.

It is also possible to select and manage only contents selected by users and remove contents which cannot be socially selected. As a result, it is possible to efficiently use and manage a space of a content storage apparatus.

In addition, it is possible to efficiently determine contents which are currently most preferred or contents which are currently most issued by managing contents by using a speed and an acceleration indicating a degree in which preference for the contents varies over time.

It is also possible to allow the user to intuitively find the preference for the contents by showing the preference for the contents for each reproduction section or for each image area.

Further, it is possible to predict the contents which may be preferred by the user with higher accuracy and recommend the predicted contents to the user by selecting the contents by using the speed and the acceleration of the contents, and predicting contents which may be preferred by the user are predicted by using information on a terminal used by the user.

FIG. 1 is a diagram illustrating a system of managing a content according to an exemplary embodiment of the present invention.

FIG. 2 is a diagram illustrating a content managing apparatus according to the exemplary embodiment of the present invention.

FIG. 3 is a diagram illustrating a method of expressing a user preference for each replay section of a content according to the exemplary embodiment of the present invention.

FIGS. 4a and 4b are diagrams illustrating a method of expressing a user preference for each video area of a content according to the exemplary embodiment of the present invention.

FIG. 5 is a diagram illustrating a process of expressing a user preference for each replay section and a user preference for each video area for a content by the content managing apparatus according to the exemplary embodiment of the present invention.

FIG. 6 is a diagram illustrating a method of managing a popular content by the content managing apparatus according to the exemplary embodiment of the present invention.

FIG. 7 is a diagram illustrating a method of recommending a content by the content managing apparatus according to the exemplary embodiment of the present invention.

FIG. 8 is a diagram illustrating a content managing apparatus according to another exemplary embodiment of the present invention.

FIG. 9 is a diagram illustrating a content server according to the exemplary embodiment of the present invention.

FIG. 10 is a diagram illustrating a content database apparatus according to the exemplary embodiment of the present invention.

In the following detailed description, only certain exemplary embodiments of the present invention have been shown and described, simply by way of illustration. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. Accordingly, the drawings and description are to be regarded as illustrative in nature and not restrictive. Like reference numerals designate like elements throughout the specification.

1. Method of managing lifespan of content

FIG. 1 is a diagram illustrating a system 1000 for managing a content according to an exemplary embodiment of the present invention.

The system 1000 for managing the content includes a terminal 100, a content server 200, a content managing apparatus 300, and a content database apparatus 400.

The terminal 100 provides or receives the content by accessing the content server 200.

The content server 200 stores and manages the content uploaded from the terminal 100 in the content database apparatus 400.

The content managing apparatus 300 grants a life to the content according to a user preference for the content and controls a lifespan of the content with the life (for example, a period when the user may use the content). Non-granting the life to the content may include at least one of deleting the content in the content server 200 or the content database apparatus 400, deleting the content at a place (for example, a site) posted with the content while existing in the content server 200, and making it look like the content is deleted to users while existing in the content server 200. Meanwhile, granting the life to the content may include continuously leaving the content in a state usable by the user without deleting the content in the content server 200 or the content database apparatus 400.

The content managing apparatus 300 may control the lifespan of the content granted with the life according to at least one of an expression degree of the preference granted to the content by the user (for example, the number of hearts, the number of 'good', the number of 'bad', the number of 'save', the number of 'delete', and the like), the number of comments sensed for the content, the calling number of times of the content in the content server 200, the play number of times of the content, a store selected by using a digital gauge by the user, and a score directly input to the content by the user.

The content database apparatus 400 stores a content uploaded by the terminal 100 accessing the content server 200, content lifespan information provided by the content managing apparatus 300, and the like.

The system 1000 for managing the content may include a function of a system 1000 for managing the lifespan of a content disclosed in Korean Patent No. 10-1519106. Further, components 100 to 400 of the system 1000 for managing the content may include functions of components 100 to 400 of the system 1000 for managing the lifespan of the content disclosed in Korean Patent No. 10-1519106. Hereinafter, functions (alternatively, operations and methods) other than the function (alternatively, an operation and a method) disclosed in the Korean Patent No. 10-1519106 among the functions (alternatively, the operations and the methods) of the system 1000 for managing the content and the components 100 to 400 thereof will be described in detail.

FIG. 2 is a diagram illustrating the content managing apparatus 300 according to an exemplary embodiment of the present invention.

The content managing apparatus 300 includes a pre-evaluation unit 310, an additional point calculation unit 320, a preference evaluation unit 330, a life granting unit 340, a lifespan control unit 350, a lifespan display unit 360, a restoration determination unit 370, a processor 380, a memory 381, and a communication device 382.

The pre-evaluation unit 310 executes pre-evaluation for the content before the user preference for the content is evaluated by the preference evaluation unit 330. In detail, the pre-evaluation unit 310 may perform an operation related to the pre-evaluation for the content disclosed in Korean Patent No. 10-1519106.

The additional point calculation unit 320 calculates an additional point to be applied to the user preference of the content when a pre-evaluation score of the content calculated by the pre-evaluation unit 310 is more than a threshold value. In detail, the additional point calculation unit 320 may perform an operation related to applying (granting) the additional point, which is disclosed in Korean Patent No. 10-1519106.

The preference evaluation unit 330 evaluates the user preference for the content. In detail, the preference evaluation unit 330 may perform an operation related to evaluating (collecting) the user preference, which is disclosed in Korean Patent No. 10-1519106.

The life granting unit 340 grants the life to the content based on the user preference of the content. In detail, the life granting unit 340 may perform an operation related to granting the life of the content, which is disclosed in Korea Patent No. 10-1519106.

The lifespan control unit 350 increases or decreases the lifespan of the content based on the user preference of the content calculated by the preference evaluation unit 330 after the life is granted to the content. In detail, the lifespan control unit 350 may perform an operation related to controlling (extending/shortening and the like) the lifespan of the content, which is disclosed in Korean Patent No. 10-1519106.

The lifespan display unit 360 visually displays degeneration or aging of the content based on a ratio between a total lifespan and a remaining lifespan of the content. Further, the lifespan display unit 360 may visually display the lifespan initially granted in the content, a remaining lifespan of the content, and a total survival time of the content. In detail, the lifespan display unit 360 may perform an operation related to visualization for the aging of the content and visualization for the lifespan of the content and the like, which is disclosed in Korea Patent No. 10-1519106.

The restoration determination unit 370 determines whether to restore the content when the life is not granted to the content by the life granting unit 340 or the lifespan of the content is expired while the life is granted to the content. In the case when the content obtains a restoration opportunity by the restoration determination unit 370, the user preference may be again evaluated by the preference evaluation unit 330. In detail, the restoration determination unit 370 may perform an operation related to restoring the content, which is disclosed in Korean Patent No. 10-1519106.

The processor 380 may be constituted to implement procedures, functions, and methods related to the content managing apparatus 300 described in this specification. The processor 380 controls respective components 310 to 370, 381, and 382 of the content managing apparatus 300.

The memory 381 is connected with the processor 380 and stores various information related to the operation of the processor. The communication device 382 is connected with the processor 380 and transmits or receives a wired signal or a wireless signal.

2. Method of expressing user preference for content

As described above, the lifespan of the content is controlled according to the preference (for example, the preference expression such as 'good'/'bad', or 'save'/'delete') granted to the corresponding content by the users from the time when the corresponding content is uploaded in the content server 200 and published to the user. The granting of the preference to the content by the user may include pressing a 'good' button or a 'bad' button for the corresponding content by the user, pressing a heart button for the corresponding content, or pressing a video area of the corresponding content while the corresponding content is replayed. The user preference for the content may be measured based on the number of preference expressions (for example, good, bad, heart, and the like) granted to the corresponding content by the user, and the number of preference expressions which may be granted to one content by the user is limited to 1 or may be limited to a plurality of numbers (a predetermined number). Hereinafter, for convenience of description, a case where, as the preference expression for the content, the user grants the heart to the corresponding content will be exemplified as an exemplary embodiment of the present invention. Here, the granting of the heart to the content includes pressing a heart button for the corresponding content, pressing a video area of the corresponding content while the corresponding content is replayed, or the like. However, this is just an example, and in addition to granting the heart to the content, pressing a 'good' button for the content, pressing a 'bad' button for the content, preparing comments for the content, chatting about the content, or the like may be applied to the exemplary embodiment of the present invention as the preference expression for the content.

Meanwhile, since the lifespan granted to the content is controlled in real time according to the user preference for the corresponding content, the lifespan of the corresponding content may repeatedly increase or decrease. In this case, a time when the lifespan of the content is negative (-) may occur. The content managing apparatus 300 may not immediately delete the corresponding content from the content server 200 at the time when the lifespan of the content is negative (-) and induce the preference expression (heart granting) of the users in order to save the corresponding content (so as to not delete the corresponding content). For example, the content managing apparatus 300 may allow a message inducing the preference expression (heart granting) for the corresponding content to be displayed on the terminal of the user while showing that the lifespan of the content is negative to the user (displaying that the lifespan of the content is negative through the terminal of the user). If the corresponding content receives the heart from the users, the lifespan of the corresponding content may be converted from the negative to positive (+) according to the number of granted hearts.

Meanwhile, when the content is a video content, the user may grant one heart (alternatively, a plurality of hearts) to a preferable replay section among the replay sections of the video content. As a result, among the replay sections of the video content, the most hearts may be granted to the most impacted replay section. For example, when the replay section where the most hearts are granted corresponds to a middle of the entire replay section of the corresponding video content, the user preference for each replay section for the video content may be expressed by a bell-shaped standard distribution curve. The content managing apparatus 300 may visually (for example, with a graph, a color, and the like) express the user preference for each replay section for the corresponding video content through the terminal 100 replaying the corresponding video content. As a result, the content managing apparatus 300 may notify which replay section of the video content is most preferred by the users to the users.

In detail, the terminal 100 determines a replay section where the hearts are granted by the user of the terminal 100 among the replay sections of the corresponding video content when the user of the terminal 100 grants the hearts to the video content, and may transmit information (for example, information of the replay section where the hearts are granted, the number of granted hearts, and the like) on hearts granted by the user of the terminal 100 to the content managing apparatus 300. The content managing apparatus 300 may evaluate (calculate) the user preference for each replay section for the corresponding video content by using the heart information received from the terminals 100. In addition, the content managing apparatus 300 may display the user preference for each replay section for the corresponding video content through the terminal 100. In detail, the content managing apparatus 300 transmits user preference data for each replay section calculated with respect to the content to the terminal 100, and the terminal 100 may display the user preference for each replay section of the corresponding content on a screen by using the received data. Alternatively, the terminal 100 requests the user preference data for each replay section for the content to the content managing apparatus 300, the content managing apparatus 300 transmits the corresponding data to the terminal 100 in response to the request of the terminal 100, and the terminal 100 may display the user preference for each replay section of the corresponding content on the screen by using the received data. For example, a running time gauge (alternatively, a time line) representing a running time (alternatively, a total replay length) of the video content is displayed on the screen of the terminal 100, and the replay sections of the video contents may be displayed on the running time gauge by using colors corresponding to the user preference for each replay section. For example, a replay section where the most hearts are granted is expressed by using a red color, a replay section where the least hearts are granted may be expressed by using a blue color, and other replay sections may be expressed by using light colors or other colors. A method of expressing the user preference for each replay section (alternatively, the user preference for each time) for the content will be described with reference to FIG. 3.

FIG. 3 is a diagram illustrating a method of expressing a user preference for each replay section of a content according to the exemplary embodiment of the present invention. In detail, in FIG. 3, a case where the total replay length of the content is 10 seconds and the user preference for each replay section of the corresponding content is expressed by a graph is exemplified. A horizontal axis represents a time and a vertical axis represents the number of granted hearts.

As illustrated in FIG. 3, in a replay section at about 7 seconds among the replay sections of the corresponding content, the most hearts are granted from the users. A video exemplified on the graph of FIG. 3 is a video corresponding to the replay section of about 7 seconds. The user may easily determine that the replay section of about 7 seconds is the most preferable section (a popular section) through the graph.

Meanwhile, when the content is a video content, the user may grant one heart (alternatively, a plurality of hearts) to a preferable area among areas of the video included in the video content (including a plurality of videos). For example, when a favorite animal of the user exists in a middle area of the areas of the video included in the video content, the user may grant the heart to the middle area (for example, the user grants the heart to the middle area by touching the middle area of the video). As a result, the most hearts may be granted to the most impacted area among areas of a specific video included in the video content. The content managing apparatus 300 may visually (for example, with a graph, a color, and the like) express the user preference for each video area for the video content through the terminal 100 replaying the corresponding video content. As a result, the content managing apparatus 300 may notify which area of the video included in the video content is most preferred by the users to the users.

In detail, the terminal 100 determines a video area where the hearts are granted by the user of the terminal 100 among the areas of the video included in the corresponding video content when the user of the terminal 100 grants the hearts to the video content, and may transmit information (information of the replay section where the hearts are granted, the number of granted hearts, and the like) on hearts granted by the user of the terminal 100 to the content managing apparatus 300. The content managing apparatus 300 may evaluate (calculate) the user preference for each video area for the corresponding video content by using the heart information received from the terminals 100. In addition, the content managing apparatus 300 may display the user preference for each video area for the corresponding video content through the terminal 100. In detail, the content managing apparatus 300 transmits user preference data for each video area calculated with respect to the content to the terminal 100, and the terminal 100 may display the user preference for each video area of the corresponding content on a screen by using the received data. Alternatively, the terminal 100 requests the user preference data for each video area for the content to the content managing apparatus 300, the content managing apparatus 300 transmits the corresponding data to the terminal 100 in response to the request of the terminal 100, and the terminal 100 may display the user preference for each replay section of the corresponding content on the screen by using the received data. For example, the user preference for each vide area for the video content may be expressed on the video area by using a color corresponding to the user preference for each vide area. In more detail, the terminal 100 may display a layer on which the user preference for each video area is expressed on a screen where the content is replayed. In the corresponding layer, a video area where the most hearts are granted may be expressed by using a red color, a video area where the heart is not granted may be expressed by using a transparent color, and other video areas may be expressed by using light colors or other colors. A method of expressing the user preference for each video area (alternatively, a user preference for each position) for the content will be described with reference to FIGS. 4a and 4b.

FIGS. 4a and 4b are diagrams illustrating a method of expressing the user preference for each video area of the content according to the exemplary embodiment of the present invention. In detail, FIG. 4a illustrates a specific vide included in the content, and FIG. 4b illustrates a color table which may be used for expressing the user preference for each video area.

In order to express the user preference for each video area, one of four color tables TB1, TB2, TB3, and TB4 illustrated in FIG. 4b may be used. In detail, the color table TB1 includes seven color hit maps of black, blue, cyan, green, yellow, red, and white, the color table TB2 includes five color hit maps of blue, cyan, green, yellow, and red, the color table TB3 includes a monochrome hit map, and the color table TB4 includes two color gradients (for example, blue toward the left side and red toward the right side).

In FIG. 4a, a case where the most hearts are granted in areas IA1, IA2, and IA3 among the areas of the video from the user is exemplified. In FIG. 4a, a case where the color table TB2 is used among the color tables TB1 to TB5 illustrated FIG. 4b is exemplified. Particularly, the center of the video areas IA1, IA2, and IA3 receives the most hearts to be expressed by the rightmost color (red) of the color table TB2, and peripheral areas gradually receive less hearts than the center to be expressed by more left colors (for example, yellow, green, and blue) based on the red of the color table TB2 toward the peripheral areas from the center. The center of the video area IA4 receives less hearts than the center of the video areas IA1, IA2, and IA3 to be expressed by yellow, and peripheral areas from the center may be gradually expressed by more left colors (for example, green and blue) based on the yellow of the color table TB2 toward the peripheral areas from the center. An area which receives no hearts at all among the areas of the video is transparently expressed (that is, the corresponding video area is displayed as it is). The user may easily determine that the video areas IA1, IA2, and IA3 are the most preferable areas (popular areas) through the user preference for each video area displayed on the screen of the terminal 100 as illustrated in FIG. 4a.

Meanwhile, the method of expressing the user preference for each replay section described above and the method of expressing the user preference for each video area (alternatively, the user reference for each position) may be used together. That is, the terminal 100 may display the user preference for each replay section, and the user preference for each video area for the content may be displayed.

The methods of expressing the user preference for each replay section and the user preference for each video area by the content managing apparatus 300 will be described again with reference to FIG. 5.

The content managing apparatus 300 receives replay section preference information and video area preference information for the content (S310). The replay section preference information may include information on the replay section where the hearts are granted, the number of granted hearts, and the like, and the video area preference information may include information on the video area where the hearts are granted, the number of granted hearts, and the like.

The content managing apparatus 300 evaluates (calculates) preference for each replay section and preference for each video area of the corresponding content by using the replay section preference information and the video area preference information received from the plurality of terminals 100 with respect to the corresponding content (S320).

The content managing apparatus 300 may control the terminal 100 so as to display the preference for each replay section and the preference for each video area of the corresponding content on the screen of the terminal 100 (S330).

Meanwhile, when the user touches the video of the corresponding content (that is, touches the screen of the terminal 100) to grant the heart while the content is replayed, the terminal 100 senses the intensity of pressure applied to the area touched by the user, and may determine a size of the preference (the number of hearts) granted to the corresponding content by the user according to the sensed intensity of the pressure. The terminal 100 transmits the determined size of the preference (the number of hearts) to the content managing apparatus 300. For example, when the user strongly presses the screen of the terminal 100 on which the content is replayed, a visual effect that the hearts are continuously granted to the corresponding content and the hearts are continuously emitted may be caused. Meanwhile, when the terminal 100 does not have a pressure sensing function, an area (width) of the area touched by the user is sensed and the user may determine the size of the preference (the number of hearts) granted to the corresponding content according to the sensed area (width). For example, when the user strongly presses the screen of the terminal 100 on which the content is replayed, an area where a user's finger touches the screen of the terminal 100 is increased, and when the user weakly presses the screen, the area is decreased. The terminal 100 senses the area of the touch area, and as a result, the same effect as sensing the pressure intensity of the touch area may be caused.

Meanwhile, the content managing apparatus 300 controls the lifespan of the content according to the user preference (for example, the number of hearts) granted to the content. For example, the content managing apparatus 300 may increase or decrease the lifespan of the corresponding content based on the user preference for each replay section granted to the content or the user preference for each video area granted to the corresponding content.

3. Method of managing popular content and method of recommending content

The content managing apparatus 300 may calculate a speed (hereinafter, a 'preference speed') representing the degree that the user preference for the content varies with time, and an acceleration (hereinafter, a 'preference acceleration') representing the degree that the preference speed varies with time. The preference speed represents an instantaneous change rate of the preference for the content, and the preference acceleration may be calculated by differentiating the preference speed.

The content managing apparatus 300 may calculate an average preference change rate of all the contents or an average preference change rate of a hash tag for each subject by using the preference speed and the preference acceleration of each content. The content managing apparatus 300 may measure which preference speed of any content or any hash tag is currently highest based on the average preference change rate of all the contents or the average preference change rate of the hash tag for each subject. The content managing apparatus 300 may use the measured result as an index capable of measuring which content/hash tag (for example, a rank of the popular content) is currently most preferable by the users or which content/hash tag (for example, a rank of the popular content) is currently most issued.

3.1. Management of popular content ranking and popular issued ranking

An index which is the reference of the ranking management may be used by a method described below.

3.1. 1. Server load optimization

A data processing server (for example, the content managing apparatus 300) may optimize a load of the server by using a minimum ranking condition (a minimum ranking for entering a league) based on a quartile or a decile representing a comparative position. In detail, the content managing apparatus 300 does not perform an operation of calculating the preference speed and the preference acceleration and an operation of predicting a recommended content by using an algorithm (for example, a regression analysis algorithm, a collaborative filtering algorithm, or the like) with respect to all the uploaded content. That is, the content managing apparatus 300 priorly selects (filters) contents capable of belonging to a ranking (a popular content candidate group) among all of the contents, and applies a complicated algorithm (for example, a regression analysis algorithm, a collaborative filtering algorithm, or the like) to the contents belonging to the ranking (the popular content candidate group). As a result, the content managing apparatus 300 may reduce the load. For example, the content managing apparatus 300 first calculates only the preference speed of each content and includes only the content of which the preference speed is more than the reference value among all of the contents in the popular content candidate group. In addition, the content managing apparatus 300 may calculate a final ranking of the content which may be preferred by the user among the contents included in the popular content candidate group by using the preference speed, the preference acceleration, and the analysis algorithm (for example, a regression analysis algorithm, a collaborative filtering algorithm, or the like) of the contents included in the popular content candidate group.

3.1. 2. Steady seller management

As described above, in the lifespan management of the content, the deletion of the content is included, but it is required to particularly manage the content receiving the user preference having a predetermined value or more. The content managing apparatus 300 may maintain the content receiving the user preference having a predetermined value or more as a steady seller so as to be continuously usable without deleting even though the lifespan of the corresponding content is expired.

In detail, the content managing apparatus 300 may determine whether the corresponding content is a content requiring the particular management by using the number of hearts granted to the content. For example, the content managing apparatus 300 may classify the corresponding content as a particular management category when the number of hearts granted to the content is more than a predetermined value (for example, one million). The content managing apparatus 300 may manage the corresponding content to not be dissipated even though the heart is not granted to the content included in the particular management category any more.

3.1. 3. Management of popular content

The content managing apparatus 300 may manage a popular content for a familiar period (a day, a week, a month, and the like) in life such as "a popular content today" or "a popular content this week".

In detail, the content managing apparatus 300 may manage the popular content by combining a speed (a preference speed) at which the hearts are accumulated in the content and an acceleration (a preference acceleration) at which the speed is changed based on an absolute time (for example, a day, a week, a month, and the like). A method of managing the popular content by the content managing apparatus 300 will be described with reference to FIG. 6.

The content managing apparatus 300 calculates a preference speed and a preference acceleration of each content (S410).

The content managing apparatus 300 selects (filters) a content (a popular content) to be included in the popular content candidate group by using the preference speed and the preference acceleration of each content. In detail, the content managing apparatus 300 may calculate an interest value of each content by using the following Math Figure 1.

In Math Figure 1, V_i,HOT represents an interest value of an i-th content among all of the contents, S_i represents a preference speed of the i-th content, AC_i represents a preference acceleration of the i-th content, S_avr represents an average preference speed of all of the contents, and AC_avr represents an average preference acceleration of all of the contents. In addition, W_s represents a weight value applied to a speed, and W_ac represents a weight value applied to an acceleration. W_s and W_ac may be the same as or different from each other and may be changed. For example, W_s and W_ac may be 0.5.

The content managing apparatus 300 may select the popular content by using the interest value V_i,HOT of each content. For example, the content managing apparatus 300 may include the content of which the interest value V_i,HOT is more than a predetermined value as the popular content in the popular content candidate group. The contents included in the popular content candidate group may be arranged in order of higher interest values V_i,HOT.

The content managing apparatus 300 may predict the most interested (preferable) content of the user among the contents (popular contents) included in the popular content candidate group by using a preference speed, a preference acceleration, an average preference speed, an average preference acceleration, a standard deviation, a regression analysis algorithm, a collaborative filtering algorithm, and the like (S430).

The content managing apparatus 300 arranges the predicted contents and may recommend some or all of the predicted contents to the user. The content managing apparatus 300 may control the terminal 100 so that a list of the recommended contents is displayed on the screen of the terminal 100 (S440). As a result, the content managing apparatus 300 may accurately predict the content suitable for the content preference tendency of each user and recommend the predicted content to the user.

3.2. Method of recommending content

A method of recommending a content which may be preferred by the user by the content managing apparatus 300 will be described with reference to FIG. 7.

First, the content managing apparatus 300 forms a pool (alternatively, a popular content candidate group) by collecting contents satisfying a minimum interest condition among all of the contents (S510). In detail, the content managing apparatus 300 may select contents to be included in the popular content candidate group by using a preferable speed and a preference acceleration of each content, as described above. The content managing apparatus 300 may rank the contents included in the popular content candidate group.

The content managing apparatus 300 may detect information on a terminal accessing the content managing apparatus 300. Here, the information on the terminal may include a resolution, a file format, a display size, a hardware specification, or an operating system (OS) which are supported by the terminal, and the like. The content managing apparatus may determine a capacity of the corresponding content by using the detected terminal information. The content managing apparatus 300 determines a content which is difficult execute (for example, replayed and displayed) by the corresponding terminal among the contents included in the popular content candidate group by using the detected terminal information, and excludes the corresponding content from the popular content candidate group. The content managing apparatus 300 may rank the contents included in the popular content candidate group again. Meanwhile, the content managing apparatus 300 may determine a content excluded from the popular content candidate group by using the preferable speed and the preferable acceleration after first selecting the contents to be included in the popular content candidate group by using the information of the terminal accessing the content managing apparatus 300.

The content managing apparatus 300 determines which user group the user of the terminal accessing the content managing apparatus 300 belongs to (S520). In addition, the content managing apparatus 300 predicts the content which may be preferred by the user of the terminal among the contents included in the popular content candidate group by using a recommended algorithm corresponding to the user group to which the user of the terminal belongs (S530).

In detail, the content managing apparatus 300 may determine a user group to which the user of the terminal belongs based on a history of using a service (hereinafter, a 'first service') provided by the content managing apparatus 300. In more detail, the content managing apparatus 300 may determine whether the user of the terminal is subscribed to the first service (user registration), the login number of times using the first service of the user of the terminal, whether the user of the terminal performs a page view for the first service (hovering), whether the user of the terminal uses the first service and grants the hearts to the content, and whether the user of the terminal directly uploads the content, and determine which user group the user of the terminal belongs to by using this information. Alternatively, the content managing apparatus 300 may classify the user of the terminal as a registered user, an ever-active user, a recent 1 week user, an everyday user, and the like based on the history information of using the first service by the user of the terminal. Hereinafter, a case where the user of the terminal belongs to any one of the existing user group and a new user group will be described as an example in the exemplary embodiment of the present invention.

The content managing apparatus 300 may classify a user using most of the first service as the existing user group and classify a user if not so as the new user group. For example, the content managing apparatus 300 may calculate the degree that the user of the terminal uses the first service based on the history information when the first service is used by the user of the terminal. In addition, the content managing apparatus 300 may determine that the corresponding user belongs to the existing user group when a calculated service use degree is more than a predetermined value, and determine that the corresponding user belongs to the new user group when the calculated service use degree is less than a predetermined value.

The content managing apparatus 300 may predict the content which may be preferred by the user of the terminal by using a collaborative filtering algorithm when the user of the terminal belongs to the existing user group. That is, the content managing apparatus 300 may predict the preference for the content which is still not viewed by the corresponding user (still not replayed) by using the collaborative filtering algorithm based on the content experience of the user. The content managing apparatus 300 may predict the content which may be preferred by the user of the terminal by using a regression analysis algorithm when the user of the terminal belongs to the new user group. That is, the content managing apparatus 300 may infer the preference for the content which is still not viewed by the corresponding user by using the regression analysis algorithm based on a basic attribute of the user of the terminal.

The content managing apparatus 300 may control the terminal 100 so that a list of the contents predicted in step S530 is displayed on the screen of the terminal 100 (S540). The content managing apparatus 300 may rank the contents predicted in step S530 and recommend high ranked contents among the ranked contents to the user of the terminal.

3.2. 1. Processing of regression analysis engine

When the content managing apparatus 300 uses a logistic regression analysis algorithm as the regression analysis algorithm, the logistic regression analysis algorithm may be defined as the following Math Figure 2.

In Math Figure 2, p represents a probability that the user of the terminal grants the hearts to the content, and X_j (1≤j≤k) is a variable representing an attribute of the user of the terminal. X_j may include information (for example, an age, a gender, a job position, a kind of job, a hobby, and the like of the user) provided (disclosed) when the user of the terminal joins as a member, information on the terminal (for example, a kind of terminal and OS information), or information which may be supposed based on the information. For example, X₁ may represent the age of the user of the terminal, and X₂ may represent a kind of terminal. In Math Figure 2, b_j ( 0≤j≤k) represents a weight value applied to the X_j variable, and b_j may be the same or different according to the X_j variable. B₀ represents an intercept.

The content managing apparatus 300 may predict how much the user of the terminal prefers the specific content by using the logistic regression analysis algorithm defined by Math Figure 2 above. The content managing apparatus 300 may predict a content which may be preferred by the user among the contents included in the popular content candidate group by applying the aforementioned regression analysis algorithm to the contents included in the popular content candidate group.

As an advantage of the regression analysis algorithm, the content managing apparatus 300 may predict the content which may be preferred by the user without activity information of the user. When the activity information of the user is added, prediction reliability is increased, but when the user belongs to the new user group, the activity information of the user is low, so the content managing apparatus 300 may use the regression analysis algorithm.

3.2. 2. Processing of collaborative filtering engine

The collaborative filtering algorithm used by the content managing apparatus 300 predicts a content to be preferable to the corresponding user (the first user) by using common rating data between the corresponding user (the first user) and other users when the user (for example, the first user) of the terminal accessing the content managing apparatus 300 uses the first service for a predetermined time or more (when there is the activity information). Here, the common rating data may include information on the content to which the user (the first user) and other users commonly grant the hearts (for example, attribution information of the content such as a category of the content, the number of granted hearts, all information which may be used for measuring the preference, and the like). The fact that the user of the terminal uses the first service for the predetermined time or more means that the user of the terminal performs granting the heart to the content, page-viewing, writing comments, requesting a challenge to another user by uploading the content, uploading the content for challenging the uploaded content by another user, or the like for the predetermined time or more.

The content managing apparatus 300 may predict how much the user of the terminal prefers a non-viewed content (alternatively, a content which still has not received the hearts) (how many hearts are granted) by applying the common rating data to the collaborative filtering algorithm. The collaborative filtering algorithm to which the common rating data is applied may be defined by the following Math Figure 3.

In Math Figure 3, PR_u,i is a prediction preference representing how much the user of the terminal (for convenience of description, referred to as a 'user UA1') prefers an i-th content among the contents included in the popular content candidate group (for example, how many the hearts are granted). The i-th content may be a content which is not still viewed by the user UA1 (for example, a content to which the hearts are not still granted). In Math Figure 3, PR_avr represents an average preference which is granted by the user UA1 (for example, the average number of hearts) with respect to contents preferable by the user UA1 (for example, contents granted with the hearts) (hereinafter, a 'UA1 preferable content').

In Math Figure 3, PR_d,i represents a preference (the number of granted hearts) granted to the i-th content by another user (hereinafter, a 'user UA2') having the most similar content preference tendency to the user UA1. For example, the content managing apparatus 300 may determine a content preference tendency similarity (representing how much the tendency preferring the content is similar) between each user UA3 and the user UA1 based on the preference (the number of granted hearts) granted to the UA1 preferable content by the users (hereinafter, a 'user UA3') preferring at least one of the UA1 preferable contents. The user UA2 may be a user having the highest content preference tendency similarity to the user UA1 among the users UA3.

In Math Figure 3, α₁ andα₂ represent adjustment constants applied to PR_d,i and PR_s,i in order to create meaningful data, and may have the same or different values.

In Math Figure 3, PR_s,i represents an average of the preference (for example, an average of the number of hearts) granted to the i-th content by the users having the similar content preference tendency to the user UA1. For example, PR_s,i may represent an average of the preference (for example, an average of the number of hearts) granted to the i-th content by a plurality of users of which the content preference tendency similarity to the user UA1 is more than a predetermined value among the users UA3.

In Math Figure 3, W₁ is a weight value (a weight value based on the similarity) applied to the PR_d,i, and W₂ is a weight value (a weight value based on the similarity) applied to the PR_s,i. W₁ and W₂ may have the same or different values. The content managing apparatus 300 may calculate the weight values (for example, W₁ and W₂) by various methods. The content managing apparatus 300 pre-cuts (selects) the similar users to the user UA1 by using a Pearson correlation coefficient for the same content (alternatively, a content which belongs to the same category) as the content receiving the hearts by the user UA1, and may use a weighted average of an actual rating (for example, heart granting) of the similar users.

The aforementioned collaborative filtering algorithm may calculate how much the user UA1 prefers the corresponding content (a predicted preference) by calculating a weighted average of the preference (alternatively, finding the average preference) granted to the content (the content which is not still viewed by the user UA1) by the users (alternatively, a cluster) having the similar content preference tendency to the user UA1.

3.2. 3. User class

The pattern in which the user grants the heart may vary depending on the user. In detail, a time (hereinafter, a 'heart granting time') for the user to grant the heart to the contents may vary depending on the user. For example, any user may tend to determine whether to grant the heart by viewing only a front part of the contents, while another user may tend to determine whether to grant the heart by viewing entirely up to an end part of the contents.

Further, since the other user intends to follow the fashion (alternatively, trend), the other user may determine whether to grant the heart by considering whether the contents have been granted with many hearts or whether the hearts are rapidly accumulated in the contents.

The content managing apparatus 300 may classify the users as an early adaptor, a general mass, or a follower based on a preference grant pattern (e.g., a heart granting time pattern, and the like). In detail, the content managing apparatus 300 may classify the users as the early adaptor, the general mass, or the follower based on a time for each user to grant the heart to the corresponding contents from the time (an opened time) when the contents are uploaded. For example, when the heart granting time of any user is equal to or less than a first reference value, the corresponding user may be classified into the earlier adaptor, when the heart granting time of another user is larger than the first reference value and is equal to or less than a second reference value, the corresponding user may be classified into the general mass, and when the heart granting time of any user is larger than the second reference value, the corresponding user may be classified into the follower.

The content managing apparatus 300 may use the user class (e.g., the early adaptor, the general mass, and the follower) as a factor of a contents recommendation algorithm (e.g., a regression analysis algorithm, a cooperation filtering algorithm, and the like). In detail, the content managing apparatus 300 may apply the heart granting time of the user of the terminal, the user class (e.g., the early adaptor, the general mass, and the follower) to which the user of the terminal belongs, and an average heart granting time of the user class to which the user of the terminal belongs to the contents recommendation algorithm.

3.2. 4. Method for recommending contents by using market basket analysis algorithm

The content managing apparatus 300 may predict contents which may be preferred by the user by using the market basket analysis algorithm.

In detail, the content managing apparatus 300 may generate data regarding all contents to which the user of the terminal grants the hearts during a predetermined period (e.g., 1 month). In addition, the content managing apparatus 300 may derive a content preference rule by applying the market basket analysis algorithm to the data. For example, the content preference rule may include "Most users who grant the heart by viewing content A grant the heart even to heart B.", "Users who grant the hearts contents A and C tend to grant the heart even to content E.", and the like.

The content managing apparatus 300 may calculate three indexes (support, confidence, and lift) for the content preference rule by using the following Math Figure 4.

In Math Figure 4, the content preference rule X=>Y means that a user who grants the heart to content X tends to grant the heart to content Y. The indexes, support, confidence, and lift are indexes for the content preference rule X=>Y.

In Math Figure 4, Support(X, Y) represents the number of users who grant the heart to both contents X and Y. In Math Figure 4, frq(X, Y) represents the number of users who grant the heart to both contents X and Y during a predetermined period (e.g., recent 1 month). In Math Figure 4, N represents the number of all users who grant the heart to at least one content during a predetermined period.

In Math Figure 4, Confidence represents a probability that the user who grants the heart to content Y will be included in users who grant the heart to content X. In Math Figure 4, frq(X) represents the number of users who grant the heart to content X during a predetermined period.

In Math Figure 4, Lift represents a ratio of the number of users who grant the heart to content Y among the users who grant the heart to content X and the number of users who grant the heart to content Y. In Math Figure 4, Support(X) may be defined as frq(X)/N. In Math Figure 4, Support(Y) may be defined as frq(Y)/N. frq(Y) represents the number of users who grant the heart to content Y during a predetermined period.

The content managing apparatus 300 may determine whether to use the content preference rule X=>Y for contents recommendation by combining the indexes Support, Confidence, and Lift calculated by Math Figure 4. When the content managing apparatus 300 use the content preference rule X=>Y, the content managing apparatus 300 may recommend content Y to the user who grants the heart to content X.

The content managing apparatus 300 may calculate three indexes (support, confidence, and lift) by using Math Figure 4 with respect to the number of all cases (e.g., content X => content Z, content X => content B, content X => content H, and the like). In addition, the content managing apparatus 300 may rank the content preference rules (e.g., content X => content Z, content X => content B, content X => content H, and the like) by using three indexes (support, confidence, and lift). In addition, the content managing apparatus 300 may predict contents which may be preferred by the user of the terminal by using a higher rule among the ranked content preference rules.

In detail, the content managing apparatus 300 applies the market basket analysis algorithm to a popular contents candidate group to predict contents which may be preferred by the user (contents to which the user may grant the heart) among contents included in popular contents candidate group.

Meanwhile, the market basket analysis algorithm is described by using a case in which the content managing apparatus 300 generates the content preference rule with a content to which the hearts of a predetermined value or more are granted as a basic unit, but this is just an example. The content managing apparatus 300 may generate the preference rule by using a category by a hash tag instead of the contents or a category which is pre-classified and managed as the basic unit.

3.2.5. Method for selecting optimal contents recommendation algorithm

As described above, when the content managing apparatus 300 predicts the contents which may be preferred by the user, the content managing apparatus 300 may rank the corresponding contents based on a prediction preference indicating how much the user prefers each of the corresponding contents. In addition, the content managing apparatus 300 may recommend some higher contents (e.g., 10 higher contents) among the ranked contents to the corresponding user.

Meanwhile, the content managing apparatus 300 may receive feedback regarding a recommendation result from the corresponding user. Herein, the feedback regarding the recommendation result may include recommended contents which the corresponding user actually prefers or the recommended contents are contents which the corresponding user does not prefer. The content managing apparatus 300 may correct the contents recommendation algorithm by using the feedback regarding the recommendation result.

Meanwhile, the content managing apparatus 300 may simultaneously operate (operate as a pilot) a plurality of contents recommendation algorithms to determine which algorithm of the plurality of contents recommendation algorithms is more effective and accurate. For example, the content managing apparatus 300 may recommend the contents to the users by using each of two contents recommendation algorithms during a predetermined period, and determine which algorithm of two contents recommendation algorithms is more accurate based on the number of contents selected (e.g., clicked, touched, and the like) by the corresponding users or to which the corresponding users grant the hearts among the recommended contents.

The content managing apparatus 300 may recommend the contents to the user, ask whether the corresponding user likes the recommended contents, and receive a response thereto from the corresponding user. For example, the content managing apparatus 300 may ask how much the user is satisfied with the contents recommendation. The content managing apparatus 300 may correct the contents recommendation algorithm or select a more accurate contents recommendation algorithm based on the response of the user.

The content managing apparatus 300 may randomly divide the users into multiple groups and apply different content recommendation algorithms to the multiple groups, respectively. In addition, the content managing apparatus 300 may determine which contents recommendation algorithm contributes more to improvement of the trend of the users or which contents recommendation algorithm has higher recommendation accuracy. Moreover, the content managing apparatus 300 may select one of the multiple contents recommendation algorithms based on the determination result.

FIG. 8 is a diagram illustrating a content managing apparatus 300 according to another exemplary embodiment of the present invention. The content managing apparatus 300 includes a speed calculation unit 390, a candidate group generation unit 391, a detection unit 392, a prediction unit 393, and a group determination unit 394.

The speed calculation unit 390 calculates a preference speed and a preference acceleration of contents. In detail, the speed calculation unit 390 may perform procedures, functions, and methods associated with an operation of calculating the preference speed and the preference acceleration described in the specification.

The candidate group generation unit 391 selects contents to be included in a popular contents candidate group by using the preference speed and the preference acceleration of the contents. Further, the candidate group generation unit 391 may select contents to be excluded from the popular contents candidate group by using information on a terminal that accesses the content managing apparatus 300. In detail, the candidate group generation unit 391 may perform procedures, functions, and methods associated with generation of the popular contents candidate group described in the specification.

The detection unit 392 detects information (e.g., a resolution, a file format, a display size, a hardware specification, OS information, and the like which are supported by the terminal) of the terminal that accesses the content managing apparatus 300. In detail, the detection unit 392 may perform procedures, functions, and methods associated with the detection of the terminal information described in the specification.

The prediction unit 393 predicts contents which may be preferred by the user. In detail, the prediction unit 393 may perform procedures, functions, and methods associated with the prediction of the contents described in the specification.

The group determination unit 394 determines to which user group the user of the terminal which accesses the content managing apparatus 300 belongs. In detail, the group determination unit 394 may perform procedures, functions, and methods associated with the determination of the user group described in the specification.

Meanwhile, the content managing apparatus 300 illustrated in FIG. 8 may further include some or all of the components 310 to 370 and 380 to 382 of the content managing apparatus 300 illustrated in FIG. 2 to perform the operations, functions, and method of the content managing apparatus 300 described in the specification.

The components 390 to 394 of the content managing apparatus 300 may be controlled by the processor 380.

Meanwhile, some of the functions of the content managing apparatus 300 described in the specification may be performed by the terminal 100.

FIG. 9 is a diagram illustrating a constitution of a content server 200.

The content server 200 includes a memory 210, a processor 220, and a wired/wireless communication device 230.

The processor 220 may be constituted to implement procedures, functions, and methods related to the content server 200 described in this specification.

The memory 210 is connected with the processor 220 and stores various information related to an operation of the processor 220.

The wired/wireless communication device 230 is connected with the processor 220 and transmits or receives a wired signal or a wireless signal.

FIG. 10 is a diagram illustrating a constitution of a content database apparatus 400.

The content database apparatus 400 includes a memory 410, a processor 420, and a wired/wireless communication device 430.

The processor 420 may be constituted to implement procedures, functions, and methods related to the content database apparatus 400 described in this specification.

The memory 410 is connected with the processor 420 and stores various information related to an operation of the processor 420.

The wired/wireless communication device 430 is connected with the processor 420 and transmits or receives a wired signal or a wireless signal.

Meanwhile, according to the exemplary embodiments of the present invention described above, contents which survive for a long time may be regarded as contents having high social preference, and various feedback including a comment and the like may be present as a social concern corresponding thereto. The feedback may be regarded as life force and an experience point of the contents, and the experience point grants a value to the corresponding contents. Consequently, the contents have the value, and as a result, the contents have a meaning as one life which lives in a digital world.

Further, according to the exemplary embodiments of the present invention described above, a lifespan as long as a time corresponding to user preference for the contents may be granted to the contents. The contents granted with the life may survive. Further, continuously, even after the life is granted, according to evaluation of the users, a quantitative change (a change in content life cycle) of an absolute time may occur in the lifespan of the contents. In addition, contents (contents having low social preference) which cannot be socially selected are removed to efficiently manage and use a data storage space of the content server 200 or the content database apparatus 400.

The above-described embodiments of the present invention may be created by a computer executable program and implemented in a general use digital computer which operates the program using a computer readable recording medium. The computer readable recording media include storage media such as magnetic storage media (for example, a ROM, a floppy disk, a hard disk, and the like), optical reading media (for example, a CD-ROM, a DVD, and the like), and a carrier wave (e.g., transmission through the Internet).

While this invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

A method for managing a content of a content managing apparatus, the method comprising:

receiving first information representing which replay section of the content a user prefers;

evaluating a user preference for each replay section for the content based on the first information;

transmitting the user preference for each replay section to a user terminal so that the user preference for each replay section is displayed by the user terminal; and

increasing or decreasing a lifespan of the content which is a period in which a user can use the content based on the user preference for each replay section.
The method of claim 1, further comprising:

receiving second information representing which area of a video included in the content a user prefers;

evaluating a user preference for each video area for the content based on the second information; and

transmitting the user preference for each video area to the user terminal so that the user preference for each video area is displayed by the user terminal,

wherein the increasing or decreasing of the lifespan of the content

comprises increasing or decreasing of the lifespan of the content based on the user preference for each replay section and the user preference for each video area.
The method of claim 1, wherein the user terminal displays replay sections of the content on a running time gauge representing a running time of the content by using a color corresponding to the user preference for each replay section.
The method of claim 2, wherein the user terminal displays the user preference for each video area on the video included in the content by using a color corresponding to the user preference for each video area.
The method of claim 2, wherein the user terminal determines how much a first user prefers the content based on at least one of the intensity of pressure applied to a touched area and an area of the touched area when the area of the video included in the content is touched by the first user and generates at least one of the first information and the second information by using the determined result.
An apparatus for managing a content, the apparatus comprising:

a calculation unit configured to calculate a speed representing the degree that a user preference for a content varies with time and acceleration of the speed;

a candidate group generation unit configured to include contents in the candidate group, in which a first value calculated by at least one of the speed and the acceleration is more than a first threshold value;

a prediction unit configured to predict a content which may be preferred by a user among the contents included in the candidate group by using at least one of regression analysis and collaborative filtering, the speed, and the acceleration; and

a processor configured to control the calculation unit, the candidate group generation unit, and the prediction unit.
The apparatus of claim 6, further comprising

a lifespan control unit controlled by the processor and configured to increase or decrease a lifespan of each predicted content based on the user preference for each predicted content.
The apparatus of claim 7, wherein

the prediction unit ranks the predicted contents and selects a predetermined number of contents among the ranked contents, and

the selected contents are displayed by a terminal.
The apparatus of claim 7, wherein the lifespan control unit controls the lifespan of a first content so as to not delete the first content over time when the user preference for the first content among the predicted contents is more than a second threshold value.
The apparatus of claim 6, wherein the candidate group generation unit calculates the first value for each of a plurality of contents by using the following Equation 1, and includes a content of which the first value is more than the first threshold value among the plurality of contents in the candidate group:

[Equation 1]

First value = (V_i * w_v + AC_i * w_ac)/(V_avr * w_v + AC_avr * w_ac)

(V_i: a speed of an i-th content among the plurality of contents, w_v: a weight value applied to a speed, AC_i: an acceleration of the i-th content, w_ac: a weight value applied to an acceleration, V_avr: an average speed of the plurality of contents, AC_avr: an average acceleration of the plurality of contents).
An apparatus for managing a content, the apparatus comprising:

a determination unit configured to determine which group a user of a terminal belongs to among a plurality of user groups by using first data representing a history of using a service provided by the content managing apparatus in the user of the terminal when the terminal accesses the content managing apparatus;

a prediction unit configured to predict a content which may be preferred by the user of the terminal by using an algorithm corresponding to the user group to which the user of the terminal belongs among a regression analysis algorithm, a collaborative filtering algorithm, and a market basket analysis algorithm; and

a processor configured to control the determination unit and the prediction unit.
The apparatus of claim 11, further comprising:

a calculation unit controlled by the processor and configured to calculate a speed representing the degree that a user preference for a content varies with time and an acceleration of the speed with respect to each of a plurality of contents; and

a candidate group generation unit controlled by the processor and configured to include contents of which a first value calculated by using at least one of the speed and the acceleration is more than a first threshold value in the candidate group among the plurality of contents,

wherein the prediction unit predicts a content which may be preferred by the user of the terminal among the contents included in the candidate group by using an algorithm corresponding to the user group to which the user of the terminal belongs of the regression analysis algorithm and the collaborative filtering algorithm.
The apparatus of claim 12, further comprising

a detection unit configured to detect information on the terminal,

wherein the candidate group generation unit determines a first content which cannot be replayed by the terminal among the contents included in the candidate group by using the information on the terminal and excludes the first content from the candidate group.
The apparatus of claim 11, wherein the determination unit calculates the degree that the user of the terminal uses the service based on the first data, determines that the user of the terminal belongs to a first user group of the plurality of user groups when the degree of using the service is more than a first reference value, and determines that the user of the terminal belongs to a second user group of the plurality of user groups when the degree of using the service is less than the first reference value.
The apparatus of claim 14, wherein the prediction unit calculates a first value which is an average of the preference granted by the user of the terminal with respect to a plurality of first contents which is preferred by the user of the terminal when the user of the terminal belongs to the first user group, determines a content preference tendency similarity between each of first users and the user of the terminal based on the preference granted to the plurality of first contents by the first users preferring at least one of the plurality of first contents, calculates a second value which is a preference granted to the second content by the first user having the highest content preference tendency similarity among the first users, calculates a third value which is an average of the preference granted to the second content by the first users of which the content preference tendency similarity is more than a second reference value among the first users, and predicts how much the user of the terminal prefers the second content by applying the first value, the second value, and the third value to the collaborative filtering algorithm.
The apparatus of claim 15, wherein the collaborative filtering algorithm applies the same or different weight values to the second value and the third value.
The apparatus of claim 14, wherein the prediction unit, when the user of the terminal belongs to the second user group, predicts how much the user of the terminal prefers the first content by applying user information provided for using the service by the user of the terminal, a kind of terminal, and a kind of operating system (OS) installed in the terminal to a logistic regression analysis algorithm which is the regression analysis algorithm.
The apparatus of claim 17, wherein the logistic regression analysis algorithm applies the same or different weight values to each of the user information, the kind of terminal, and the kind of OS.
The apparatus of claim 11, wherein the prediction unit calculates a first value which is the number of users granting the preference to at least one of a plurality of contents for a first period, calculates a second value which is the number of users granting the preference to both a first content and a second content among the plurality of contents for the first period, calculates a third value which is the number of users granting the preference to the first content for the first period, calculates a fourth value which is the number of users granting the preference to the second content for the first period, calculates a fifth value by dividing the third value by the first value, calculates a sixth value by dividing the fourth value by the first value, and applies a first index value by dividing the second value by the first value, a second index value by dividing the second value by the third value, and a third index value by dividing the first index value by a multiple of the fifth value and the sixth value to the basket analysis algorithm to predict how much the user of the terminal prefers the second content when the user of the terminal prefers the first content.