KR20090080857A - System And Method For Recommending Contents Based On Social Network And Contents Providing Server - Google Patents

Abstract

A system for recommending contents based on the social network and a contents offering system are provided to induce the spontaneous participation of a plurality of Internet users. A system for recommending contents based on the social network comprises one or more user terminals(100-1 ~ 100-N) and a contents providing server(300). The user terminals produce the contents information and user data. The contents information comprises data about the grade which is given to the read contents. The user data comprises an identification number. The contents providing server receives the user data from the user terminals, determines the ranking of each content which obtains at least one grade, produces the first high-ranking contents, and then determines the ranking of each user which gives the grade to the contents by using the weighted value corresponding to each ranking of contents and the grade.

Description

System and Method for Recommending Contents Based On Social Network And Contents Providing Server}

The present invention relates to a social network-based content recommendation system, method and content providing server, and in particular, by repeating the process of calculating only the information within a certain rank based on the rating given by the user terminal users with high reliability content The present invention relates to a social network-based content recommendation system and method for providing a terminal and a content providing server.

In the modern information society, computers and the Internet have emerged as essential tools for new communication as a medium for transmitting various information. Accordingly, computer and internet users are required to be self-protected from the various injuries and adverse functions of informatization contrary to the convenience of informatization, and to quickly adapt to the rapid change of environment and to continuously change new ones.

The Internet is rapidly changing from Web 1.0 to Web 2.0. Existing Web 1.0 collects and displays information on the Internet. In this Web 1.0, as the amount of information increases exponentially, an excessive amount of information occurs, which causes users not only to spend too much time searching for the necessary information but also to find the information they need.

In contrast, Web 2.0 builds content using the voluntary participation of users and their collective intelligence, and provides information based on it, thereby providing more accurate information that users want.

However, the web page still provides information using a specific editor, search engine, recommendation engine, etc. of the web page rather than a large number of users participate in providing information. For example, when a news list is displayed on the main screen of the portal site, the news list specified by the editor of the portal site is displayed. This is contrary to the Web 2.0 perspective.

Accordingly, users want to be able to participate voluntarily in accordance with the viewpoint of Web 2.0 and to receive accurate information that users want.

An object of the present invention is to build content using voluntary participation and collective intelligence of a plurality of Internet users, and to repeat the process of calculating only the information falling within a certain rank based on a rating given by the user terminal. It is to provide a social network-based content recommendation system to provide more accurate.

Another object of the present invention is to build content using voluntary participation and collective intelligence of a plurality of Internet users, and repeat the process of calculating only the information falling within a certain rank based on a rating given by the user terminal. To provide a content recommendation method based on social networks to provide them more accurately.

Another object of the present invention is to build content using voluntary participation and collective intelligence of a plurality of Internet users, and repeat the process of calculating only the information falling within a certain rank based on a rating given by the user terminal, thereby increasing reliability. It is to provide a social network based content providing server for providing contents more accurately.

To this end, the present invention provides a social network-based content recommendation system and method and a content providing server as follows.

The social network-based content recommendation system according to the present invention includes at least one user terminal and a content providing server. The user terminal generates user data including content information and identification number including data on the rating given to the browsed content. After receiving the user data from the user terminals, the content providing server ranks at least one scored content by score and calculates first upper content items within an upper predetermined rank, and ranks the higher contents. Determine the ranking of users who have given a rating to the higher contents by using the weights and the ratings given by the users, calculate the top users within a certain rank among the determined rankings of users, and the weight and the top users according to the ranking of the top users. The rankings of the contents are determined by using the ratings given by the third and upper second contents that fall within a predetermined rank among the determined contents.

The content providing server extracts a rating of the contents from the received user data, and each of the first upper content from the calculated first higher contents, the calculated upper users, and the calculated second upper contents, respectively. Generate a list, an upper user list, and a second upper content list.

The content providing server may store the user data, the ranking of the contents, the ranking of the users, the first upper content list, the upper user list, and the second upper content list, and include them in the second upper content list. At least one of the contents, the second upper content list, and the upper user list to the user terminal, and generates a web page in which the second upper content is disposed in an area having high recognition rate, thereby generating the web page. Send to the terminal.

In addition, the content providing server detects contents viewed by the upper users, generates an opinion list including the detected contents, and transmits the opinion list to the user terminal.

The content information includes identification information of the content and rating information of the user for the content.

In accordance with another aspect of the present invention, there is provided a method for recommending a content based on a social network, including: receiving, by a content providing server, at least one user data including content information of which a user terminal has been rated and user information about the user terminal; Ranking the at least one rated content by score and calculating first upper content within an upper predetermined rank; Determining a ranking of users who have given a rating to higher contents by using a weight according to the ranking of the higher contents and a rating given by the user, and calculating upper users within a predetermined rank among the determined rankings of the users; And determining the ranking of the contents using weights according to the rankings of the upper users and the ratings given by the upper users, and calculating second upper contents within a predetermined rank among the determined rankings of the contents.

The social network-based content recommendation method of the present invention includes the steps of storing the user data, the ranking of the contents, the ranking of the upper users; Storing an opinion list that is a list of contents viewed by the upper-level users; The method may further include transmitting the opinion list to the user terminal.

The social network-based content recommendation method of the present invention comprises the steps of: generating a second upper content list including the second higher contents; And transmitting the second upper content list to the user terminal.

The social network-based content recommendation method of the present invention includes the steps of: disposing the second higher-level contents in an area of high recognition rate among areas of a web page for transmitting to the user terminal; And transmitting the web page to the user terminal.

The social network-based content providing server according to the present invention receives user data from user terminals and extracts the ratings of the contents from the user data. A content reliability score determiner configured to generate a first upper content list by calculating first upper content items within a predetermined high rank, and using the weights according to the ranks of the first upper content and the ratings assigned by the users. A user evaluation score determination unit that determines a ranking of users who have given a rating to the user, and calculates top users within a predetermined rank among the determined users' rankings to generate a top user list, weights according to the ranks of the top users, and the top users Ratings Used to determine the order of the content, and a high reliability content selector, for generating a second high-content list to yield a second parent content that appeals within certain of the content of the determined priority ranking.

The apparatus may further include an evaluation score log storage configured to store the user data, the ranking of the contents, the ranking of the users, the first upper content list, the upper user list, and the second upper content list.

The apparatus may include a content providing unit configured to transmit at least one of contents included in the second upper content list, the second upper content list, and the upper user list to the user terminal.

In addition, the content providing unit generates a web page in which the second higher-level contents are arranged in an area having a high recognition rate, and transmits the web page to the user terminal.

The apparatus may further include an opinion leader selecting unit that detects contents viewed by the upper users.

In addition, the opinion leader selecting unit generates an opinion list including contents read by the detected upper-level users, and the content providing unit transmits the opinion list to the user terminal.

In addition, the content information includes identification information of the content and rating information of the user for the content.

According to the present invention, the social network-based recommendation system, method, and content providing server may induce voluntary participation of a plurality of Internet users, thereby exchanging more active opinions.

In addition, the social network-based recommendation system, method, and content providing server induce voluntary participation of a large number of Internet users, thereby making it possible to build content using their collective intelligence.

In addition, social network-based recommendation system and method and content providing server can induce voluntary participation of a large number of Internet users and build contents with their collective intelligence to provide more accurate and reliable information that users want. .

In addition, the social network-based recommendation system, method, and content providing server assigns a rating to content constructed by voluntary participation of a plurality of Internet users, and repeats a process of calculating only information that falls within a predetermined rank based on the given rating. By doing so, it is possible to extract highly reliable contents.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, in describing the embodiments, descriptions of technical contents that are well known in the technical field to which the present invention belongs and are not directly related to the present invention are omitted. This is to more clearly communicate without obscure the core of the present invention by omitting unnecessary description.

On the other hand, in the accompanying drawings, some components may be exaggerated, omitted, or schematically illustrated, and the size of each component may not fully reflect the actual size. Like reference numerals refer to like or corresponding elements throughout the accompanying drawings.

1 is a diagram schematically showing a social network-based content recommendation system according to an embodiment of the present invention.

Referring to FIG. 1, a social network-based content recommendation system includes at least one user terminal 100-1 to 100 -N, N is a natural number, hereinafter, 100, a communication network 200, and a content providing server 300. It is configured by.

The user terminal 100 is connected to the content providing server 300 through the communication network 200, receives the content from the content providing server 300 to display. After reading the content displayed through the user terminal 100, the user gives a rating to the content. In this case, the user terminal 100 generates user data (User_Data) including content information (Contents_Info), a rating given to the corresponding content, and user information (User_Info). Here, the contents information (Contents_Info) may include a content ID or a content address in order to distinguish the contents read and rated by the user terminal 100. The user information (User_Info) includes a number of the user terminal 100, an IP address of the user terminal 100, a user ID, or the like, so that the content providing server 300 distinguishes the user terminal 100 that has given a rating to the content. can do. The user terminal 100 transmits the above-described user data (User_Data) to the content providing server 300 through the communication network 200 in order to transmit the result of assigning a rating to a specific content to the content providing server 300.

The user terminal 100 includes a wired communication module or a wireless communication module according to the communication method of the communication network 200, a display unit for displaying the content, an input unit for the user to input the rating for the content after viewing the content, A memory capable of storing the content may further include a controller for controlling each component of the user terminal 100. The user terminal 100 having such a configuration is connected to the content providing server 300 via the communication network 200 using the communication module under the control of the controller. To this end, the memory may store a web browser application, and the controller of the user terminal 100 may load the web browser stored in the memory and access the content providing server 300 in response to a key input signal from the input unit. The user terminal 100 may output a web page provided by the content providing server 300 to a display unit using a web browser. As a result, the display unit of the user terminal 100 outputs various contents information (Contents_Info) provided by the content providing server 300. When a user browses a specific content among the contents output on the display unit and assigns a rating to the content, the user terminal 100 includes user data including contents information (Contents_Info) and user information (User_Info) of the selected content. Create (User_Data). The generated user data User_Data is transmitted to the content providing server 300 via the communication network 200 under the control of the controller.

Through the above-described process, the user terminal 100 provides the content providing server 300 with a preference and reliability for the content, and may be recognized as a high ranker for content selection from the content providing server 300 through an appropriate process. . As a result, the user terminal 100 receives a web page from the content providing server 300 that can preferentially check contents preferred by a plurality of user terminals and gives high ratings, or recommends a list of reliable contents to other user terminals. You may. Detailed process for this will be described later.

On the other hand, the user terminal 100 is a terminal that can be connected to the content providing server 300 through the communication network 200, a mobile communication terminal, a digital broadcasting terminal, a personal digital assistant (PDA), a smart phone terminal and It will be apparent that all the information and communication devices such as UMPC (Ultra Mobile Personal Computer) and multimedia devices, and applications thereof.

The communication network 200 connects the user terminal 100 and the content providing server 300. The communication network 200 may be applied to any communication network such as a mobile communication network, a wireless communication network, a wired communication network, and the like, but is not limited to any particular communication network. For example, when the user terminal 100 is a desktop computer that can be connected to a wired network, the communication network 200 may be a wired Internet, and when the user terminal 100 is a mobile communication terminal, the communication network 200 The Internet may include a content providing server 300 and a mobile communication network providing a mobile communication function. The communication network 200 may be a variety of networks capable of supporting a network capable of bidirectional transmission of contents, for example, a WCDMA network, an IMT-2000 network, a GSM / PDSN network, and the like. Accordingly, the communication network 200 of the present invention is not limited to a specific network, but transmits the content of the content providing server 300 to the user terminal 100 and provides the user data (User_Data) of the user terminal 100. Any network that can form a data channel that can be transmitted to the server 300 may be any communication network.

The content providing server 300 stores the content and provides the content list and the contents of the highly reliable contents to the user terminal 100. Here, the content may be a news article, an advertisement, a shopping item of a shopping mall, etc. displayed on a web page, and may be applied to the application thereof.

The content providing server 300 provides various contents to the user terminal 100 connected through the communication network 200 and collects user data (User_Data) including information from the user terminal 100 including information for which a specific content is rated. do. The content providing server 300 may calculate reliability of specific contents based on user data (User_Data), and select a user terminal 100 having high reliability among various user terminals 100. After the selection of the content is completed, the content providing server 300 is configured to be arranged in the area of high reliability, such as the upper area of the web page or the central area of the web page, the content of high reliability content using the user terminal 100 The web page may be transmitted to the user terminal 100. In addition, the content providing server 300 is a list of the content that is viewed by the highly reliable user terminal 100 or the content that is given a high rating among the content viewed by the user terminal 100 with high reliability After generating the list, the list may be provided to other user terminals 100. Detailed configuration of the content providing server 300 will be described in more detail with reference to FIG. 2.

2 is a diagram schematically showing an internal configuration of a content providing server 300 according to an embodiment of the present invention.

2, the content providing server 300 of the present invention is a user data extraction unit 310, evaluation score log storage unit 320, content reliability score determination unit 330, user evaluation score determination unit 340 , A reliable content selection unit 350, an opinion leader selection unit 360, and a content providing unit 370. The content providing server 300 may further include a storage device for storing content.

The user data extracting unit 310 receives the user data User_Data from the user terminal 100 through the communication network 200 and stores the user data User_Data in the evaluation score log storage unit 320. That is, the user data extracting unit 310 calculates information on a rating given by the user with respect to various contents provided by the content providing server 300 and stores the information in the evaluation score log storage 320. To this end, the user data extracting unit 310 extracts contents information (Contents_Info) and user information (User_Info) included in the user data (User_Data). The user data extracting unit 310 transmits the extracted content information Contents_Info and the user information User_Info to the evaluation score log storage 320.

The evaluation score log storage 320 stores contents information (Contents_Info) and user information (User_Info) delivered by the user data extraction unit 310, respectively. In this case, the evaluation score log storage 320 may classify and store user information for each content information (Contents_Info) or classify and store content information (Contents_Info) for each user information.

The content reliability score determination unit 330 selects the higher-ranking content by using the user data User_Data, that is, the user information User_Info and the content information Contents_Info stored in the evaluation score log storage 320. The content reliability score determination unit 330 checks the specific content using the contents information (Contents_Info) to collect the sum of the ratings of the users for the specific content, extracts the rating for the corresponding content from the user information, and then extracts the ratings from the other user information. Can be summed with the extracted ratings. As a result, the content reliability score determination unit 330 may list the content information (Contents_Info) in order of the high rating content.

On the other hand, the content reliability score determination unit 330 may select the top content in the upper schedule ranking among the contents listed in the order of high rating, and generate a first top content list (Contents_List_1) for the selected top content. have. The generated first upper content list (Contents_List_1) is transmitted to the user terminal 100 via the content providing unit 370 and stored in the evaluation score log storage unit 320 and the user evaluation score determination unit 340. To pass. In this case, the first upper content list Contents_List_1 is a content list in which the rating for the initial content is initialized. In addition, the content reliability score determination unit 330 periodically generates the first upper content list (Contents_List_1) according to a predetermined period or generates a situation according to the generation of the content, so that the initial content is not necessarily the initial content. You can initialize the rating for any content you want to use.

When the user evaluation score determiner 340 receives the first upper content list Contents_List_1 from the content reliability score determiner 330, the user rating score determiner 340 ranks the top user list for the top users who fall within a predetermined rank among the users who rated the higher content. Create (User_List). In more detail, the user evaluation score determiner 340 collects information on users who have rated the contents included in the first upper content list Contents_List_1, and compares the ratings and contents that the user has assigned to the contents. The reliability score between each user is calculated using the weight corresponding to the rank. The user evaluation score determiner 340 may transmit the upper user list User_List to the user terminal 100.

For example, a calculation formula for calculating a reliability score for each user is shown in Equation 1 below.

이용자 X에 대한 신뢰도 점수 =

Confidence score for user X =

Here, m is a natural number and the total number of contents included in the first upper content list (Contents_List_1), W is a weight according to the order of the contents included in the first upper content list (Contents_List_1), and S is a user X for each content. The rating given. If no rating is given for specific content, S may be zero.

That is, the reliability scores of the users are calculated by summing the products of the user's own ratings for each content and the weight of the corresponding content. The user evaluation score determiner 340 classifies the users by the reliability scores when the reliability scores of the users are calculated. In addition, the user evaluation score determiner 340 generates a higher user list (User_List) by using users within a predetermined rank. The user evaluation score determiner 340 may transfer the generated upper user list User_List to the evaluation score log storage 320 and store the evaluation score.

The highly reliable content selecting unit 350 receives a higher user list (User_List) and contents information (Contents_Info) related to the users from the evaluation score log storage unit 320 and based on this, the second higher content list having high reliability. Create (Contents_List_2). In this case, the second upper content list Contents_List_2 is a list having information including the ratings assigned to the contents after the initialization ratings given to the contents of the first upper content list Contents_List_1. That is, the highly reliable content selection unit 350 assigns weights according to the rankings of the users included in the upper user list User_List, multiplies the weight value with the ratings assigned to the contents, and gives a credit score for the specific content. Calculate On the other hand, the highly reliable content selection unit 350 may calculate the reliability score by unifying the content to which a rating is equal to or greater than a predetermined score as a reference score. For example, a credit score may be calculated by unifying the rating to four points and multiplying the content to four points out of five points to multiply the weight value given to the corresponding content.

The highly reliable content selecting unit 350 sorts each of the contents by the reliability scores, and generates a second upper content list (Contents_List_2) including contents falling within a predetermined rank among the sorted contents. The reliable content selection unit 350 may directly transmit the second upper content list Contents_List_2 to the user terminal 100.

For example, a calculation formula for selecting the second higher contents list Contents_List_2 generated for each higher reliability score is as shown in Equation 2 below.

콘텐츠 X에 대한 신뢰도 점수 =

Confidence score for content X =

Here, m is a natural number and the total number of contents included in the second upper content list Contents_List_2, V is a weight according to the order of each user included in the upper user list User_List, and H is a rating given to each content X. (Or reference score). If a particular user has not rated a content X or has given a rating below a certain score, H may be zero.

That is, the reliability score for the content is calculated by adding up the product of the weight (W) of the user who rated the content and the rating (or reference score) assigned to the content.

As described above, the highly reliable content selecting unit 350 calculates a reliability score for each content using Equation 2 described above, and stores the calculated reliability score in the evaluation score log storage 320. In addition, the highly reliable content selecting unit 350 generates a list of these contents, that is, a second upper content list (Contents_List_2) by selecting contents falling within a predetermined rank among the contents sorted by the calculated reliability scores, Transfer to study (370).

The opinion leader selecting unit 360 selects users included in the upper group among the reliability scores of the users stored in the evaluation score log storage unit 320, that is, users included in the upper user list (User_List) as the opinion leaders. First, they generate an opinion list (Opinion_List), which is a list of contents, and transmits them to the content provider 370.

The content provider 370 sets the contents transmitted to the user terminals 100 based on the second upper content list Contents_List_2 and the opinion list Opinion_List stored in the evaluation score log storage 320, Organize web pages with When the construction of the web page is completed, the content providing unit 370 may provide a corresponding web page to the user terminals 100 requesting access.

A function of the content providing server 300 according to an exemplary embodiment of the present invention having such a configuration will be described with reference to FIG. 3.

3 is a view showing an example for explaining the function of the content providing server 300 according to an embodiment of the present invention.

Prior to the description, it will be assumed that the rating can be given from 1 point to 5 points in the following description, and 5 points is assumed to mean a relatively high score compared to 1 point. In the following description, the ratio for determining the ranking will be described as 20% or 5%, but the present invention is not limited thereto. The ratio may be arbitrarily set by the program designer.

As shown, it will be assumed that there are a total of 100 users corresponding to the user terminal 100, and that there are 100 contents that are read by the users. Users access the content providing server 300 using the user terminal 100, and the content providing server 300 transmits a web page including 100 contents to the user terminal 100. Then, the user terminal 100 browses the specific content among the 100 contents, and gives a rating to the viewed content. In addition, the user terminal 100 transmits user data (User_Data) including information on the content to which the rating is assigned, to the content providing server 300. Referring to the figure, user 1 assigns 0 points to the first content, 5 points to the third content, and 4 points to the 100th content. In addition, user 2 assigns 3 points to the second content and 5 points to the last 100th content. The 99th user assigns 0 points to the 99th content. The 100th user gives 3 points to the last content. In this manner, 100 users may give a rating to specific content viewed through the user terminal 100 and transmit user data (User_Data) including information about the content to the content providing server 300.

The content providing server 300 receives user data (User_Data), extracts information on the ratings assigned to the contents, and calculates the rating sum information of all the contents. Thereafter, the content providing server 300 sorts the ranking of all 100 contents on the basis of the rating sum information. The content providing server 300 generates a first upper content list Contents_List_1 including contents within a predetermined rank, for example, 20%, among the 100 contents. Then, the first upper content list Contents_List_1 may be a list including a total of 20 contents.

When the first upper content list Contents_List_1 is generated, the content providing server 300 extracts users who have rated 20 contents based on user data. In this case, the content providing server 300 assigns weights according to the order of the 20 contents. For example, the content providing server 300 may assign a weight of 20 to the 1st content, a weight of 19 to the 2nd content, and a weight of 1 to the 20th content. Weighting is arbitrary.

When the weighting is completed, the content providing server 300 multiplies the weight of the content with the ratings given to the corresponding contents to calculate a reliability score of the user who has given a rating to the specific content. For example, if a user A has given 4 points to 1st content, 5 points to 2nd content, and 3 points to 20th content, user A has 80 + 95 + 3 = 178 points. You will have a confidence score of. In addition, assuming that a user B gives 4 points to 3rd content, 4 points to 4th content, 4 points to 5th content and 4 points to 6th content, user B is 72 You can have a confidence score of + 68 + 64 + 60 = 370 points. If the number of users who rated 20 contents in this way was 70, for example, the order is sorted based on the reliability scores of 70 people. Thereafter, the content providing server 300 may generate a higher user list (User_List) including 14 users within a predetermined rank, for example, 20% of the 70 users.

Here, the opinion leader selection unit 360 may generate an opinion list (Opinion_List) based on the upper user list (User_List), but may adjust the number of people included in the upper user list (User_List) by adjusting a ratio of a predetermined rank. have. That is, the opinion leader selecting unit 360 may adjust the ratio to 5% to calculate a list including 3 to 4 top users, based on the list of contents viewed by the top users included in the list. Opinion_List may be created to include.

Meanwhile, when 14 upper user lists (User_List) are generated in the content providing server 300, weights are sequentially assigned to each user included in the upper user list (User_List). For example, the content providing server 300 has a weight of 14 for the first user and a weight of 13 for the second user. A weight of 1 may be assigned to the fourth-placed user. After the weighting is completed, the content providing server 300 calculates a reliability score for the content that each user has rated. For example, if the 1st user has a rating of 4, the 2nd user has a rating of 4, and the 10th user has a rating of 4, the confidence score of the content "a" is 96 + 72 + 20 = 188 points. In the same way, the 4th user rated 3, the 5th user rated 4, the 6th rated 5, the 7th rated 5, and the 8th rated Given a rating of 5, the “I” confidence score can be 33 + 40 + 45 + 40 + 35 = 193 points. In this way, a confidence score is calculated for contents that 14 users have rated. Here, if there are 30 pieces of content that are rated by 14 users, the content providing server 300 sorts the 30 pieces of content by the reliability score and generates a second upper content list (Contents_List_2) including the respective pieces of content.

In order to more easily organize data in the above-described calculation process, that is, the first high content list (Contents_List_1) and the second high content list (Contents_List_2), only users who have given a rating of 4 or more to the content It can also be extracted. For example, when calculating a content reliability score of "I" in the second upper content list (Contents_List_2), the rating of the fourth-ranked user who rated 3 may be ignored when calculating the content reliability score of "I". .

As described above, the social network-based content recommendation system of the present invention calculates the first upper content list (Contents_List_1) based on a rating given by users for various contents, and then calculates the calculated first upper content list ( The upper user list User_List is calculated based on Contents_List_1. The social network-based content recommendation system of the present invention calculates a second upper content list (Contents_List_2) based on the upper user list (User_List). As a result, the social network-based content recommendation system has a high reliability, and many users can read more and support users with high ratings. In this case, the upper user list User_List may use information about users who have rated the contents of the first upper contents list Contents_List_1, weights corresponding to the ratings and rankings of the contents that the user has given the contents. It includes the confidence score between each calculated user.

In the above, embodiments of the social network-based content recommendation system and system have been described through the embodiments of the present invention. Hereinafter, a description of the method for recommending a content based on a social network of the present invention will be described while omitting as much as possible from the above description. From the description below, the social network based content recommendation system and its effects will also be more apparent.

4 is a diagram illustrating a signal flow between system configurations for generating a first upper content list (Contents_List_1).

Referring to FIG. 4, the content providing server 300 provides at least one content to the at least one user terminal 100 in step S101. In this case, since the content transmitted to the user terminal 100 is transmitted at the stage before the rating or the reliability score is given, the content providing server 300 may provide the content at random or provide the content requested by the user.

The user terminal 100 generates content information (Contents_Info) including data obtained by reading a specific content and giving a rating in step S103 and user data (User_Data) including user information (User_Info) for the user terminal 100. do. Thereafter, the user terminal 100 transmits user data (User_Data) to the content providing server 300 in step S105.

When the content providing server 300 receives the user data (User_Data) from the user terminal 100, in step S107 extracts the content information (Contents_Info) and the user information (User_Info), and the evaluation score log storage unit 320 Store in Thereafter, the content providing server 300 calculates a score sum score for the plurality of contents based on the plurality of contents information (Contents_Info) delivered from the plurality of user terminals 100 in step S109, and scores in step S111. Sort a plurality of contents based on the sum score.

Finally, the content providing server 300 extracts upper contents falling within a predetermined rank among the plurality of contents arranged in step S113, and generates a first upper contents list Contents_List_1 including the extracted upper contents. In this case, the first upper content list Contens_List_1 is a content list in which the initial rating of the initial content is initialized.

A process of generating an upper user list User_List using the first upper content list Contents_List_1 in the content providing server 300 will be described with reference to FIG. 5.

FIG. 5 is a flowchart illustrating a process of generating an upper user list User_List based on a first upper content list Contents_List_1.

Referring to FIG. 5, in operation S201, the content providing server 300 assigns weights to respective contents included in the first upper content list Contents_List_1 in order. In addition, the content providing server 300 extracts users who have given a rating to the first higher content list Contents_List_1 in step S203, and contents that belong to the first higher content list Contents_List_1 that a specific user has rated in step S205. Detects contents to which a rating is equal to or greater than a predetermined score. Here, step S205 may be omitted, and the order of steps S201 and S203 may be reversed.

The content providing server 300 calculates a reliability score of the specific user by multiplying the weights assigned to the content by the weight of the content that the specific user has given a rating in step S207. If there are a plurality of contents rated by a specific user, a reliability score for the specific user may be calculated by adding up all the values calculated by multiplying the rating given to each content by the weight of the corresponding content.

In step S209, the content providing server 300 checks whether all the reliability scores of the users extracted in step S203 have been calculated. If there are remaining users whose reliability score is not calculated, the process branches to step S207 and repeats the process. However, when all the reliability scores of the users extracted in step S203 is calculated, the content providing server 300 sorts each reliability score calculated in step S211 by score.

Finally, the content providing server 300 extracts users who fall within a predetermined rank based on the sorted order in step S213, and generates a higher user list (User_List). Meanwhile, the content providing server 300 further generates an opinion list (Opinion_List) based on the list of contents read by the user included in the upper user list (User_List) after generating the upper user list (User_List). Can be done.

FIG. 6 is a flowchart illustrating a process of generating a second upper content list Contents_List_2 based on the upper user list User_List.

Referring to FIG. 6, the content providing server 300 assigns a weight to each user included in the upper user list User_List in step S301 according to the ranking, and in step S303, the contents to which the users have been rated Detect. Here, the content providing server 300 may detect contents to which upper users have given a predetermined score or more. At this time, the content to which a rating below a predetermined predetermined score is given is excluded when performing the reliability check calculation.

The content providing server 300 calculates a reliability score by multiplying the ratings given by the user to the specific content among the contents detected in step S305 and the weights of the users who gave the ratings. If there are a large number of users who have given a certain score or more to a certain content, the score can be calculated by multiplying the score given by the user with the specific content, and the reliability score can be calculated by adding the values together. On the other hand, the rating of the predetermined score or more given to the content may be unified to one reference score to calculate the reliability score.

Thereafter, the content providing server 300 checks whether the reliability score has been calculated for all contents in step S309. If the content with no reliability score calculated, the process branches to step S207 to repeat the process. However, if the credit score is calculated for all the contents, the content providing server 300 sorts the order of each content by the reliability score in step S311.

In operation S313, the upper content items within a predetermined rank may be extracted to generate a second upper content list Contents_List_2, and in step S315, a web page based on the second upper content list Contents_List_2 may be extracted. It may be prepared and transmitted to the user terminal 100.

So far, the social network-based content recommendation system, method, and content providing server according to the present invention have been described. In the present specification and drawings, preferred embodiments of the present invention have been disclosed, and although specific terms have been used, these are merely used in a general sense to easily explain the technical contents of the present invention and to help the understanding of the present invention. It is not intended to limit the scope. It will be apparent to those skilled in the art that other modifications based on the technical idea of the present invention can be carried out in addition to the embodiments disclosed herein.

According to the present invention, more reliable content can be constructed by using voluntary participation of users and their collective intelligence. In particular, since the content is built based on the participation of users, it is possible to provide more accurate and reliable content information when other users request the content.

1 is a view schematically showing the configuration of a social network-based content recommendation system according to an embodiment of the present invention;

Figure 2 shows the configuration of the content providing server of the present invention in more detail,

3 is a view for explaining the function of the content providing server of the present invention;

4 is a signal flow diagram between system configurations for generating a first upper content list;

5 is a flowchart illustrating a process of generating an upper user list based on a first upper content list;

6 is a flowchart illustrating a process of generating a second upper content list based on a higher user list.

Claims (19)

In the content recommendation method, Receiving, by the content providing server, at least one user data including content information of which the user terminal has given a rating and user information about the user terminal; Ranking the at least one rated content by score and calculating first upper content within an upper predetermined rank; Determining a ranking of users who have given a rating to higher contents by using a weight according to the ranking of the higher contents and a rating given by the user, and calculating upper users within a predetermined rank among the determined rankings of the users; Determining the ranking of the contents by using the weights according to the rankings of the upper users and the ratings given by the upper users, and calculating second upper contents within a predetermined rank among the determined rankings of the contents; How to recommend content based on your network. The method of claim 1, And storing the user data, the ranking of the contents, and the ranking of the upper users. The method of claim 1, And storing an opinion list, which is a list of contents viewed by the upper-level users. The method of claim 3, The method of claim 1, further comprising the step of transmitting the opinion list to the user terminal. The method of claim 1, Generating a second upper content list including the second higher contents; And And transmitting the second upper content list to the user terminal. The method of claim 1, Disposing the second higher-level contents in an area having a high recognition rate among areas of a web page for transmitting the second upper content to the user terminal; And And transmitting the web page to the user terminal. In the content recommendation system, At least one user terminal for generating user data including identification information and content information including data on a rating given to the viewed content; After receiving the user data from the user terminals, at least one rating is ranked by the scores by the score and the first higher content within the predetermined schedule ranking, the weight and the user according to the ranking of the higher content The ratings given by the users are used to determine the ranking of users who rated the upper contents, calculate the top users within a certain rank among the determined rankings of users, and the weights according to the ranking of the top users and the ratings given by the top users. And a content providing server for determining the ranking of the contents using the ranking and calculating second higher contents within a predetermined ranking among the determined ranking of the contents. The method of claim 7, wherein The content providing server Extracting the ratings of the contents from the received user data, and extracting a first upper content list and a higher user list from the calculated first upper contents, the calculated upper users, and the calculated second upper contents, respectively. And a second upper content list, the social network based content recommendation system. The method of claim 8, The content providing server And storing the user data, the ranking of the contents, the ranking of the users, the first upper content list, the upper user list, and the second upper content list. The method of claim 8, The content providing server Transmitting at least one of contents included in the second upper content list, the second upper content list, and the upper user list to the user terminal, and a web page on which the second upper content is disposed in an area having high recognition rate. And generating and transmitting the web page to the user terminal. The method of claim 7, wherein The content providing server Detecting contents viewed by the upper users, generating an opinion list including the detected contents, and transmitting the opinion list to the user terminal. The method of claim 7, wherein The content information Social network-based content recommendation system comprising the identification information of the content and the user's rating information on the content. A user data extracting unit for receiving user data from user terminals and extracting ratings of contents from the user data; A content reliability score determination unit which ranks the at least one rated content by score and calculates first upper content items within a higher predetermined rank to generate a first upper content list; The rank of the users who have given the rating to the first higher content by using the weight according to the ranking of the first higher content and the rating given by the user is determined, and the upper user by calculating the top users within a certain rank among the determined ranking of the users A user evaluation score determination unit for generating a list; The ranking of the contents is determined using the weights according to the rankings of the upper users and the ratings given by the upper users, and the second upper contents list is generated by calculating second upper contents within a predetermined rank among the determined contents. A reliable content selection unit; Social network-based content providing server comprising a. The method of claim 13, An evaluation score log storage unit for storing the user data, the ranking of the contents, the ranking of the users, the first upper content list, the upper user list, and the second upper content list; Social network-based content providing server, characterized in that it further comprises. The method of claim 13, And a content providing unit configured to transmit at least one of contents included in the second upper content list, the second upper content list, and the upper user list to the user terminal. Providing server. The method of claim 15, The content providing unit And a web page in which the second upper content items are arranged in an area having a high recognition rate, and transmitting the web page to the user terminal. The method of claim 13, An opinion leader selecting unit which detects contents viewed by the upper users; Social network-based content providing server, characterized in that it further comprises. The method of claim 17, The opinion leader selection unit Generating an opinion list including contents viewed by the detected upper-level users, The content providing unit Social network based content providing server, characterized in that for transmitting the opinion list to the user terminal. The method of claim 13, The content information Identification information of the content and Social network-based content providing server comprising the user's rating information for the content.

Images