KR20120130962A - System and method for analyzing similar inclination - Google Patents
System and method for analyzing similar inclination Download PDFInfo
- Publication number
- KR20120130962A KR20120130962A KR1020110049080A KR20110049080A KR20120130962A KR 20120130962 A KR20120130962 A KR 20120130962A KR 1020110049080 A KR1020110049080 A KR 1020110049080A KR 20110049080 A KR20110049080 A KR 20110049080A KR 20120130962 A KR20120130962 A KR 20120130962A
- Authority
- KR
- South Korea
- Prior art keywords
- user
- similarity
- questionnaire
- categories
- calculating
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
Abstract
Description
The present invention relates to a technique for analyzing similarity between users, and more particularly, to a method and system for providing a questionnaire to a user and analyzing the similarity between users based on the contents answered to the questionnaire.
Today, due to the development of communication networks and communication terminals, various types of services are provided through communication systems. Such services vary from User Created Contents (UCC) services and Video On Demand (VOD) services. Among them, social networking services have recently been in the spotlight.
The social networking service is an online social networking service for the purpose of forming a new network with an unspecified person using an internet communication medium such as a mini homepage or strengthening a network with friends. Such social networking services are provided to users in various forms in almost every portal service company (e.g., Naver's MeToday, Daum, Nate's Cyworld, etc.).
These social networking services also serve to form online communities among people with similar common interests. For example, people interested in inline skates exchange each other's online contacts (eg, mini homepage address, blog address, etc.), thereby sharing knowledge related to inline skates on their community.
However, a user using a social networking service should grasp the main interests of other users by accessing other users' homepages, blogs, or checking other user's profiles in order to form contacts with others with similar interests. That is, the user checks whether the interests of other users are the same as his or her interests by accessing a homepage, a blog, or a profile in order to form a network with others with similar interests. However, since this method depends on the passive method of the user, not only does it take a long time to form a network, but also depends on the user's intuition and thus, there is a problem that the interests are actually inconsistent.
The present invention has been proposed to solve such a conventional problem, and provides a similar propensity analysis method and system for analyzing similar propensities among users based on data stored in a database, and recommending similar users with similar interests based on the analyzed information. Its purpose is to.
In particular, it is another object of the present invention to provide a method and system for analyzing similarity between users by calculating interest similarity and consensus similarity between users using response information of a user described in an online questionnaire.
Other objects and advantages of the present invention can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. It will also be readily apparent that the objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.
Method for analyzing the similarity between users in the similarity analysis system according to the first aspect of the present invention for achieving the above object, the number of questionnaire categories by extracting the category of the questionnaire responded by the first user and the second user for each user A questionnaire category summing step of summing for each user; Calculating the number of overlapping questionnaire categories overlapping between the first user and the second user; And a questionnaire category of each user summed in the questionnaire category adding step to a ratio of the number of duplicate questionnaire categories calculated in the step of calculating the number of duplicates compared to the number of questionnaire categories of the first user summed in the questionnaire category adding step. And a similarity of interest calculation step of calculating a similarity of interest of the second user to the first user by multiplying the relative ratio of the number as a weight.
The method for analyzing similarity between users in the similarity analysis system according to the second aspect of the present invention for achieving the above object includes a category list of questionnaires responded to by the first user and the second user, and the first user and the first user. A category summing step of extracting at least one or more lists from the category list of the contents accessed by the second user for each user and summing the number of categories recorded in the corresponding category list for each user; Calculating a number of duplicates for calculating the number of categories overlapping between the first user and the second user; And weighting the relative ratio of the number of categories of each user summed in the category adding step to the ratio of the number of duplicate categories calculated in the duplication number calculating step relative to the number of categories of the first user summed in the category adding step. And calculating interest similarity by multiplying as to calculate interest similarity of the second user with respect to the first user.
The method for analyzing the similarity between users in the similarity analysis system according to the third aspect of the present invention for achieving the above object, by extracting the questionnaire questions answered by the first user and the second user for each user, the extracted questionnaire A questionnaire question summing step of adding up the number of questions for each user; An equal answer calculating step of calculating the number of questions in the questionnaire answered by the first user and the second user equally; And a ratio of the number of questionnaires of the respective users summed in the questionnaire question adding step to the ratio of the number of questions calculated in the same answer calculation step to the number of questionnaire questions of the first user summed in the questionnaire question adding step. And a consensus similarity calculating step of calculating a consensus similarity of the second user with respect to the first user by multiplying the ratio as a weight.
Similarity analysis system according to a fourth aspect of the present invention for achieving the above object, the database for storing the questionnaire response information for each user; And using the questionnaire response information of the first user and the second user extracted from the database, checking the category of the questionnaire responded by the first user and the second user for each user, and adding up the number of questionnaire categories for each user, The number of questionnaire categories of each of the users summed up to the ratio of the number of questionnaire categories overlapping the sum of the questionnaire categories of the first user by calculating the number of questionnaire categories overlapping between the user and the second user. And similarity calculating means for calculating a similarity of interest of the second user with respect to the first user by multiplying the relative ratio of as a weight.
The present invention analyzes users who have similar tendencies with each other and provides the analyzed results to the users, thereby helping them to easily and accurately grasp the other person's dispositions online as well as contributing to social networking on social networks. There is an advantage.
In addition, the present invention has an advantage of providing a more intuitive similarity tendency analysis result to users by calculating the similarity of interest similarity and consensus similarity between users based on questionnaire response information and content access information accessed by the user.
In addition, the present invention has an advantage of rationally quantifying similarity between users by applying interest weight and consensus similarity by applying variable weights depending on user's basic data.
BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments of the invention and, together with the description, serve to explain the principles of the invention. And shall not be construed as limited to such matters.
1 is a view showing the configuration of a similarity analysis system according to an embodiment of the present invention.
2 is a diagram illustrating a configuration of a similarity analysis server and a database according to an embodiment of the present invention.
3 is a flowchart illustrating a method of analyzing similarity between users based on questionnaire response information and content access information in the similarity analysis system according to an embodiment of the present invention.
4 is a flowchart illustrating a method of analyzing similarity between users based on questionnaire response information and content access information in the similarity analysis system according to another embodiment of the present invention.
5 is a flowchart illustrating a method of analyzing similarity between users based on questionnaire response information and content access information in a similarity analysis system according to another embodiment of the present invention.
6 is a flowchart illustrating a method of analyzing similarity between users based on questionnaire response information and content access information in the similarity analysis system according to another embodiment of the present invention.
The foregoing and other objects, features and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings, in which: There will be. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.
Prior to describing a method and system for analyzing similarity tendency according to an embodiment of the present invention, terms to be described below are described.
Interest similarity is a quantification of the overlapping rate for the kind of content that two users have in common.
The consensus similarity is a digitized portion of two users empathizing with each other. Specifically, the sympathy similarity is a quantification of the proportion of content accessed by two users and the response rate of a questionnaire responding to the same.
Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.
1 is a view showing the configuration of a similarity analysis system according to an embodiment of the present invention.
As shown in FIG. 1, the similarity analysis system according to an embodiment of the present invention includes a
The
The communication terminal 100 -N receives an online questionnaire from the
The
The
2 is a diagram illustrating a configuration of a similarity analysis server and a database according to an embodiment of the present invention.
As shown in FIG. 2, the
The subscriber
The access
The category information indicates the classification information of the content, and category information for classifying a video such as sports, drama, movie, entertainment, documentary, etc. may be added for each video content, and the category of clothing, mobile phone, MP3, PMP, notebook, etc. Information may be added for each shopping content. In addition, category information for classifying music such as jazz, dance, hip hop, classical, ballad, etc. may be added for each audio content, and various category information may be added to the corresponding content and stored in the
Meanwhile, the access
The
The
The subscriber manager 220 performs a function of creating, modifying, and deleting member information. In detail, the subscriber management unit 220 receives data such as a name, ID / password, home address, blog address, and homepage address from a user who has requested a new membership, and stores the data in the subscriber
The questionnaire provider 230 provides an online questionnaire to a member or a non-member using the
The access information manager 240 maps the identification information of the content accessed by the user and the category information of the content and stores the mapping information in the
The payment processor 270 pays a service fee or a product purchase cost for a user who uses a paid service or purchases a specific product. In this case, the payment processing unit 270 may receive a corresponding service use fee or a product purchase cost from a corresponding user by using various payment means such as a credit card, a mobile phone, a passbook deposit, a gift certificate, and the like.
The similarity calculator 250 calculates interest similarity and consensus similarity between a plurality of users using data stored in the
n (A): Number of categories of content accessed by comparison user + Number of categories of questionnaire responded by user
n (B): number of categories of content accessed by the reference user + number of categories of questionnaire responded by the reference user
n (A∩B): Sum of the number of duplicate content categories and the number of questionnaire categories between the reference user and the comparison user.
s is the smaller of n (A) and n (B)
l: the larger of n (A), n (B)
In addition, the similarity calculator 250 calculates the consensus similarity by substituting the questionnaire response information and the content access information of the two users, which are respectively prepared by the reference user and the comparison target user, into Equation 2 below.
n (X): Number of questions in questionnaire responded by comparison user + Number of contents accessed by comparison user
n (Y): Number of questions in each questionnaire responded by the reference user + Number of content accessed by the reference user
n (X∩Y): Number of questions answered by the reference user and the comparison user equally + Number of contents accessed by the reference user and the comparison user the same
N: integer that discarded the fractional part from 1 / m × n (X) -n (Y) |, with N = 2 when N <2
(The initial value of m is 1, which is a random constant that increases with the sum of the total number of content stored in the database and the total number of questions answered)
s: the smaller of n (X), n (Y)
l: the larger of n (X), n (Y)
In Equation 2, the arbitrary constant 'm' is a constant that increases so that the sum of the total number of contents stored in the
The network manager 260 performs a function of providing the reference user with contacts of other users having similarity with the reference user by using at least one of interest similarity and consensus similarity calculated by the similarity calculator 250. . In this case, the network manager 260 may refer to the similarity of interest for each user calculated by the similarity calculator 250 to determine whether there is a user whose interest similarity exceeds a predetermined first threshold (eg, 50%). If there is a check and existence, contact information such as a user's profile, phone number, and email address may be provided to the reference user as contact information. Alternatively, the network manager 260 may refer to the consensus similarity for each user calculated by the similarity calculator 250, and if there is a user whose consensus similarity exceeds the second threshold (eg, 70%), The contact information of this user can be provided to the reference user.
In addition, the network management unit 260 also serves to form a network between a plurality of users. In detail, the network manager 260 registers other users in a friend list of a specific user, thereby forming a network of users.
Hereinafter, a method of analyzing similarity between users in the similarity analysis system according to the present invention will be described in more detail with reference to FIGS. 3 to 6.
3 is a flowchart illustrating a method of analyzing similarity between users based on questionnaire response information and content access information in the similarity analysis system according to an embodiment of the present invention.
Referring to FIG. 3, the questionnaire providing unit 230 extracts one or more online questionnaires from the
Subsequently, when the survey period ends, the similarity calculator 250 checks one or more user information (eg, IDs) that responded to the survey during the survey period, and access information of the
Next, the similarity calculator 250 checks the content category information from the extracted content access information, and adds the total number of categories for the content accessed by each user for each user (S305). In this case, when there is duplicate category information in the content category information of a specific user, the similarity calculator 250 may calculate the total number of categories by adding the duplicate category information by one number rather than adding them individually. . For example, when three categories of information such as 'music', 'game', and 'game' are extracted as category information on the content of user 'A', the similarity calculator 250 calculates a duplicate 'game'. The category information may be calculated as one number, and the total number of categories for the 'A' user may be added to two.
Next, the similarity calculator 250 extracts the questionnaire response information collected during the survey period by dividing the questionnaire response information by the user from the
Next, the similarity calculator 250 selects a specific user as a reference user, and selects users other than the reference user as the comparison target user (S311). Here, the reference user is a user who is a reference for analyzing the similarity, and a user who answers the questionnaire may be sequentially selected from the reference target user. In addition, the comparison target user is a user who compares and analyzes how much similarity the user has with the reference user.
After selecting the reference user, the similarity calculator 250 calculates a sum of the number of content categories overlapping between the reference user and the specific comparison target user and the number of categories of the questionnaire. Subsequently, the similarity calculating unit 250 overlaps the number of content categories of the reference user, the number of categories of questionnaires participated by the reference user, the number of content categories of specific comparison target users, the number of categories of questionnaires participated by a specific comparison target user, and overlaps between the two users. By substituting the sum of the number of content categories and the number of questionnaire categories into Equation 1, interest similarity between the reference user and the first comparison target user is calculated (S313). Thus, using Equation 1, the similarity calculator 250 sequentially calculates similarity of interests between the comparison target user and the reference user other than the specific comparison target user.
For example, the content category of the reference user stored in the access
The similarity of interest calculated through Equation 1 is high when the category of the accessed content matches between the reference user and the comparison target user or the category of the answered questionnaire matches. Also, interest similarity is a weight that is assigned to the category match rate between the reference user and the comparison user (
In this manner, when the similarity calculator 250 calculates the interest similarity between all the comparison target users and the reference user, the network manager 260 may determine a similarity degree of interest with the reference user (eg, 50%). Check whether there is a comparison target user exceeding (S315). Network management unit 260, if there is a comparison target user whose interest similarity exceeds a threshold value, the contact information such as blog address, email address, profile information, mobile phone number of the
On the other hand, the network manager 260 notifies the reference user that there is no user matching the Guam similarity when there is no comparison user whose interest similarity exceeds a threshold (eg, 50%).
4 is a flowchart illustrating a method of analyzing similarity between users based on questionnaire response information and content access information in the similarity analysis system according to another embodiment of the present invention.
In the following description with reference to FIG. 4, portions overlapping with the contents of FIG. 3 will be described by compressing and the differences will be mainly described.
Referring to FIG. 4, the questionnaire providing unit 230 extracts the online questionnaire from the
Subsequently, when the survey period ends, the similarity calculator 250 checks user IDs responding to the survey during the survey period, and accesses the content access information mapped to the identified user IDs. Extracted at 320 (S403). In this case, the similarity calculator 250 may select and extract only content access information stored in the
Next, the similarity calculator 250 adds the total number of contents accessed by the user based on the content identification information included in the extracted content access information (S405). Subsequently, the similarity calculator 250 extracts the questionnaire response information collected during the survey period by dividing the questionnaire response information by the user in the questionnaire storage unit 330 (S407). Next, the similarity calculation unit 250 adds up the total number of questionnaire questions answered by the user based on the extracted questionnaire response information for each user (S409).
Next, the similarity calculator 250 selects a specific user as a reference user, and selects users other than the reference user as the comparison target user (S411). Subsequently, the similarity calculator 250 calculates the number of content identification information (that is, the number of contents accessed by two users equally) and the number of questionnaires answered by two users identically between the reference user and a specific comparison target user. Calculate. Subsequently, the similarity calculating unit 250 may include the number of contents accessed by the reference user, the number of questions answered by the reference user, the number of contents accessed by a specific comparison target, the number of questions answered by the specific comparison target user, By substituting the number of contents accessed by both the reference user and the specific comparison target user and the number of question items answered by the two users in Equation 2, the consensus similarity between the reference user and the comparison target user is calculated (S413). . Thus, using Equation 2, the similarity calculator 250 sequentially calculates similarity of interests between other comparison target users and the reference user.
For example, the content identification information of the reference user stored in the access
In this case, 'www.XXX.com/aaa.asf' is duplicated in the content identification information, and five of the questionnaire response information are duplicated, and '6' is substituted into n (X∩Y) in Equation (2). . In addition, since the number of content identification information stored in the access
When each of these numbers is substituted into Equation 2, the similarity calculator 250 calculates a consensus similarity between the reference user and the first comparison target user as 19.28 (%).
The similarity of the consensus is calculated high when the number of similarly accessed contents between the reference user and the comparison target user is large or the number of questionnaires answered with the same answer is large. In addition, consensus similarity is weighted based on the ratio of the number of matching content and the number of question items between the reference user and the comparison user.
In this manner, when the consensus similarity calculation between all the comparison target users and the reference user is completed by the similarity calculator 250, the network manager 260 may have a consensus similarity level with the reference user (eg, 70%). Check whether there is a comparison target user exceeding (S415). If there is a comparison target user whose interest similarity exceeds a threshold, the network manager 260 extracts one or more user contacts corresponding to the interest from the subscriber
On the other hand, the
5 is a flowchart illustrating a method of analyzing similarity between users based on questionnaire response information and content access information in a similarity analysis system according to another embodiment of the present invention.
Hereinafter, in the description with reference to FIG. 5, portions overlapping with the contents of FIGS. 3 and 4 will be described by compressing and the differences will be mainly described.
Referring to FIG. 5, the questionnaire providing unit 230 transmits the online questionnaire extracted from the
Subsequently, the similarity calculator 250 checks the user IDs responding to the survey during the survey period, and accesses the content access information mapped to the identified user IDs from the access
Next, the similarity calculator 250 checks category information in each extracted content access information, and adds the total number of categories for the content accessed by the user for each user (S505). Subsequently, the similarity calculator 250 extracts the questionnaire response information of the user who responded during the survey period by dividing the questionnaire by the user from the
Next, the similarity calculator 250 selects a specific user as a reference user, and selects users other than the reference user as the comparison target user (S511). Next, after the similarity calculator 250 selects the reference user, the number of content categories of the reference user, the number of categories of the questionnaire participated by the reference user, the number of content categories of the specific comparison target user, and the questionnaire in which the specific comparison target user participates By substituting the sum of the number of categories and the total number of content categories duplicated between the two users and the number of questionnaire categories in Equation 1, interest similarity between the reference user and each comparison target use is calculated (S513).
Next, the similarity calculator 250 determines whether there is a comparison target user whose interest similarity exceeds the first threshold value (eg, 50%), and selects the user as a consensus analysis target user if it exists. (S515). Subsequently, the similarity calculator 250 adds the total number of contents accessed by the consensus analysis target user and the reference target user for each user based on the content access information extracted in step S503 (S517). Next, the similarity calculator 250 adds the total number of questionnaire questions answered by the reference user and the consensus analysis target user for each user based on the questionnaire response information extracted in step S507 (S519).
Subsequently, the similarity calculating unit 250 may include the number of contents of the reference user, the total number of questions answered by the reference user, the number of contents of the consensus analysis target user, the total number of questions answered by the consensus analysis target user, and the reference user. And the consensus similarity between the reference user and the consensus analysis target user is calculated by substituting the number of content identification information accessed by all the consensus analysis target users and the number of questionnaire questions answered by the two users in Equation 2. S521). In this way, using Equation 2, the similarity calculator 250 sequentially calculates similarity of interests between the user of another consensus analysis target and the reference user.
When the consensus similarity calculation is completed by the similarity calculator 250, the network manager 260 checks whether or not the user whose consensus similarity exceeds the second threshold value (eg, 70%) exists. The contact information of the user is extracted from the subscriber information storage unit 310 (S523 and S525). Next, the network manager 260 provides the network recommendation information to the extracted reference user (S527).
Through the method of FIG. 5, a user whose interest similarity is greater than or equal to the first threshold and the consensus similarity is greater than or equal to the second threshold is recommended as a network recommendation user of the reference user. Branches can make connections with other users.
Meanwhile, the
6 is a flowchart illustrating a method of analyzing similarity between users based on questionnaire response information and content access information in the similarity analysis system according to another embodiment of the present invention.
Hereinafter, in the description with reference to FIG. 6, a user who owns communication terminal 1 (100-1) is referred to as a first user, and a user who owns communication terminal 2 (100-2) is referred to as a second user.
Referring to FIG. 6, the communication terminal 1 100-1 transmits a connection request message including an ID and a password to the
Subsequently, after successfully logging in, the communication terminal 1 100-1 requests the
Then, the similarity calculator 250 of the
Next, the similarity calculator 250 checks category information in the extracted content access information, and adds the total number of categories for content accessed by the first user and the second user for each user. Subsequently, the similarity calculator 250 checks category information in each questionnaire response information, and based on the category information, sums the total number of categories for the questionnaire answered by the first user and the second user for each user.
Next, the similarity calculator 250 selects the first user as the reference user and selects the second user as the comparison target user. Next, the similarity calculator 250 may include the number of content categories of the first user (ie, the reference user), the number of categories of questionnaires in which the first user participates, the number of content categories of the second user, and the second user (ie, comparison). The interest similarity between the first user and the second user is calculated by substituting the sum of the category number of the questionnaire participated by the target user, the sum of the number of content categories overlapping between the two users, and the number of questionnaire categories in Equation 1 (S609).
Next, the similarity calculator 250 adds the total number of contents accessed by the first user and the second user for each user based on the content access information extracted in operation S607. Next, the similarity calculator 250 adds up the total number of questionnaire questions answered by the first user and the second user for each user based on the questionnaire response information extracted in step S607. Subsequently, the similarity calculator 250 may include the number of contents accessed by the first user (ie, the reference user), the total number of questions answered by the first user, the number of contents of the second user (ie, the comparison target user), and the like. By substituting Equation 2 into the total number of questionnaires answered by the second user, the number of contents accessed by both the first user and the second user, and the number of questionnaires answered by both users in Equation 2, The consensus similarity between the second users is calculated (S611).
The similarity calculator 250 transmits the calculated interest similarity and consensus similarity to the communication terminal 1 100-1 using the
Then, the network manager 260 of the
As described above, the
Meanwhile, in the above-described embodiments, the interest similarity and the consensus similarity have been described using the questionnaire response information collected during the survey period, but the present invention is not limited thereto, and the questionnaire collected during the period set by the administrator is described. By extracting the response information and the content access information, it is made clear that interest similarity and consensus similarity between users can be calculated based on the extracted information.
In addition, in the above-described embodiments, it has been described that both the questionnaire response information and content access information are used when calculating the consensus similarity and interest similarity, but the present invention uses any one of the questionnaire response information and content access information. It should be clear that interest similarity and consensus similarity between users can be calculated. In detail, when the
In addition, the
While the specification contains many features, such features should not be construed as limiting the scope of the invention or the scope of the claims. In addition, the features described in the individual embodiments herein may be combined and implemented in a single embodiment. Conversely, various features described in the singular < Desc / Clms Page number 5 > embodiments herein may be implemented in various embodiments individually or in combination as appropriate.
Although the operations have been described in a particular order in the figures, it should be understood that such operations are performed in a particular order as shown, or that all described operations are performed to obtain a sequence of sequential orders, or a desired result . In certain circumstances, multitasking and parallel processing may be advantageous. It should also be understood that the division of various system components in the above embodiments does not require such distinction in all embodiments. The above-described program components and systems can generally be implemented as a single software product or as a package in multiple software products.
The method of the present invention as described above may be implemented as a program and stored in a recording medium (CD-ROM, RAM, ROM, floppy disk, hard disk, magneto-optical disk, etc.) in a computer-readable form. Since this process can be easily implemented by those skilled in the art will not be described in more detail.
It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention. The present invention is not limited to the drawings.
100-N: communication terminal 200: similarity analysis server
210: transceiver unit 220: subscriber management unit
230: questionnaire provision unit 240: access information management unit
250: similarity calculation unit 260: network management unit
270: payment processing unit 300: database
310: subscriber information storage unit 320: access information storage unit
330: survey storage 400: network
Claims (11)
A questionnaire category summing step of extracting categories of questionnaires answered by the first user and the second user for each user and summing the number of questionnaire categories for each user;
Calculating the number of overlapping questionnaire categories overlapping between the first user and the second user; And
The number of questionnaire categories of each user added in the questionnaire category adding step to the ratio of the number of duplicated questionnaire categories calculated in the step of calculating the number of duplicates compared to the number of questionnaire categories of the first user summed in the questionnaire category adding step And calculating a similarity of interest of the second user to the first user by multiplying the relative ratios of the weighted values as a weight.
The interest similarity calculating step,
The similarity propensity analysis method of calculating a similarity of interest of the second user to the first user according to Equation 3 below.
(3)
n (A): Number of categories of questionnaire responded by second user
n (B): Number of categories of questionnaire responded by first user
n (A∩B): Sum of the number of questionnaire categories overlapping between the first user and the second user
s is the smaller of n (A) and n (B)
l: the larger of n (A), n (B)
After the interest similarity calculation step,
Checking whether the calculated interest similarity exceeds a preset threshold; And
And if the calculated interest similarity exceeds the threshold, extracting contact information of the second user and providing the contact information as social networking recommendation information to the first user. Characteristic similarity analysis method characterized by.
At least one or more lists are extracted from the category list of the questionnaire responded by the first user and the second user, and the category list of the content accessed by the first user and the second user, respectively, and recorded in the corresponding category list. A category adding step of adding up the number of categories for each user;
Calculating a number of duplicates for calculating the number of categories overlapping between the first user and the second user; And
The ratio of the number of duplicate categories calculated in the duplication number calculation step to the number of categories of the first users summed in the category adding step is weighted by the relative ratio of the number of categories of each user added in the category adding step. And a similarity of interest calculation step of calculating a similarity of interest of the second user to the first user by multiplying.
The interest similarity calculating step,
The similarity propensity analysis method of claim 2, wherein the similarity of interest of the second user to the first user is calculated according to Equation 4 below.
(4)
n (A): Number of categories of content accessed by the second user + Number of categories of questionnaire responded by the second user
n (B): number of categories of content accessed by the first user + number of categories of questionnaire responded by the first user
n (A∩B): Sum of the number of content categories duplicated between the first user and the second user and the number of questionnaire categories
s is the smaller of n (A) and n (B)
l: the larger of n (A), n (B)
A questionnaire step summing step of extracting questionnaire questions answered by the first user and the second user for each user, and adding the extracted questionnaire questions to each user;
An equal answer calculating step of calculating the number of questions in the questionnaire answered by the first user and the second user equally; And
The ratio of the number of questionnaires of each user summed in the questionnaire question sum step to the ratio of the number of questions calculated in the same answer calculation step to the number of questionnaire questions of the first user summed in the questionnaire question sum step And calculating a consensus similarity of the second user with respect to the first user by multiplying by a weight.
The consensus similarity calculating step,
A similarity propensity analysis method according to Equation 5 below, calculating the consensus similarity of the second user with respect to the first user.
(5)
n (X): Number of questions in questionnaire responded by second user
n (Y): Number of questions in each questionnaire responded by first user
n (X∩Y): Number of questions answered by the first and second users equally
N: integer that discarded the fractional part from 1 / m × n (X) -n (Y) |, with N = 2 when N <2
(The initial value of m is 1 and is a random constant that increases with the sum of the total number of questions stored in the database.)
s: the smaller of n (X), n (Y)
l: the larger of n (X), n (Y)
After the consensus similarity calculating step,
Checking whether the calculated consensus similarity exceeds a preset threshold; And
And if the calculated consensus similarity exceeds the threshold, extracting contact information of the second user and providing the contact information to the first user as contact recommendation information. Characteristic similarity analysis method characterized by.
Extracts at least one or more lists from the list of questionnaire questions answered by the first user and the second user, and the list of contents accessed by the first user and the second user for each user, and displays the detailed data recorded in the list. A summing step of summing each other;
Calculating at least one or more of the number of contents accessed by the first user and the second user in the same way and the number of questions in the questionnaire answered by the first user and the second user in the same manner;
The ratio of the number calculated in the overlapping number calculating step to the number of detailed data of the first user added in the adding step is multiplied by the relative ratio of the number of detailed data of each user added in the adding step as a weight. A similarity similarity calculating step of calculating a consensus similarity of the second user with respect to the first user.
The consensus similarity calculating step,
The similarity propensity analysis method of calculating a consensus similarity of the second user with respect to the first user according to Equation 6 below.
(6)
n (X): Number of questions in questionnaire responded by second user + Number of content accessed by second user
n (Y): Number of questions in each questionnaire responded by first user + Number of content accessed by first user
n (X∩Y): Number of questions answered by the first and second users equally + Number of contents accessed by the first and second users equally
N: integer that discarded the fractional part from 1 / m × n (X) -n (Y) |, with N = 2 when N <2
(The initial value of m is 1, which is a random constant that increases with the sum of the total number of content stored in the database and the total number of questions answered)
s: the smaller of n (X), n (Y)
l: the larger of n (X), n (Y)
Using the questionnaire response information of the first user and the second user extracted from the database, the category of the questionnaire responded by the first user and the second user is checked for each user, and the number of questionnaire categories is summed for each user, and the first user Calculates the number of questionnaire categories overlapping with each other and the second user, and the ratio of the number of questionnaire categories of each summed user to the sum of the number of questionnaire categories compared to the sum of the number of questionnaire categories of the first user is calculated. And similarity calculation means for calculating a similarity of interest of the second user to the first user by multiplying the relative ratio as a weight.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020110049080A KR20120130962A (en) | 2011-05-24 | 2011-05-24 | System and method for analyzing similar inclination |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020110049080A KR20120130962A (en) | 2011-05-24 | 2011-05-24 | System and method for analyzing similar inclination |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| KR20120130962A true KR20120130962A (en) | 2012-12-04 |
Family
ID=47514922
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR1020110049080A KR20120130962A (en) | 2011-05-24 | 2011-05-24 | System and method for analyzing similar inclination |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR20120130962A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20180023921A (en) * | 2018-02-06 | 2018-03-07 | 노유현 | Apparatus and method for providing personalication information |
| KR20230053444A (en) | 2021-10-14 | 2023-04-21 | 승문수 | Metaverse-based tendency classification display system |
| CN117520522A (en) * | 2023-12-29 | 2024-02-06 | 华云天下(南京)科技有限公司 | Intelligent dialogue method and device based on combination of RPA and AI and electronic equipment |
-
2011
- 2011-05-24 KR KR1020110049080A patent/KR20120130962A/en not_active Application Discontinuation
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20180023921A (en) * | 2018-02-06 | 2018-03-07 | 노유현 | Apparatus and method for providing personalication information |
| KR20230053444A (en) | 2021-10-14 | 2023-04-21 | 승문수 | Metaverse-based tendency classification display system |
| CN117520522A (en) * | 2023-12-29 | 2024-02-06 | 华云天下(南京)科技有限公司 | Intelligent dialogue method and device based on combination of RPA and AI and electronic equipment |
| CN117520522B (en) * | 2023-12-29 | 2024-03-22 | 华云天下(南京)科技有限公司 | Intelligent dialogue method and device based on combination of RPA and AI and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11875363B2 (en) | Method and system for generating recommendations | |
| TWI534733B (en) | Based on interpersonal distance ranking, search methods and devices | |
| JP5945369B2 (en) | Method and system for recommending target article information | |
| US8768759B2 (en) | Advertising based on influence | |
| US7908184B2 (en) | Method of providing customized information of commodity for on-line shopping mall users | |
| JP4005100B2 (en) | Advertisement distribution system and apparatus, method, and advertisement distribution program | |
| Xu et al. | Integrated collaborative filtering recommendation in social cyber-physical systems | |
| JP6753578B2 (en) | Methods and equipment for selecting and recommending presentation targets on electronic distribution platforms | |
| CN103391302B (en) | Information sending method and system | |
| US8856177B2 (en) | Personal profile data repository | |
| CN102903047A (en) | Method and device for releasing commodity information | |
| KR101737919B1 (en) | Advertisement method using algorithm of selecting user for merchandise experience through sns | |
| WO2014008765A1 (en) | Network transaction platform system and method having merchandise subscription structure | |
| CN107808314B (en) | User recommendation method and device | |
| Lee et al. | Trustor clustering with an improved recommender system based on social relationships | |
| US20230053063A1 (en) | Statistics-aware sub-graph query engine | |
| JP2004326634A (en) | Recommended data selecting device using confidential relationship between users | |
| KR20120130962A (en) | System and method for analyzing similar inclination | |
| CN109829593B (en) | Credit determining method and device for target object, storage medium and electronic device | |
| US20160070806A1 (en) | A system and method for providing organized search results on a network | |
| JP4962950B2 (en) | Recommendation method, recommendation server and program for users on network | |
| Han et al. | Analyze users' online shopping behavior using interconnected online interest-product network | |
| JP7189277B1 (en) | Provision device, provision method, and provision program | |
| CN114118651A (en) | Evaluation method, device, equipment and computer storage medium | |
| Malik et al. | A new personalized approach in affiliate marketing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A201 | Request for examination | ||
| E902 | Notification of reason for refusal | ||
| E601 | Decision to refuse application |