KR102330724B1 - Method and apparatus for measuring influence based on social network - Google Patents

Abstract

A method and apparatus for measuring influence based on a social network are provided. The method comprises the steps of: collecting SNS data from a social network service (SNS); and a user influence index, attribute influence value, and Including; measuring the product influence value; and curating content including product information based on the product influence value; It is characterized in that it is measured based on the number of comments per post.

Description

Method and device for measuring influence based on social networks

The present invention relates to a method and apparatus for measuring influence based on a social network.

A social network service (SNS) is being provided that supports online relationships and communication with unspecified people. In addition, due to the widespread use of smartphones, online accessibility is increasing, and accordingly, the use rate of SNS services is also increasing. These SNS services began to function as a new purchase channel, and the contents existing on SNS also began to be used for marketing purposes.

On the other hand, the patent document (US Patent Publication No. 2017-0132688) provides a method of calculating an ad price based on the influence of an influential user on SNS, that is, an influencer. However, there is no mention of a method for directly evaluating content. Therefore, there is a need for a method to evaluate how much value content can be utilized on SNS.

US Patent Publication No. 2017-0132688, 2017.05.11

An object of the present invention is to provide a method and apparatus for measuring influence based on a social network.

SUMMARY OF THE INVENTION An object of the present invention is to provide a method and apparatus for dynamically evaluating the influence of a product based on the influence and image tagging on a social network.

SUMMARY OF THE INVENTION An object of the present invention is to provide a method and apparatus capable of recommending products and posting advertisements by dynamically reflecting influence on a social network.

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

The present invention for solving the above problems is a method for measuring influence performed by an apparatus for measuring influence using a social network, the method comprising: collecting SNS data from a social network service (SNS); Measuring influence index, attribute influence value, and product influence value, and curating content including product information based on the product influence value; including, wherein the user influence index is the SNS It is characterized in that it is measured based on the number of followers of the corresponding user and the number of comments per post.

In this case, the influence measuring device measures the user influence value through (the number of followers of the corresponding user compared to the average number of followers of all users) * (the root mean square (RMS) of the number of comments per post), (the user influence Index) * (the attribute is an item extracted from an image included in a post within a preset period), the influence value of the attribute can be measured.

In addition, the influence measuring apparatus, when measuring the influence value of the attribute, obtains image information within a preset period related to the user's account information from the SNS data, and obtains image information in a learning image generated based on the image information At least one attribute included in the image information may be extracted by applying tagging, and an attribute influence value for each of the at least one attribute may be measured.

In addition, when extracting the at least one attribute, the influence measuring apparatus derives at least one influential user based on the user influence index, and extracts the at least one influential user from content information related to the account information of the at least one influential user. Obtain image information within a preset period, extract the at least one attribute, and measure the attribute influence value for each of the at least one attribute based on the user influence index for each of the at least one influential user have.

In addition, when measuring the influence influence value of the attribute, the influence influence value of the attribute is a ratio of the at least one attribute in the corresponding class based on the corresponding class to which each of the at least one attribute belongs. can be normalized, and the attribute and the corresponding attribute influence value can be mapped and stored in an attribute database.

In addition, the influence measuring apparatus, when measuring the product influence value, obtains the content including the product information from the user, and applies image tagging to a learning image generated based on the content including the product information to extract at least one product attribute included in the product information, obtain an attribute influence value corresponding to each of the at least one product attribute from the attribute database, and based on the obtained normalized value of the attribute influence value, By reflecting the weight, it is possible to measure the product influence value for the content including the product information.

In addition, the influence measuring apparatus, when measuring the product influence value, maps the product attribute and the product influence value corresponding thereto, respectively, and stores it in a product attribute database, and when acquiring content including the product information , content having a product influence value matching a preset condition is detected and obtained from the product attribute database, wherein the preset condition is set based on the ranking of the product influence value, or product information included in the content and It may be set based on a similar degree.

Also, the influence measuring apparatus may collect and update new SNS data at regular intervals from the social network service when collecting the SNS data.

In addition, the social network-based influence measuring apparatus according to the present invention includes a data collection unit in which a data collection unit collects SNS data from a social network service (SNS), and a user influence index, attribute influence value and An influence measuring unit for measuring the product influence value, and a content curation unit for curating content including product information based on the product influence value, wherein the user influence index is the user's It is characterized in that it is measured based on the number of followers and the number of comments per post.

According to the present invention, it is possible to evaluate the influence on the content based on the influence of the influencer on the SNS. In addition, by continuously monitoring the content of the influencer on the SNS, it is possible to evaluate the influence of the content in real time, and through this, products or advertisements related to the content can be provided in a timely manner.

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

1 is a diagram schematically illustrating an influence measurement system using a social network according to an embodiment of the present invention.
2 is a diagram showing the configuration of an apparatus for measuring influence using a social network according to an embodiment of the present invention.
3 is a flowchart schematically illustrating a method for measuring influence using a social network according to an embodiment of the present invention.
4 is a diagram illustrating an example of applying a method for measuring influence using a social network according to an embodiment of the present invention.
5 and 6 are diagrams illustrating another example in which the method for measuring influence using a social network according to an embodiment of the present invention is applied.
7 is a diagram illustrating another example of applying a method for measuring influence using a social network according to an embodiment of the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully understand the scope of the present invention to those skilled in the art, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used herein will have the meaning commonly understood by those of ordinary skill in the art to which this invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless specifically defined explicitly.

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

1 is a diagram schematically illustrating an influence measurement system using a social network according to an embodiment of the present invention.

Referring to FIG. 1 , an influence measuring system 10 (hereinafter referred to as an influence measuring system) using a social network according to an embodiment of the present invention is a social network service (hereinafter referred to as SNS) server 100, a terminal device ( 200 ), and may include an influence measuring device ( 300 ).

The SNS server 100 is a server device of a service provider (eg, Twitter, Facebook, Instagram, Kakao Story, etc.) that provides SNS, and provides social networking and information exchange services to SNS users through the terminal device 200 . can do. In addition, the SNS server 100 allocates an SNS account to users (eg, individual users, content providers, etc.) who want to use SNS, and posts various contents through the assigned account or inquires contents posted by others. To provide access to a variety of services. For convenience of explanation, one SNS server 100 is illustrated in FIG. 1 , but the SNS server 100 in the present specification includes various SNS servers constructed for each SNS service provider.

The terminal device 200 is a terminal device of a user using the SNS provided through the SNS server 100 , and may be, for example, various computing devices such as a smart phone, a tablet PC, a desktop, a notebook computer, and a smart TV.

The influence measuring device 300 may collect various SNS data posted on the SNS from the SNS server 100, and through this, evaluate the influence on various information included in the SNS data. In addition, the influence measuring apparatus 300 may provide curated content to users with various purposes based on the influence evaluation on various information included in the SNS data. A detailed operation process for the influence measuring apparatus 300 will be described later.

2 is a diagram showing the configuration of an apparatus for measuring influence using a social network according to an embodiment of the present invention.

Referring to FIG. 2 , an apparatus 300 for measuring influence using a social network according to an embodiment of the present invention (hereinafter referred to as an influence measuring apparatus) includes a data collection unit 310 , an influence measurement unit 320 , and content curation. It may include a unit 330 .

The data collection unit 310 may collect various SNS data posted on the SNS from the SNS server 100 . In an embodiment, the data collection unit 310 includes an account information collection unit 311 for collecting account information of an SNS user from the SNS server 100, and a content information collection unit for collecting content information posted by the SNS user ( 312) may be included.

The influence measuring unit 320 may evaluate the influence of various types of information included in the SNS data. In an embodiment, the influence measurement unit 320 may include a user influence measurement unit 321 , an attribute influence measurement unit 322 , and a product influence measurement unit 323 .

The user influence measurement unit 321 may measure the user influence index based on the SNS user account information and related content information collected by the data collection unit 310 . In this case, the influence measuring apparatus 300 may map each SNS user account information and a corresponding user influence index and store it in the user account database 340 .

The attribute influence measurement unit 322 may measure an attribute influence value for an attribute included in the image information based on image information among the contents collected by the data collection unit 310 . In this case, the influence measuring apparatus 300 may map the attribute and the corresponding attribute influence value and store it in the attribute database 341 .

The product influence measurement unit 323 may measure a product influence value for product information based on a user influence index and an attribute influence value for content including product information. In this case, the influence measuring apparatus 300 may map product information and a product influence value corresponding thereto and store it in the product attribute database 342 .

The content curation unit 330 may curate content including product information based on the product influence value and provide it to users with various purposes.

The operation process of each component of the influence measuring apparatus 300 shown in FIG. 2 will be described in detail below.

3 is a flowchart schematically illustrating a method for measuring influence using a social network according to an embodiment of the present invention. The method of FIG. 3 may be performed by the influence measuring apparatus 300 of FIG. 2 .

Referring to FIG. 3 , the data collection unit 310 may collect various SNS data posted on the SNS from the SNS server 100 ( S400 ).

In an embodiment, the account information collection unit 311 may collect account information of the SNS user registered in the SNS server 100 . The content information collecting unit 312 may collect various content data posted on the SNS from the SNS server 100 . In this case, the account information collection unit 311 and the content information collection unit 312 may collect and update user account information and content information at regular intervals, and thus may maintain recent SNS data.

The influence measuring unit 320 may measure a user influence index, an attribute influence value, and a product influence value based on the collected SNS data (S410 to S430).

More specifically, the user influence measurement unit 321 may derive a user influence index based on the SNS user's user account information and related content information (S410). Here, the user account information may include a user account and personal information related to the user account (eg, followers, following, etc.). Content information related to user account information includes additional information such as the number of content posted through that user account and interest in content posted through that user account (e.g. rating information such as number of comments, likes, etc.) can do.

As an embodiment, the user influence index may be derived as in Equation 1, which may be used as a criterion for evaluating how much influence the user has on the SNS.

Here, n is the total number of users, followers is the number of followers of the user account, post is the number of posted content, and engagements may be the interest (number of likes, comments, etc.) for the posted content. Therefore, Equation 1 represents (the number of followers of the corresponding user compared to the average number of followers of all users) * (root mean square (RMS) of interest per post).

For example, user pool n(user) = 2, item = {coat, skirt, padding}, color = {beige, gray}, user A has 5,000,000 followers, 3 posts, likes and Assume that the number of comments is 1150, 1300, and 1000 respectively, user B has 2,000,000 followers, 2 posts, and 1000 and 500 likes and comments on each post respectively. The user influence index for user A and user B can be calculated as follows when Equation 1 is applied.

The user influence measurement unit 321 may map the user account information derived as described above and the corresponding user influence index and store it in the user account database 340 .

The attribute influence measuring unit 322 may derive the attribute influence value for the attribute included in the image information based on the image information among the content data ( S420 ).

In an embodiment, the attribute influence measuring unit 322 may obtain recent content data related to user account information from the content information collecting unit 312 . According to an embodiment, the attribute influence measuring unit 322 searches for and derives at least one influential user (influencer) from the user account database 340 based on the user influence index, and derives at least one You can obtain data on recent content published through the user accounts of influential users.

The image processing unit 324 may obtain image information from recent content data and extract at least one attribute included in the image information by applying an image tagging technology to the obtained image information. For example, the image processing unit 324 forms a trained model using deep learning, a neural network, etc. for image information, and applies image tagging to the learned model of this image to include it in image information At least one attribute can be extracted.

The attribute influence measuring unit 322 may derive an attribute influence value for each of at least one attribute. In one embodiment, the attribute influence value can be derived as shown in Equation 2, which can be used as a criterion for evaluating how frequently the corresponding attribute appears in the content of influential users and has high impact. .

According to Equation 2, the attribute influence measurement unit 322 determines the attribute influence value based on the user influence index for each of the at least one influential user and the frequency of each attribute included in the image of the at least one influential user. derive

For example, for user A, the user influence index is 1652.1, the attributes extracted from the image included in the recent post are item = {coat 2, skirt 1}, color = {gray 2, beige 1}, for user B, the user Assume that the influence index is 451.8, and the attributes extracted from images included in recent posts are item = {long padding 1, skirt 1}, color = {beige 2}. The attribute influence value for each of the attributes based on the influential users A and B may be calculated as follows according to Equation (2).

According to an embodiment, the attribute influence measurement unit 322 may normalize the attribute influence value based on the class to which the corresponding attribute belongs. As an embodiment, the process of normalizing the attribute influence value can be derived as shown in Equation 3, which indicates which attribute within one class is by adjusting the influence value of the corresponding attribute within the class to which the attribute belongs. The degree of influence can be measured.

According to Equation 3, the attribute influence measuring unit 322 normalizes the attribute influence value by the ratio occupied by the attribute in the class to which the attribute belongs.

For example, in the class composed of items, the influence value of the attribute of the coat attribute is 3304.2, the influence value of the attribute of the skirt attribute is 2103.9, the influence value of the attribute of the long padding attribute is 451.8, and the influence value of the attribute of the gray attribute is the influence value of the attribute in the class composed of colors Assume that 3304.2, the attribute influence value of the beige attribute is 2555.7. If the attribute influence value for each attribute is normalized according to Equation 3, it can be calculated as follows.

The attribute influence measurement unit 322 may map the attribute derived as described above, the attribute influence value corresponding thereto, and the normalized value of the attribute, and store it in the attribute database 341 .

The product influence measurement unit 323 may derive the product influence value for the product included in the content based on the user influence index and the attribute influence value for the content including product information (S430).

In one embodiment, the product influence measuring unit 323 may obtain content including product information from the user. For example, the product influence measuring unit 323 may obtain content including product information from a user who sells a product online through an online shopping mall, a user who wants to advertise the product, or the like. In this case, the user may separately build and use a database 343 for storing his/her product information, and accordingly, the influence measuring apparatus 300 may obtain product information from the database 343 of each user.

The image processing unit 324 obtains image information from content including product information, and applies image tagging through deep learning, a neural network, etc. to the acquired image information to extract at least one product attribute from the image information. .

The product influence measurement unit 323 may measure the product influence value for content including product information based on the extracted attribute influence value for each of the at least one product attribute. In this case, the product influence measurement unit 323 may obtain an attribute influence value corresponding to the product attribute from the attribute database 341 to derive the product influence value. As an embodiment, the product influence value may be derived as shown in Equation 4, which may be used as a criterion for evaluating how influential the content including the product is and reflects the trend.

According to Equation 4, the product influence measuring unit 323 calculates a weight (eg, the degree of interest, importance of the class to which the product attribute belongs) based on the normalized value of the attribute influence value for each product attribute included in one content. value) to derive the product influence value. Here, the weight can be variably changed through empirical data.

For example, the normalized value of the attribute influence value for coat is 0.9433, the normalized value of the attribute influence value for skirt is 0.1051, the normalized value of the attribute influence value for long padding is -1.0484, and the normalized value of the attribute influence value is gray. If the value is 0.7071, the normalized value of the attribute influence value for beige is -0.7071, the weight of the item class is 1, and the weight of the color class is 0.5, the first content obtained from the user has item = {coat}, color = Assume that a product attribute of {beige} is included, and the second content includes a product attribute of item = {skirt} and color = {gray}. The product influence value for the product included in the first content and the second content may be calculated as follows according to Equation (4).

The product influence measurement unit 323 may map the product attribute derived as described above and the product influence value corresponding thereto and store it in the product attribute database 342 .

The content curation unit 330 may curate content including product information based on the product influence value (S440).

In an embodiment, the content curator 330 may detect product content having a product influence value that meets a preset condition from the product attribute database 342 and provide it to a user who wants to utilize it. For example, when the ranking of product influence value is set as a condition, the content curation unit 330 acquires, from the product attribute database 342, a product attribute having a high product influence value among various contents provided by the user, Content curated with product attributes can be provided. Alternatively, when the similarity of products is set as a condition, the content curation unit 330 acquires product attributes similar to products included in the content provided by the user from the product attribute database 342, and It can also be used for content curation by calculating the similarity.

4 is a diagram illustrating an example of applying a method for measuring influence using a social network according to an embodiment of the present invention.

Referring to FIG. 4 , the data collection unit 310 may collect new SNS data every preset period ( S500 ). According to an embodiment, the data collection unit 310 may detect an influential user based on the user influence index from the user account database 340 and periodically collect the detected user's SNS data.

The influence measuring unit 320 may obtain an image from the collected SNS data and extract at least one attribute included in each image (S510). According to an embodiment, the influence measuring unit 320 may extract attributes from only the newly collected image from the SNS data of the influential user.

The influence measuring unit 320 may measure an attribute influence value for an attribute included in each image by using the user influence index for each influential user ( S520 ).

For example, as shown in FIG. 4 , the influence measuring unit 320 extracts coat, beige, and belt attributes from the first image 501 of the first user, and the second image 502 of the second user. You can extract wrap skirt, gray, and destroyed attributes from The influence measuring unit 320 may calculate the influence value of each attribute by using the user influence indices of the first and second users for each of the extracted attributes.

The influence measuring unit 320 may store each attribute and the corresponding attribute influence value in the attribute database 341 ( S530 ).

When the attribute database 341 is updated, the influence measurement unit 320 may derive a product influence value by using the attribute and the attribute influence value updated in the attribute database 341 , and accordingly, the product attribute database 342 . can be updated (S540).

In addition, when content including product information is obtained from the user, the influence measuring unit 320 measures the product influence value for the product included in the user's content by using the attribute and the attribute influence value stored in the attribute database 341. There is (S550).

According to an embodiment of the present invention, since influence is evaluated by periodically collecting new SNS data, product influence can be continuously updated, and thus the latest trends can be effectively reflected and utilized in a timely manner.

5 and 6 are diagrams illustrating another example in which the method for measuring influence using a social network according to an embodiment of the present invention is applied.

5 and 6 , the influence measuring unit 320 may obtain content including product information from the user product database 343 ( S600 ). In an embodiment, the influence measuring unit 320 may acquire content from the user product database 343 in units of stock keeping units (SKUs).

The influence measuring unit 320 extracts an attribute from each of the acquired contents (S610), measures the attribute influence value for each of the extracted attributes, and then measures the product influence value for each content based on the influence influence value of each attribute. There is (S620).

The content curation unit 330 may determine whether or not a preset influence value is satisfied based on the product influence value of each acquired content, and may filter the contents meeting the condition according to the determination result ( S630 ). For example, it is possible to select and filter content having a high ranking of influence value.

The content curator 330 may curate the filtered content and provide it to the user (S640). For example, it is possible to reflect and provide a product with the highest influence on a social network among contents owned by the user.

As described above, according to the update of the attribute database 341 and the product attribute database 342 , the attribute influence value and the product influence value may be changed. In this case, the content curator 330 may filter the content by determining whether it meets a preset condition again based on the product influence value updated from the product attribute database 342 . Accordingly, it is possible to continuously monitor the trend change on the social network, and accordingly, the content reflecting the trend can be dynamically updated.

7 is a diagram illustrating another example of applying a method for measuring influence using a social network according to an embodiment of the present invention.

Referring to FIG. 7 , the influence measuring unit 320 may acquire content including new product information from the user ( S700 ).

The influence measuring unit 320 extracts a product attribute from the content including the new product information (S710), and based on the attribute influence value for each of the extracted product attributes, the product influence value of the content including the product information can be measured. There is (S720).

Here, the influence measuring unit 320 may evaluate the product influence value by comparing a product included in the content obtained from the user and a similar product.

In an embodiment, the influence measuring unit 320 may obtain a product 701 similar to a product attribute in the content obtained from the user from the product attribute database 342 ( S730 ). The influence measuring unit 320 extracts product attributes 702 and 703 from each of the similar products 701 (S740), and the degree of similarity between each of the extracted product attributes 702 and 703 and each product attribute of the content obtained from the user can be compared (704, 705) (S750). According to the degree of similarity 704 and 705 between these attributes, it is possible to derive the degree of similarity of each similar product. The influence measuring unit 320 may measure the product influence value by reflecting the product similarity for each similar product 701 .

The content curator 330 may determine whether the content is similar to the content obtained from the user based on the product influence value reflecting the product similarity, and may filter the content including the similar product according to the determination result (S760).

The content curator 330 may curate the filtered content and provide it to the user (S770).

According to the present invention, since the influence can be evaluated by reflecting the similarity of products, it can be effectively used when recommending similar products or matching similar products, and can be used, for example, when posting advertisements.

The steps of a method or algorithm described in relation to an embodiment of the present invention may be implemented directly in hardware, as a software module executed by hardware, or by a combination thereof. A software module may contain random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any type of computer-readable recording medium well known in the art to which the present invention pertains.

As mentioned above, although embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains can realize that the present invention can be embodied in other specific forms without changing its technical spirit or essential features. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

Claims (10)

In the influence measurement method performed by the influence measurement apparatus using a social network,
collecting SNS data from a social network service (SNS);
measuring a user influence index, an attribute influence value, and a product influence value based on the SNS data; and
Including; curating content including product information based on the product influence value;
The user influence index is measured based on the number of followers of the corresponding user of the SNS and the number of comments per post,
The influence measuring apparatus, when measuring the influence value of the attribute, obtains image information within a preset period related to the user's account information from the SNS data, and image tagging the learning image generated based on the image information extracting at least one attribute included in the image information by applying it, and measuring an attribute influence value for each of the at least one attribute;
When measuring the influence influence value of the attribute, the influence measuring apparatus normalizes the influence influence value of the attribute by a ratio occupied by each of the at least one attribute in the corresponding class based on the corresponding class to which each of the at least one attribute belongs A method of measuring influence based on social networks. According to claim 1,
The influence measuring device,
(The number of followers of that user compared to the average number of followers of all users) * (Root mean square (RMS) of the number of comments per post) measures the user influence value
(The user influence index) * A social network-based influence measurement method that measures the influence value of the attribute through (item, which is an attribute extracted from an image included in a post within a preset period). delete According to claim 1,
The influence measuring device, when extracting the at least one attribute,
Derive at least one influential user based on the user influence index,
obtaining image information within a preset period from content information related to the account information of the at least one influential user, and extracting the at least one attribute;
A social network-based influence measurement method for measuring an attribute influence value for each of the at least one attribute based on a user influence index for each of the at least one influential user. delete According to claim 1,
The influence measuring device, when measuring the attribute influence value,
A method for measuring influence based on a social network, wherein the attribute and the corresponding attribute influence value are mapped and stored in an attribute database. 7. The method of claim 6,
The influence measuring device, when measuring the product influence value,
Obtaining content including the product information from the user,
extracting at least one product attribute included in the product information by applying image tagging to a learning image generated based on the content including the product information;
obtaining an attribute influence value corresponding to each of the at least one product attribute from the attribute database;
A social network-based influence measuring method of measuring a product influence value for content including the product information by reflecting a weight based on the obtained normalized value of the attribute influence value. 8. The method of claim 7,
The influence measuring device,
When measuring the product influence value, the product attribute and the product influence value corresponding thereto are respectively mapped and stored in a product attribute database,
When acquiring the content including the product information, by detecting and acquiring content having a product influence value that meets a preset condition from the product attribute database,
The predetermined condition is
A social network-based influence measurement method that is set based on the ranking of the product influence value, or is set based on a degree similar to product information included in the content. According to claim 1,
The influence measuring device, when collecting the SNS data,
A social network-based influence measurement method for collecting and updating new SNS data at regular intervals from the social network service. a data collection unit that the data collection unit collects SNS data from a social network service (SNS);
an influence measuring unit for measuring a user influence index, an attribute influence value, and a product influence value based on the SNS data; and
A content curation unit that curates content including product information based on the product influence value;
The user influence index is measured based on the number of followers of the corresponding user of the SNS and the number of comments per post,
When measuring the attribute influence value, image information within a preset period related to the user's account information is obtained from the SNS data, and image tagging is applied to the learning image generated based on the image information to the image information. extracting the included at least one attribute, measuring an attribute influence value for each of the at least one attribute,
When measuring the influence influence value of the attribute, the attribute influence value is normalized by the ratio occupied by each of the at least one attribute in the corresponding class based on the corresponding class to which each of the at least one attribute belongs. influence measuring device.

Images