KR20210089887A - System for distributing profit of review contents - Google Patents

Abstract

The present invention relates to a system for distributing the profits of review content and, more specifically, to a system for distributing the profits of review content, which extracts the contribution of review content by analyzing a review content usage pattern of a consumer before a product is purchased in a shopping platform. The system comprises: a weight calculation unit for calculating and quantifying the contribution of a review video according to a predetermined algorithm in order to evaluate the contribution of the review video to sales that is provided to an Internet shopping platform; and a fee calculation unit for setting fees according to the contribution of the video. The weight calculation unit comprises: a variable setting module for setting, as a variable, one selected from the playback time of a video provided on a video page, the number of subscribers therefor, the number of "like" clicks, the number of recommendations, or the number of time of times of playback; a variable calculation module for calculating the value of the variable for all videos in the platform; a variable calculation module for calculating the mean and standard deviation of the variable calculated for each video; a standard score calculation module for calculating a standard score through the mean and standard deviation calculated by the variable calculation module; a distance calculation module for calculating each Euclidean distance by using the standard score of any video among all the videos and the standard score of each of the remaining videos and summing the same; a video selection module for finding a video of which a value summed by the distance calculation module is lowest among all the videos; and a weighting factor module for setting a weighted factor for each variable according to a preset criterion.

Description

System for distributing profit of review contents

The present invention relates to a review content revenue distribution system, and more particularly, to a review content revenue distribution system for extracting the contribution of review content by analyzing a consumer's review content utilization pattern before purchasing a product in a shopping platform.

With the proliferation of PCs and smartphones, the online shopping mall market is gradually becoming huge. As the center of the distribution market is rapidly shifting from offline to online, the existing online shopping industry and traditional distribution giants are also making efforts to take the lead in the online shopping market.

Characteristics of the current online shopping market include the continued entry of new strong players in the fusion of open market and social commerce, and the shift in growth paradigm from PC-centered markets to mobile shopping. In addition, the shift from price competition to customer value is also a key feature.

As of 2017, the domestic online shopping transaction amount is estimated to be 78.2 trillion won, which is a 7.3-fold increase from 10.6 trillion won in 2005, or an average annual growth of about 20%. The domestic online shopping market is growing about 2.3 times in a five-year cycle. .

According to a research report from the Information and Communication Policy Institute, the online shopping market is expected to reach a minimum of 176 trillion won and a maximum of 190 trillion won by 2022 if the current trend continues. Thanks to the explosive growth of the online shopping market, its share in the overall shopping market is also increasing.

Accordingly, each company is also presenting various marketing plans to promote online shopping. Major marketing measures include ‘VR Store’ that applies virtual reality technology, ‘deep scan’ that recommends products using artificial intelligence (AI) technology, and makeup service using augmented reality.

On the other hand, one of the most important factors for actual consumers to make purchase decisions in online shopping is ‘review’. It was found that 79% of consumers who enjoy online shopping make a purchase decision based on product reviews.

For example, it was found that the first factor that consumers consider when purchasing cosmetics is the efficacy and effect of the product, and the second is reviews. This can be seen as a result of reflecting the recent trend toward substantial value consumption centered on the younger generation.

In addition, as the number of smart consumers increases, the influence of ‘review’ on consumption is getting stronger day by day. There was also a survey that more than half of Korean consumers shared their experience of using products or services they used with others around them. In particular, online sharing was more common than offline sharing, and the sharing methods were found to be in the following order: writing reviews for online shopping malls, sharing computer messengers, sharing SNS, writing reviews on blog boards, and reviewing company and brand websites.

As such, the biggest reason consumers rely on reviews is the abundance of information. In general, the online shopping market is overflowing with advertisements and marketing that unilaterally promote products. Because only the merits of the product are emphasized, the actual purchased product often does not meet expectations, and those who have experienced it are changing into smart consumers.

Rather than being swayed by information swarming, consumption is decided by meticulously checking product reviews posted by other consumers after using them.

There are already many examples of strategic use of reviews across industries. For example, there is a service that finds a restaurant through a review that evaluates a restaurant, and a service that informs the satisfaction of hospital treatment, friendliness, and facility satisfaction through reviews.

As the influence of reviews of other consumers on purchases increases in online shopping, the importance of data accumulation and management of actual product reviews is increasing. According to market research firm Muchneeded, 68% of consumers believe that positive online reviews increase their confidence in a product, and more than 50% of 18-34 year olds believe that positive online reviews play a decisive role in purchasing rather than acquaintance recommendations. has been investigated.

Therefore, sellers are incentivizing them to write reviews by spending money such as reserves and coupons to secure high-quality reviews. However, the amount of writing is often small compared to the effort put in. This is because it is difficult to accurately analyze the contribution of product reviews, so a comprehensive compensation plan is being maintained. According to this situation, the consumer does not have the motivation to leave a detailed review, and even a situation in which no review is left at all occurs frequently.

On the other hand, with the explosive growth of video platforms, product reviews through video are also increasing gradually. In the case of young people, the proportion of searching for videos to obtain desired information is gradually increasing, and unlike texts, they tend to prefer videos that can be directly viewed and heard and experienced indirectly. In other words, video review itself can be marketing content for product sales.

However, it is not common for real consumers to actively create and leave a video review for the reasons described above. The fact that consumers are not rewarded enough for the time and effort they put in to create a video and leave a review, and that there is no system that calculates the contribution to sales and provides compensation even if the seller wants to give a reasonable compensation to overcome this Because there is a problem.

The prior art Republic of Korea Patent Registration No. 10-0457670 relates to a revenue compensation distribution system according to the user's content use, and the member identification information, the client execution module installation information for each member, and the client execution module generated by the A member information database that stores flower garden information including contribution point information for each member, content use information generated by the execution of the client execution module installed in the user's computer, and the member who installed the corresponding client execution module are searched for, and the content usage information A client execution module that executes the content provided by the content providing server by including a contribution point reflection unit that determines the contribution score of the member who has installed the client execution module from the member and reflects the determined contribution score in the member's contribution score information It includes a configuration that enables the distribution of the client execution module to be activated by making the installer who installs the installation process a predetermined compensation for the distribution of the client execution module, but as described above, the problem of distribution of revenue through video contribution determination was difficult to solve.

In another prior art, Korean Patent Application Laid-Open No. 10-2017-0105309, user registration is done in a recommendation-subscribed spillover method that allows new users recommended by them to be included under their umbrella and automatically fills vacancies. , When a specific person generates points, in network marketing in which users of a higher organization or a lower organization share points, the distribution of points is distributed so that the distribution of points is not concentrated on any one user. It relates to a profit distribution method of network marketing. , when points are generated for a specific user, the number of virtual avatars corresponding to the points is infinitely accumulated every week until the deadline, and the previously accumulated virtual avatars are arranged in the sub-organizations of the specific user in the future. Points are distributed as evenly as possible to all users included in the lower ranks, and in arranging the avatar, the random arrangement method is alternately performed in addition to the fixed arrangement method at regular time intervals, thereby allowing users to obtain higher profits. Although disclosed that can induce the creation of, there was a difficulty in solving the above-mentioned problems.

Republic of Korea Patent Registration No. 10-0457670 Republic of Korea Patent Publication No. 10-2019-0105309

An object of the present invention is to provide a system capable of calculating the contribution of a specific review to product sales.

Another object of the present invention is to provide a system capable of distributing revenue according to the contribution of each review to product sales.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems to be solved by the present invention not mentioned here are those of ordinary skill in the art to which the present invention belongs from the description below. can be clearly understood by

Review content revenue, including a weight calculator that calculates and quantifies the contribution of a video according to a predetermined algorithm in order to evaluate the sales contribution of a review video provided to an internet shopping platform, and a fee calculator that sets a fee according to the contribution of the video In the distribution system, the weight calculation unit comprises: a variable setting module for setting as a variable one selected from the play time, the number of subscribers, the number of 'like' clicks, the number of recommendations and the number of reproductions of the video provided to the video page; A variable calculation module for inputting and storing the value of the variable for all videos in the platform, a variable calculation module for calculating the average and standard deviation for the value of the variable input and stored for each video, and the variable calculation module A standard score calculation module that calculates a standard score through mean and standard deviation, a distance calculation module that calculates each Euclidean distance using the standard score of any video among all videos and the standard score of each remaining video, and adds them all together; , a video selection module for finding a video whose sum value of the distance calculation module is the lowest value among all videos, and a weighting factor module for setting a weighted factor for each variable according to a preset criterion.

Preferably, the variable calculation module is

는 설정된 변수에 대한 해당 동영상의 변수 값이고,

는 도수이고,

은 확률이므로

는

을 확률로 나타낸 것이다.)(where N is the total number of videos,

is the variable value of the video for the set variable,

is the frequency,

is the probability

is

is expressed as a probability.)

This includes calculating the mean and standard deviation for all videos through

Preferably, the standard score calculation module is based on the result value calculated by the variable calculation module, the following formula

(Where Z is the Z score, X is the raw score, M is the mean, and σ is the standard deviation.)

It can be characterized by calculating the standard score.

Preferably, the Euclidean distance calculated by the distance calculation module is expressed by the following equation

는 동영상의 기 설정된 방식으로 매겨진 동영상 고유번호이고,

는 동영상의 표준점수이다.)(here,

is a video unique number assigned in a preset method of the video,

is the standard score of the video.)

It can be characterized by calculating through .

Preferably, the fee calculation unit may include a video log module for tracking the video history watched by the orderer and a fee calculation module for calculating the fee by evaluating the influence of the video in a preset manner.

Preferably, the fee calculation module is expressed by the following equation

은 총 시청 횟수이고,

은 시청 횟수이고,

은 총 동영상 시간이고,

은 재생시간이고,

은 총 라스트 클릭이고,

은 라스트 클릭이고,

은 상품 n에 m번째 동영상이 미친 영향력이고,

은 동영상이 상품 구매 n 구매에 미친 영향력이고,

은

의 주문에

의 스트리밍 수수료이고,

은

의 주문에

의

스트리밍 수수료이고,

는 동영상 변수

의 가중치이다.)(where n is a product variable, m is a video variable,

is the total number of views,

is the number of views,

is the total video time,

is the playing time,

is the total last clicks,

is the last click,

is the influence of the mth video on product n,

is the influence of video on product purchase n purchase,

silver

on the order of

is the streaming fee of

silver

on the order of

of

streaming fee,

is the video variable

is the weight of ).

It may be characterized in that it is calculated through

The review content revenue distribution system according to the present invention has an effect of providing a system capable of calculating the contribution of a specific review video to product sales through a weight calculation unit.

The review content revenue distribution system according to the present invention has an effect of providing a system capable of distributing revenue according to the contribution of each review video to product sales through a fee calculation unit.

1 shows the configuration of a weight calculator according to an embodiment of the review content revenue distribution system according to the present invention.
2 shows the configuration of a fee calculation unit according to an embodiment of the review content revenue distribution system according to the present invention.
3 shows a weighting factor calculation algorithm applied according to an embodiment of the review content revenue distribution system according to the present invention.
Figure 4 shows, according to an embodiment of the review content revenue distribution system according to the present invention, assuming video A and B, and calculating the standard score distance between the two videos through the Euclidean distance formula.
5 is a conceptual diagram illustrating a process of selecting a video in which the sum of each Euclidean distance is the lowest, assuming videos A, B, C, D, and E, according to an embodiment of the review content revenue distribution system according to the present invention. is shown as

Specific details including the problems to be solved for the present invention as described above, the means for solving the problems, and the effects of the invention are included in the embodiments and drawings to be described below. 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.

The review content revenue distribution system according to the present invention, as shown in FIGS. 1 to 5, calculates and quantifies the contribution of the video according to a predetermined algorithm to evaluate the sales contribution of the review video provided to the Internet shopping platform. In the review content revenue distribution system comprising a weight calculation unit (1) and a fee calculation unit (2) for setting a fee according to the contribution of the video, the weight calculation unit (1) comprises: , play time, number of subscribers, number of 'like' clicks, number of recommendations, or number of playbacks as a variable setting module 11, and input and storage of the variable value for all videos in the platform the variable calculation module 12, the variable calculation module 13 for calculating the mean and standard deviation for the values of the variables input and stored for each video, and the average and standard deviation calculated by the variable calculation module 13 The standard score calculation module 14 for calculating the standard score through the standard score calculation module 14, and the distance calculation module 15 for calculating each Euclidean distance using the standard score of any video among the entire video and the standard score of each of the remaining videos and adding them all together (15) and a video selection module 16 that finds a video whose sum of the distance calculation module 15 has the lowest value among all videos, and a weighted factor module that sets a weighted factor for each variable according to a preset criterion (17) may be included.

First, as shown in FIG. 1 , the variable setting module 11 sets as a variable one selected from the play time, the number of subscribers, the number of 'like' clicks, the number of recommendations or the number of plays of the video provided on the video page as a variable. can be set.

That is, in the review content revenue distribution system according to the present invention, a variable means a parameter that can be expressed numerically in each video and is used for calculating the contribution of the video.

Next, the variable calculation module 12 may input and store the value of the variable for all videos in the platform.

Specifically, it includes assigning a variable value of each video to the variable set by the variable setting module 11 .

In one embodiment, it is assumed that there are videos A, B, and C, and when the variable setting module 11 sets the 'play count' as a variable, the variable 'play count' for the videos A, B, and C is set respectively. will be entered

Here, the variable may be the playback time, the number of subscribers, the number of 'like' clicks, and the number of recommendations.

Next, the variable calculation module 13 may calculate the average and standard deviation of the values of the variables input and stored for each video.

Specifically, the variable calculation module 13 is expressed in Equation 1 as follows:

는 설정된 변수에 대한 해당 동영상의 변수 값이고,

는 도수이고,

은 확률이므로

는

을 확률로 나타낸 것이다.)(where N is the total number of videos,

is the variable value of the video for the set variable,

is the frequency,

is the probability

is

is expressed as a probability.)

It is possible to calculate the mean and standard deviation for all videos.

That is, the standard deviation is calculated by finding the mean and variance of the probability distribution. Here, the average is the sum of the given variable values divided by the number of videos.

The variance is the mean of the squares of the deviations, which means the degree to which the values of a variable are spread.

The standard deviation is the positive square root of the variance. As the variance increases, the standard deviation appears larger because the values of each variable are farther from the mean.

Next, the standard score calculation module 14 may calculate a standard score through the average and standard deviation calculated by the variable calculation module 13 .

Specifically, the standard score calculation module 14 is based on the result calculated by the variable calculation module 13 as follows: Equation 2

(Where Z is the Z score, X is the raw score, M is the mean, and σ is the standard deviation.)

can be used to calculate the standard score.

In addition, Z means the standard score of the corresponding video.

The standard score of each video for the variable set by the variable setting module 11 can be obtained through the above formula.

Next, the distance calculation module 15 may calculate each Euclidean distance by using a standard score of any video among all the videos and the standard scores of the remaining videos and sum them all up.

Specifically, the Euclidean distance calculated by the distance calculation module 15 is as follows: Equation 3

는 동영상의 기 설정된 방식으로 매겨진 동영상 고유번호이고,

는 동영상의 표준점수이다.)(here,

is a video unique number assigned in a preset method of the video,

is the standard score of the video.)

can be calculated through

In the same manner as described above, the Euclidean distance of an arbitrary moving image and each of the remaining moving images may be calculated.

As an embodiment, as shown in FIG. 4 , assuming that there are a video A and a video B, the distance between the video A and the video B may be calculated using the Euclidean distance formula for calculating the distance between two points.

Next, the video selection module 16 may find a video in which the sum of the distance calculation module 15 is the lowest value.

The video for which the sum of the Euclidean distances of all videos has the lowest value is the video with the highest importance for that variable.

The video selection module 16 can find a video having the lowest sum of Euclidean distances by comparing the sum of Euclidean distance values for all videos.

As a specific embodiment, as shown in FIG. 5 , assuming that there are videos A, B, C, D, and E, and selecting video A as an arbitrary video, video A and B, C, D, E, respectively Euclidean distance with and can be calculated, and all calculated values are summed. Iteratively, the sum of the Euclidean distance calculations of video B and the rest of the videos, video C and the rest of the videos, video D and the rest of the videos, and video E and the rest of the videos is calculated, and the video with the smallest sum is selected.

That is, it turns for each video and becomes a random video as a standard once, and calculates the Euclidean distance from the rest of the videos. And the video selection module 16 selects the video having the smallest sum of the Euclidean distance calculation values with the remaining videos.

Next, the weighting factor module 17 may set a weighted factor for each variable according to a preset criterion.

As described above, the process of selecting the video having the lowest value of the sum of the Euclidean distances may be repeatedly performed for each variable, that is, for each parameter.

Specifically, the sum of the Euclidean distances is performed for each variable including the play time, the number of subscribers, the number of ‘like’ clicks, the number of recommendations or the number of plays.

Then, a weighting factor is set for each variable according to a preset criterion. That is, different weights are assigned to each variable.

This means that the importance of each variable is assumed differently.

In a specific embodiment, the sum of the Euclidean distances of the video may be obtained for each variable, and the variables may be weighted in the order of having the lowest value.

As another embodiment, the sum of the Euclidean distances of the moving images for each variable may be obtained, and when a common variable of each moving image is the lowest value, a weight may be assigned to each variable.

The reason why the weight factor module 17 sets weights for each variable is to determine which elements of the video to focus on to distribute revenue to distribute revenue when a video review affects product sales.

Next, the commission calculator 2 may be provided to distribute profits according to the sales contribution of the videos.

Specifically, as shown in FIG. 2 , the fee calculation unit 2 may include a video log module 21 and a fee calculation module 22 .

The video log module 21 may track the video history watched by the orderer.

Specifically, in order to classify video reviews that affect product sales, video viewing history is tracked and stored.

In addition, it is possible to set the weight of the video variable for the sale product by calculating the standard score and the weight factor from the video list collected through the video log module 21 .

In addition, the fee calculation module 22 may calculate the fee by evaluating the influence of the video in a preset manner.

In more detail, the fee calculation module 22 is expressed by the following Equation 4

은 총 시청 횟수이고,

은 시청 횟수이고,

은 총 동영상 시간이고,

은 재생시간이고,

은 총 라스트 클릭이고,

은 라스트 클릭이고,

은 상품 n에 m번째 동영상이 미친 영향력이고,

은 동영상이 상품 구매 n 구매에 미친 영향력이고,

은

의 주문에

의 스트리밍 수수료이고,

은

의 주문에

의

스트리밍 수수료이고,

는 동영상 변수

의 가중치이다.)(where n is a product variable, m is a video variable,

is the total number of views,

is the number of views,

is the total video time,

is the playing time,

is the total last clicks,

is the last click,

is the influence of the mth video on product n,

is the influence of video on product purchase n purchase,

silver

on the order of

is the streaming fee of

silver

on the order of

of

streaming fee,

is the video variable

is the weight of ).

can be calculated through

Specifically, the video associated with the actual product to be sold and the video log of the buyer until an order to sell the product is collected, and the influence of each video is calculated and revenue is distributed.

3, the review content revenue distribution system according to the present invention may include the following algorithm.

First, one of the parameters such as the weighting factor target video variable to be obtained, that is, the playback time and the number of playbacks, is set.

Next, variable values are assigned to the entire video data.

Next, the average and standard deviation of the variable values are calculated for each video.

Next, a standard score is calculated through the average and standard deviation of each calculated video variable value.

Next, the Euclidean distance between the standard scores and the rest of the videos is calculated based on any one video among all the videos, and these are summed.

For each video, the Euclidean distance from the rest of the video is calculated and summed independently to find the video with the lowest value.

The relationship between the video having the lowest price and the product and order is identified through a preset standard.

The preset criterion may be different for each variable, and in one embodiment, a variation affecting sales may be set, such as an increase in the order rate when the reproduction time is long, or an increase in the order rate when the number of reproductions is large.

If it is determined that the correlation between the video having the lowest value with the product and the order is out of the preset standard and the correlation is not large, the process of finding the video with the lowest value by changing the first selected variable is repeated. can

Finally, the weighting factor of the video is extracted and used to calculate the influence of the video on the purchase to pay a fair fee.

In other words, it is to find the optimal weight and use it for profit distribution.

As such, those skilled in the art to which the present invention pertains will understand that the above-described technical configuration of the present invention may be implemented in other specific forms without changing the technical spirit or essential characteristics of the present invention.

Therefore, the embodiments described above are to be understood as illustrative and not restrictive in all respects, and the scope of the present invention is indicated by the following claims rather than the above detailed description, and the meaning and scope of the claims and their All changes or modifications derived from the concept of equivalents should be construed as being included in the scope of the present invention.

1: weight calculator
11: Variable setting module
12: variable calculation module
13: variable calculation module
14: standard score calculation module
15: distance calculation module
16: video selection module
17: weight factor module
2: Fee calculator
21 : video log module
22: Fee calculation module

Claims (6)

Review content revenue comprising a weight calculator that calculates and quantifies the contribution of a video according to a predetermined algorithm to evaluate the contribution to sales of review videos provided on an internet shopping platform, and a fee calculator that sets a fee according to the contribution of the video A distribution system comprising:
The weight calculation unit,
a variable setting module for setting as a variable one selected from the play time, the number of subscribers, the number of 'like' clicks, the number of recommendations or the number of plays of the video provided on the video page;
a variable calculation module for inputting and storing the values of the variables for all videos in the platform;
a variable calculation module for calculating mean and standard deviation of the variable values input and stored for each video;
a standard score calculation module for calculating a standard score through the mean and standard deviation calculated by the variable calculation module;
a distance calculation module for calculating each Euclidean distance using the standard score of any video among all the videos and the standard score of each of the remaining videos and summing them up;
a video selection module for finding a video whose sum of the distance calculation module has the lowest value among all videos; and
Review content revenue distribution system comprising a; a weighting factor module for setting a weighted factor for each variable according to a preset criterion. According to claim 1,
The variable calculation module has the following formula

(where N is the total number of videos,

is the variable value of the video for the set variable,

is the frequency,

is the probability

is

is expressed as a probability.)
Review content revenue distribution system, characterized in that for calculating the average and standard deviation for all videos through. 3. The method of claim 2,
The standard score calculation module is based on the result value calculated by the variable calculation module, the following formula

(Where Z is the Z score, X is the raw score, M is the mean, and σ is the standard deviation.)
Review content revenue distribution system, characterized in that for calculating the standard score. 4. The method of claim 3,
The Euclidean distance calculated by the distance calculation module is expressed by the following equation

(here,

is a video unique number assigned in a preset method of the video,

is the standard score of the video.)
Review content revenue distribution system, characterized in that calculated through. According to claim 1,
The fee calculator,
a video log module that tracks the video history watched by the orderer; and
a fee calculation module for evaluating the influence of a video in a preset manner and calculating a fee;
Review content revenue distribution system comprising a. 6. The method of claim 5,
The fee calculation module is expressed by the following equation

(where n is a product variable, m is a video variable,

is the total number of views,

is the number of views,

is the total video time,

is the playing time,

is the total last clicks,

is the last click,

is the influence of the mth video on product n,

is the influence of video on product purchase n purchase,

silver

on the order of

is the streaming fee of

silver

on the order of

of

streaming fee,

is the video variable

is the weight of ).
Review content revenue distribution system, characterized in that calculated through.

Images