KR102319118B1 - Product display update method of shopping mall related to e-commerce and apparatus using said method - Google Patents

Abstract

One aspect of the present invention discloses a method for updating a product display of a shopping mall related to e-commerce. The method includes the steps of calculating a product ranking change index based on the hourly product rankings of at least some of the plurality of products registered in the first shopping mall associated with the e-commerce transaction (here, the product ranking change index is the first time point It is determined by a similarity between the arrangement according to the ranking of at least some of the plurality of products and the arrangement according to the ranking of at least some of the plurality of products at a second time point, wherein the second time point is the first time point. ), calculating a product display update time of a web page for e-commerce of the first shopping mall using the product ranking change index, and based on the calculated product display update time, the first shopping mall updating the product display of

Description

Method and apparatus for updating product display in a shopping mall related to e-commerce

The present invention relates to a product display update method in an e-commerce transaction, and more particularly, to a method for determining an optimal product display update time to promote sales of shopping malls.

In the process of running an online e-commerce (which may be called a shopping mall), shopping malls periodically adjust the order in which products are displayed on an online product page (order of display) in order to promote sales. Products that require active exposure, such as launching new products or discounting, are placed in the upper rank, and products that do not are placed in the lower order to increase sales.

In another example, in the case of online advertising, the advertiser considers the ranking in which his or her advertisement is searched as important, and may pay a higher advertising fee to secure a target ranking by participating in a keyword auction to increase the ranking.

Information displayed in rankings, such as display rankings or search rankings, is subject to constant monitoring, and although it is a very important indicator because it is closely related to sales revenue or advertising expenditure due to changes in the ranking, in order to control it well, the ranking In reality, there are not many suitable methods to quantify the amount of change over time with a single number. In other words, despite the fact that the display ranking or search ranking of a specific product has changed a lot over time, shopping mall operators or advertisers do not clearly understand this as an indicator, so there are cases where they cannot respond quickly. . As a result, there is a problem that results in a decrease in sales performance or decrease in advertising efficiency due to an insufficient response.

An object according to an aspect of the present invention for solving the above-described problems is to optimize product display using a quantified ranking change index that is easy for users to intuitively understand the changed information when the data sorted according to the ranking fluctuates. An object of the present invention is to provide a method and apparatus for updating product display of a shopping mall related to e-commerce for updating.

According to an aspect of the present invention for achieving the above object, there is provided a product display update method of a shopping mall related to an e-commerce transaction, based on a product ranking by time of at least some of a plurality of products registered in a first shopping mall related to an e-commerce transaction, Calculating a product ranking change index (wherein the product ranking change index is an arrangement according to the ranking of at least some of the products of the plurality of products at a first time, and at least some of the products of the plurality of products at a second time (determined by the degree of similarity with the arrangement according to their ranking, and the second time point is a time point subsequent to the first time point) The method may include calculating a display update time and updating the product display of the first shopping mall based on the calculated product display update time.

The method includes calculating a product display update cycle for updating the product display of the first shopping mall based on the product ranking change index, and comparing the calculated product display update cycle with the product display update cycle of the second shopping mall The method may further include a step of recommending to update the display of the first shopping mall based on the step and the comparison result.

A product display update period for updating the product display of the first shopping mall may be set in proportion to a value of the product ranking change index.

The method may include comparing the product ranking change index of the first shopping mall with a preset reference value and generating a product display update notification generated by the product display update time based on the comparison result.

The method further comprises the step of calculating a change rate over time of the product ranking change indicator, wherein the product display update notification generated by the product display update time is higher or lower than a preset reference value in the product display update notification It may be generated in response to a state.

Calculating the product ranking change index includes, based on the time-based product ranking of products existing at at least one _{of a first time point (t 1} ) and a second time point (t _{2 ) among the plurality of products, (i)} The weight (λ), which is a numerical value indicating that the upper rank is considered more important than the lower rank, (ii) the rank at the first time point (Rank(t ₁ )), and (iii) the rank at the second time point (Rank(t)) ₂ )) may include calculating the display similarity for each product and calculating a product ranking change index using the display similarity for each product.

The product ranking by time may include a sales ranking of at least some of the plurality of products displayed in the shopping mall by time or a display ranking of products in an online product page of the shopping mall.

The method further comprises the step of visualizing the product ranking change index over time as a graph of the product ranking change index as the first axis as a time axis and the second axis as a graph of the product ranking change index, wherein a plurality of graphs are divided according to the weight value and displayed on one screen You can control the display.

The product ranking change index and the product display update cycle calculated by the product ranking change index are input, and a learning data set is generated using the product sales quantity achieved by the input value as an output value to correspond to the product ranking change index A deep learning model can be trained to infer the optimal shelf update cycle for

The value of the product ranking change index and the product display update cycle value corresponding thereto are stored in a table, and the product display update cycle is determined by extracting the product display update cycle value corresponding to the calculated product ranking change index value from the table. can decide

The method may include calculating a product display update period of a plurality of shopping malls excluding the first shopping mall using a product ranking change index and using at least one of statistical values related to the product display update period of the plurality of shopping malls. The method may further include calculating a product display update cycle of the first shopping mall.

According to another aspect of the present invention for achieving the above object, there is provided an apparatus for updating product display of a shopping mall related to e-commerce, a storage unit holding information related to display and sales of products of a plurality of shopping malls related to e-commerce, and the A product ranking change index is calculated based on the hourly product rankings of at least some of the plurality of products registered in the first shopping mall held in the storage, and the electronic commerce of the first shopping mall is performed using the product ranking change index and a processor for calculating the product display update time of the web page for updating the product display of the first shopping mall based on the calculated product display update time, wherein the product ranking change indicator is It is determined by the similarity between the arrangement according to the ranking of at least some of the products and the arrangement according to the ranking of at least some of the plurality of products at a second time point, wherein the second time point is subsequent to the first time point It may be the time to do it.

The processor calculates a product display update cycle for updating the product display of the first shopping mall based on the product ranking change index, and compares the calculated product display update cycle with the product display update cycle of the second shopping mall, , it may be recommended to update the display of the first shopping mall based on the comparison result.

A product display update period for updating the product display of the first shopping mall may be set in proportion to a value of the product ranking change index.

The processor may compare the product ranking change index of the first shopping mall with a preset reference value, and generate a product display update notification generated by the product display update time based on the comparison result.

The processor calculates a change rate over time of the product ranking change indicator, but the product display update notification generated by the product display update time is generated in response to a state in which the rate of change over time is higher or lower than a preset reference value can be

The processor, based on the time-based product ranking of products existing at least one of a first time point (t ₁ ) and a second time point (t _{2 ) among the plurality of products, (i) a higher rank is more important than a lower rank} Each product using the weight (λ), (ii) the ranking at the first time point (Rank(t ₁ )), and (iii) the ranking at the second time point (Rank(t ₂ )) The display similarity may be calculated, and a product ranking change index may be calculated using the display similarity for each product.

The product ranking by time may include a sales ranking of at least some of the plurality of products displayed in the shopping mall by time or a display ranking of products in an online product page of the shopping mall.

The first axis is a time axis and the second axis is a graph of the product ranking change index, which visualizes the product ranking change index over time, and a plurality of graphs are divided according to the weight value and can be controlled to be displayed on one screen.

The processor receives the product ranking change index and the product display update cycle calculated by the product ranking change index as inputs, and generates a learning data set using the product sales quantity achieved by the input value as an output value to rank products A deep learning model can be trained to infer the optimal shelf update cycle corresponding to the fluctuation index.

The storage unit holds the value of the product ranking change index and the product display update cycle value corresponding thereto in a table, wherein the processor receives the product display update cycle value corresponding to the calculated value of the product ranking change index in the table By extracting, the product display update cycle can be determined.

The processor is configured to calculate a product display update period of a plurality of shopping malls excluding the first shopping mall using a product ranking change index, and use at least one of statistical values related to a product display update period of the plurality of shopping malls to determine the second shopping mall. 1 It is possible to calculate the product display update cycle of the shopping mall.

According to another aspect of the present invention for achieving the above object, there is provided a product display update system of a shopping mall associated with e-commerce, a user terminal that transmits a request related to product display update of a shopping mall related to e-commerce, and In response to the request related to the product display update, a product ranking change index is calculated based on the hourly product rankings of at least some of the plurality of products registered in the first shopping mall, and the product ranking change index is used to calculate the first and a product display update device that calculates a product display update time of a web page for e-commerce in the shopping mall, and updates the product display of the first shopping mall based on the calculated product display update time, wherein the product ranking change indicator is determined by the similarity between the arrangement according to the ranking of at least some of the products of the plurality of products at the first time point and the arrangement according to the ranking of at least some of the products of the plurality of products at the second time point, The time point may be a time point subsequent to the first time point.

According to the method and apparatus for updating product display in a shopping mall related to e-commerce of the present invention, the product arrangement is arranged at an appropriate time by using the similarity of the ranking arrangement over time from ranking information such as the product sales ranking, display ranking, and advertisement search ranking. It can be easily identified whether advertising cost investment is required to change and change the advertisement search ranking, thereby maximizing product sales efficiency and advertising efficiency.

1 is a conceptual diagram schematically showing a system to which a product ranking change index calculation method according to an embodiment of the present invention is applied;
2A is an exemplary view in which the product ranking is changed less;
2b is an exemplary view in which the product ranking has changed a lot;
3 is a flowchart illustrating a method for calculating a product ranking change index according to an embodiment of the present invention;
4 is a view showing the flow of ranking change of a plurality of products according to the passage of time;
5 is a detailed flowchart specifically showing a display similarity calculation method;
6A is a view showing the display of two different dates by way of example;
Figure 6b is a view showing the calculation of the ranking of the corresponding date for all products,
6c is a diagram for explaining a method of calculating the similarity for each product;
7 is a view for explaining a method of standardizing display similarity according to another embodiment;
8 is a diagram for explaining a change in a similarity value according to a change in a weight;
9 is a flowchart illustrating a method for calculating a product display update time point according to an embodiment of the present invention;
10 is a diagram exemplarily showing a process of deriving an update time according to the method of FIG. 9;
11 is a graph in which standardized similarities of a plurality of products with respect to different weight values are calculated over time;
12 is a block diagram illustrating an apparatus for calculating a product ranking change index according to an embodiment of the present invention.

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail.

However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the present invention.

Terms such as first, second, etc. may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.

When a component is referred to as being “connected” or “connected” to another component, it is understood that the other component may be directly connected or connected to the other component, but other components may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that no other element is present in the middle.

The terms used in the present application are only used to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present application, terms such as “comprise” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. In describing the present invention, in order to facilitate the overall understanding, the same reference numerals are used for the same components in the drawings, and duplicate descriptions of the same components are omitted.

1 is a block diagram illustrating a system to which a product ranking change index calculation method according to an embodiment of the present invention is applied. As shown in FIG. 1 , the system according to an embodiment of the present invention may include shopping malls 110 - 1 to 110 -N, a network 120 , and a product ranking change index calculation device 130 .

Referring to FIG. 1 , shopping malls 110-1 to 110 to N refer to online shopping malls that sell products and/or services through the Internet. That is, they are shopping malls related to e-commerce. The shopping malls 110-1 to 110-N have a specific URL and sell products to visitors to the corresponding address on the Internet. The shopping mall operator operates the shopping mall in such a way that at least a portion of the amount of the sold product is obtained through his/her own account associated with the shopping mall.

In FIG. 1, shopping malls 110-1 to 110-N are implemented as a web page of a specific address (eg, URL) on the Internet, a server handling information related to the shopping mall, or a terminal of an operator of the shopping mall, as described above. can be The terminal may include a computing device such as a smartphone or a PC.

According to an embodiment of the present invention, the plurality of shopping malls 110 - 1 to 110 -N may interwork with the product ranking change index calculating device 130 through the network 120 . The shopping malls 110-1 to 110-N display products, promote event products of the shopping mall, and operate a web site for selling a plurality of products. In this case, in order to promote the sale of a plurality of products registered in the shopping mall, it is preferable to appropriately change the display of the products. A change in display includes a change in the display position of products on the product display page, such as changing the display of products in the order of ABCD to ABDE (eg, changing the order from top to bottom when displayed from top to bottom of a web page); or changing the order in which products are displayed over time. The product display indicates an arrangement of products on a specific web page, an arrangement of products returned in response to a specific search term, and an order formed through arrangement of specific items such as product sales ranking and product search ranking. Assuming such an environment in which the display changes with time, the product ranking change index calculation device 130 calculates the similarity between the two displays at different time points in the environment in which the display of goods changes over time as described above, and ranks Calculate how much of the change is as a quantified indicator. The display similarity (more specifically, the standardized display similarity) indicating how similar two displays are can be used in the same or similar meaning as the product ranking change indicator.

When the operator of the shopping malls 110-1 to 110-N wants to check the ranking change index of a product related to the shopping mall, the operator may request the product ranking change index calculation device 130 using his/her terminal. In this case, when making a request for calculating a ranking change index according to a specific search word, it is preferable to provide search word information together. In addition, when it is desired to calculate the product ranking change index at a certain period, the index calculation period may be set and provided to the product ranking change index calculation device 130 . In another example, the shopping mall operator may directly set a weight value having a great influence on the calculation of the ranking change index and provide it to the product ranking change index calculation device 130 .

On the other hand, the operator of the shopping malls 110-1 to 110-N queries the optimal product display update time based on the product ranking change index or automatically sets the product display update cycle. can be requested to You can also request that the product display be changed automatically at the time or period of the product display update corresponding to the request.

The network 120 is a communication network connecting the shopping malls 110-1 to 110-N and the product ranking change index calculating device 130, and may include a wired network and/or a wireless network (Internet).

The product ranking change index calculation device 130 receives a ranking change index calculation request or a product display update-related request from at least one of the shopping malls 110-1 to 110-N through the network 120, and adds the information to the received information. Based on the calculation of the product ranking change index, the product display can be updated based on this. The product ranking change index calculating device 130 may calculate the product ranking change index according to a value set by a manager who manages the shopping mall platform or a command of the manager without a request from the shopping mall operator. In the present specification, the product ranking change index calculation device 130 may be expressed as an index calculation device, a server, or a device. In another example, since the actual product display can be updated using the calculated index, it may be referred to as a product display update device or a product display update time (or period) calculation device.

In one example, the product ranking change index calculation device 130 may be implemented as a computing device. The computer device referred to herein may be implemented as a server-class computer terminal. The computer device may include an input device, a display device, a networking device, a hard disk, a memory for storing a program, and a processor for executing the program stored in a typical computer terminal. However, it is not necessarily implemented as a server-grade computer terminal.

The product ranking change index calculation device 130 may be a server that distributes and manages a platform for opening and operating a shopping mall, or a server interworking with the server. Therefore, the product ranking change index calculation device 130 not only handles information related to one shopping mall 110-1, but also handles information related to a plurality of shopping malls 110-1 to 110-N, and big data Based on this, various information can be analyzed and learned. In addition, in order to calculate a product ranking change index in response to a request of a user (a shopping mall operator or a shopping mall platform manager), optimal information among information of a plurality of shopping malls 110-1 to 110-N is used to change the product ranking. It is used for index calculation.

The product ranking change index calculation device 130 generates a product list of a specific shopping mall at two different time points. If possible, the ranking of the corresponding product is calculated for all products handled by the corresponding shopping mall. Then, the similarity of the products with respect to the products existing at at least one of the two time points is calculated using the weight, the first time point rank, and the second time point rank. Then, by standardizing the sum of the calculated similarities, the similarity of product display at two time points is calculated. The display similarity calculated in this way functions as an index indicating how much product ranking changes. The higher the display similarity, the less the ranking change, and the lower the display similarity, the more severe the ranking change.

According to another embodiment of the present invention, the terminal associated with the shopping malls 110-1 to 110-N downloads the application provided by the product ranking change index calculation device 130 and calculates the index and updates the product display through the application. can be done That is, the user terminal associated with the shopping malls 110-1 to 110-N calculates the ranking change index and updates the product display by using the weight value, the index calculation period, and the factors related to the index calculation that it considers optimal. can be performed in the terminal. When the processor (not shown) of the terminal executes the downloaded application without a separate server device, it is possible to calculate the ranking change index and update the product display accordingly.

In another example, the method of calculating the product ranking change index of the present invention is not necessarily limited to being applied only to the calculation of the ranking change index for products handled in the shopping mall. For example, the method of calculating the ranking change index of the present invention in a situation where the similarity between two lists expressed through the ranking or order is compared with a lag such as the order (or ranking) of products returned according to a specific search term in an online advertisement or a portal site. It will be apparent to those skilled in the art that the present invention can be used. It does not necessarily have to be a ranking for a product. Other objects such as a specific word, company name, person name, and place name may be the target of calculating the ranking change index. For example, when a place name such as Jeju-do in the first place, Gangwon-do in the second place, and Busan in the third place is searched for in response to the search for travel destination ranking, if the order of the place names returned for the travel destination search word is changed according to time, the order of place names according to time In order to derive the degree of change, the method for calculating a ranking change index according to an embodiment of the present invention may be used.

FIG. 2A is an exemplary diagram in which the product ranking is changed slightly, and FIG. 2B is an exemplary diagram in which the product ranking is relatively changed.

Referring to FIGS. 2A and 2B , the situation of FIG. 2A shows that the display order of products is not significantly changed, while the situation of FIG. 2B shows that the product display order is significantly changed. In the case of FIG. 2A, only the first-ranked products are newly entered, and from the second, the order of the previous products is followed. In the product ranking change index calculation method according to an embodiment of the present invention, a change in a higher rank and a change in a lower rank may be treated differently. This is achieved by weighting. A high weight indicates that a change in a higher rank is considered more important than a change in a lower rank, and a low weight indicates that a change in the upper rank or a change in the lower rank is considered similarly. This is due to the characteristic that people feel very sensitive to changes in rankings of 1st and 2nd places, but not so sensitively to changes in rankings of 7th and 8th places. In the case of Figure 2b, three new products, such as Z, W, and V, enter the ranks of 1st to 3rd, and since the rankings from 4th place do not follow the existing order, it can be seen that the ranking has changed very much, , the device allows the user to intuitively understand how large the liquidity of the display is over time by quantifying the severe degree of such ranking fluctuation as a single value considering the weight.

It is desirable that the index expressing such display fluidity satisfies the following requirements.

- It is possible to calculate the similarity of two display rankings through one indicator.

- In the case of the same display, the similarity is 1, and if there is no common product between the two displays, the similarity is 0.

- When there is a change in the upper rank, the similarity decreases significantly compared to the same type of change in the lower rank.

- The indicator is symmetric (the same value is obtained even if the role of before/after is switched).

The detailed process of the method of calculating the product ranking change index for satisfying the above requirements will be described in more detail with reference to FIG. 3 .

3 is a flowchart illustrating a method for calculating a product ranking change index according to an embodiment of the present invention.

Referring to FIG. 3 , the device obtains product ranking information by time ( S310 ). The product ranking information includes an arrangement of products and/or an order of products. That is, not only information related to ranking according to the sales quantity, but also an arrangement of products in a specific shopping mall at a specific point in time or an arrangement of products returned in response to a specific search term may be included. Such information may be stored by time in a database interworking with the device as log data.

Regarding product ranking information,

The device receives a weight value and a calculation period (S320). As described above, the weight has a value of 0 to 1 as a numerical value indicating that a higher rank is considered more important than a lower rank. The weight may mean a degree to which the similarity decreases when the ranking is lowered one by one. The calculation cycle can be arbitrarily set by the user at regular time intervals such as minutes, hours, 6 hours, 12 hours, daily (24 hours), weekly, and monthly. However, when the period for updating the product display is set, it is preferable to set the period to be shorter than the display update period. Then, when it is determined that the display is not appropriate, it can be recommended to update the display of the product even before the next update cycle arrives.

When the weight value and the calculation period (or calculation request) are input, the device uses the product ranking information at the first time point and the product ranking information at the second time point (or the current time point), and the weight value according to the corresponding cycle, the product ranking change index is calculated (S330). The calculation process of the product ranking change index will be described in more detail with reference to FIG. 5 .

4 is a diagram illustrating a flow of ranking change of a plurality of products according to the passage of time.

Referring to FIG. 4 , when “skirt” is searched for in a shopping mall between July 2, 2018 and July 11, 2018, the display order in which products are displayed can be checked. At this time, in the period 410 between July 3 and 5, the product display order is actively changing, while between July 9 and 11 420, the change in the display order is not active. can be checked The device secures the ranking information of such a product and quantifies it as a single number. As shown in FIG. 4 , it is appropriate to secure the secured ranking information as product ranking (sales ranking, display order, search order, etc.) information according to the passage of time.

5 is a detailed flowchart specifically illustrating a method of calculating the display similarity.

Referring to FIG. 5 , the device generates a list of products of two time points (eg, dates) that are the calculation targets of the ranking change index ( S510 ). The product list includes ranking information of products that exist even at at least one of the two time points. 6A is a view showing the display of two different dates by way of example. Referring to FIG. 6A , when compared to Display 1, in Display 2, a new product Z enters the first display order, and thereafter, A to G are arranged side by side in the same manner as in Display 1.

After securing the product list at two different time points, the device calculates the rank of the corresponding product for all possible products ( S520 ). 6B is a view showing the calculation of the ranking of the corresponding date for all products. Referring to FIG. 6B , a list is generated by calculating the number of ranks of the corresponding product on the corresponding day. That is, the rank of time 1 is arranged in the first column of the list, and the rank of time 2 is arranged in the second column of the list.

Next, the device calculates the product similarity (S530). The product similarity is calculated through the square of the weights (rank of the first time + the rank of the second time). 6C is a diagram for explaining a method of calculating a similarity for each product. Referring to FIG. 6C , for example, in the case of product A, the 1st rank is 1, whereas the 2nd rank is 2, so the product similarity is weight^(1+2) = the third power of the weight, at this time, Assuming the weight is 0.7, (0.7) ³ becomes 0.343. In the same way, the display similarity for each product is calculated by referring to the weight, the first time point, and the second time point ranking for all products.

After calculating the display similarity for each product, the device calculates the sum of the similarities by summing the calculated display similarities for each product ( S540 ). If the display similarity values by product of all products from product A to product Z are added, the total is 0.668.

After calculating the first sum, the device calculates the similarity by product assuming that the rankings of the first time point and the second time point for the same product group are the same, and calculates the second sum by summing the calculated similarities for each product (S550) . According to the embodiment of FIG. 6C , the sum of the similarities in the same display is 0.958.

Based on the two sums, the device divides the display similarity of the current actual display calculated in step S540 by the sum of the similarity similarities calculated in step S550 to the same display assumption and normalizes it to a value between 0 and 1 (S560). The standardized rank sequence similarity is 0.668/0.958 = 70%. This is determined as the final display similarity value (S570).

7 is a view for explaining a method of standardizing display similarity according to another embodiment.

Referring to FIG. 7, when the weight is 0.93 and the change of the first and second days exists as in the embodiment of FIG. 7, the device first generates a daily ranking list of products A to I, and calculates the sum of the two daily rankings. Calculate. Then, the similarity for each product is calculated using this as an index of the weight. Next, the degree of similarity for each product is summed. As in step S540 of FIG. 5 , the column expressed as "changed" in FIG. 7 is a calculation of the similarity for each product in which the first-day and second-day rankings are changed, and the column expressed "if still go" As in step S550 of FIG. 5, assuming that the first-day ranking and the second-day ranking are the same, the degree of similarity for each product is calculated. In the above case, the sum of similarities is calculated as 3.911 and 4.397, respectively, and it can be seen that when the division operation is performed with the two sums, the similarity is 0.889, that is, about 89%.

8 is a diagram for explaining a change in a similarity value according to a change in a weight.

Referring to FIG. 8 , when considering the situation where a new product enters the 1st rank, the rank change of the 1st and 2nd places, the rank change of the 2nd and 3rd places, and the rank change of the 9th and 10th places, the similarity is the highest. What changes a lot is that a new product enters the 1st place, which is also the highest ranking. Next, in turn, the ranking change situation of the 1st and 2nd places, the ranking change situation of the 2nd and 3rd places, and the ranking change situation of the 9th and 10th places significantly change the value of the display similarity. Here, many changes may mean a decrease in the similarity value. Similarly, considering the weight of 0.6, it can be seen that the similarity level of 0.795 for a new product entering the 1st place drops to 0.795, while other situations do not change significantly with 1.000, 0.991, and 0.985. Also, it can be seen that the similarity value varies according to the difference in the weight values even for the same situation.

To summarize the weights, when a new product enters the N-th in the middle of the product display order, and the lower ranks are pushed down by one, the similarity of the display becomes N squared of the weight. And, when the weight is close to 1, more product changes contribute to the similarity, and when the weight is close to 0, only the change in the upper rank contributes to the weight change.

In a situation where the ranking is shifted, the final similarity value has the following characteristics. When the display is shifted (eg, from A-B-C-D to Z-A-B-C), the similarity is calculated as follows.

- In case of shift from 1st priority, similarity = weight

- In case of shift from 2nd rank, similarity = 2 power of weight

- In case of shift from 3rd rank, similarity = 3 power of weight

It is assumed that there are many products, and if there are not many, some errors may occur. Accordingly, the device can easily calculate the display similarity value by using the relationship between the similarity and the weight in the shifted situation as described above.

In conclusion, the display similarity can be expressed by the following equation if the ranking of product P _{i at time (t 1} ) and time (t ₂ ) is called Rank(P _i , T ₁ ), Rank(P _i , t _{2 )} can

Here, λ may mean a weight.

In this case, when the number of products is very large (eg, 20 (N>20) or more), the display similarity may be calculated as follows.

9 is a flowchart illustrating a method for calculating a product display update time point according to an embodiment of the present invention.

Referring to FIG. 9 , the device may calculate a product ranking change index using the method of FIG. 5 ( S910 ). Then, it is possible to derive a time when the display of the product should be updated using the product ranking change index (S920). This may be accomplished through a process such as comparison with a threshold value or calculating a differential value of a product ranking change index. When the update time is calculated, the device updates (changes) the display of the product (S930).

10 is a diagram exemplarily illustrating a process of deriving an update time point according to the method of FIG. 9 .

Referring to FIG. 10 , the device selects a date to be calculated as a product ranking index. In the embodiment of FIG. 10 , it is assumed that “day 1” is used. The calculation target date may be a date corresponding to a basic display update cycle when the display of the shopping mall is periodically updated. For example, assuming that the display is updated in units of 10 days, the 1st, 11th, and 21st days may be the days to be compared for the display similarity with the ranking of the following days. In this case, the product rankings of the 1st and 11th days may indicate the order of products displayed on a specific web page. The device updates the display according to the sales ranking on the first day, and thereafter continuously acquires the daily sales ranking information to calculate the display similarity with the first day arrangement. Then, the device may set a threshold value for product display update, compare the threshold value with the display similarity by date, and control to change the product display once again when the display similarity falls below the threshold value. In the embodiment of FIG. 10 , assuming that the threshold value is 70%, since the display similarity is higher than 70% on the 2nd, 3rd, and 4th days, a warning or notification about the update is not generated. However, in the case of the 5th day, since the display similarity falls below 70%, it may generate a notification that an update is required at this time. Then, on the 6th day, the display similarity may be calculated based on the display ranking of the 5th day, which is the time of previous display update. In this way, the device may control so that the first time point is automatically set as the time point at which the display update occurs.

In another example, the device may set the date (or time point) of the immediately previous calculation period to be the first time point based on the index calculation period. In the embodiment of FIG. 10 , assuming that the index calculation cycle is 1 day, the second day calculates the similarity with the display of the first day, the third day calculates the display and the similarity of the second day, and the fourth day is the third day Display and similarity can be calculated. This can be used when updating the product display on a daily basis, and when you want to judge the appropriateness of the update based on the similarity with the previous display. For example, if the product display is to be changed while maintaining a degree of similarity with the existing display, the display may be controlled to determine that the product display is not appropriate in response to the similarity falling below a threshold value after the display change. After determining that it is not appropriate, a warning or notification signal is generated so that the shopping mall operator can understand it. The opposite is also possible. If the shopping mall user wants to exclude the display of the same or a certain form, the shopping mall user changes the display every day, but when the similarity with the previous date is more than a certain value, it is judged that the product display has not been appropriately changed and a warning or notification signal is generated can do it This can be appropriately adjusted by changing the user settings according to the user's needs.

In another embodiment, the device may calculate the product display update cycle based on the product ranking change index of a plurality of shopping malls. Then, the update period of the first shopping mall may be compared with the update period of other shopping malls except for the first shopping mall, and display update of the first shopping mall may be recommended based on the comparison result. In one example, the update period may be determined as a value that is inversely proportional to the value of the product ranking change index. If the product ranking change index is a low value (eg, 0.5), it is preferable that the update cycle is set to a low value so that the display is updated quickly. Conversely, if the product ranking change index is a value close to 1 (0.95), it is preferable that the update cycle is set to a high value so that it is updated relatively slowly. In this case, the update cycle of a specific shopping mall may be set to correspond to the update cycle of other shopping malls. The update period of other shopping malls may be set to appropriate statistical values, such as average, minimum, and maximum values of update periods of all shopping malls. In some cases, the device learns the relationship between the product ranking change index and the update period to infer the optimal update period corresponding to the specific product ranking change index, and updates corresponding thereto based on the product ranking change index of the first shopping mall cycle can be determined. In this case, the learning data may use the product ranking change index and the update period as input values and the product sales quantity achieved by the display update as an output value. The device may learn the relationship between the product ranking change index and the update period by finding an optimal hyper parameter based on the input value and the output value. In this case, a deep learning model, a machine learning model, or the like may be used for learning. It is desirable to consider the weight value as well.

In another example, an update period corresponding to the value of the product ranking change index may be set and held as a table. Then, the device may extract an update period corresponding to a specific index value from the table and control the display to be updated according to the extracted period.

Meanwhile, in response to a notification signal for an update cycle or update time, the display of products may be automatically arranged based on the sales ranking, the display ranking, or the search ranking at the time. That is, when a notification signal is generated, the product may be optimally displayed according to the sales ranking, etc. automatically without a separate input or request by the user.

11 is a graph in which standardized similarities of a plurality of products with respect to different weight values are calculated over time.

Referring to FIG. 11 , the device may continuously calculate and store the standardized display similarity over time. In this case, it is preferable to store the weight and the display similarity calculated accordingly. This is because it can be utilized in consideration of the relationship with the weight. The device may visualize the stored display similarity in the form of a graph. In this case, the horizontal axis may be the time axis and the vertical axis may be the display similarity axis, and different graphs may be visualized together according to weights. The device displays a display similarity graph based on a plurality of weights together so that a user can intuitively check a noticeable change in display similarity. Changes in display similarity are displayed differently according to weights even in the same time period, so that the user can understand them while viewing them on one screen.

Meanwhile, as described above, the stored display similarity is stored together with various factors such as weights and used for learning. Through machine learning or deep learning-based learning, it is possible to infer not only the update cycle and relationship, but also the relationship between a specific product or specific search term and the corresponding ranking change index, and the relationship between weights and the ranking change index. An update period, a weight value, and the like can be extracted.

In another example, as shown in FIG. 11 , the device may calculate the rate of change of the index through differential calculation based on the change over time of the index of change in product ranking. Then, it is possible to control to generate a notification signal for display update based on the change rate. For example, when the rate of change of the indicator falls above or below the threshold at a specific point in time, the device determines this as an abnormality and generates a notification signal to request a display change. The user may update the product display in response to the notification signal. In this case, it is preferable to set the threshold value serving as the upper limit of the change rate and the threshold value serving as the lower limit to different values.

12 is a block diagram illustrating an apparatus for calculating a product ranking change index according to an embodiment of the present invention. 12 , the device may include a communication unit 1210 , a processor 1220 , a memory 1230 , and an input/output module 1240 .

Referring to FIG. 12 , the communication unit 1210 is a component for exchanging information with the user terminal through the network 120 of FIG. 1 . The communication unit 1210 includes a communication device associated with a wired or wireless network. The communication unit 1210 may include an antenna.

The processor 1220 is a component that obtains the information related to the index calculation request and the product ranking change index calculation received from the user terminal through the communication unit 1210, calculates the index according to the programming, and determines the update period or time point. . The processor 1220 is connected to the memory 1230 through a signal line, and executes a program stored in the memory 1230 . As described above, the processor 1220 generates a list of products according to time, and calculates a product ranking for all possible products. Then, the similarity value for each product is calculated using the weight of the products existing on at least one of the two dates as the base and the sum of the rankings of the two time points as the index. Then, the processor 1220 sums up and standardizes the calculated similarity values for each product to derive a final display similarity value, and uses this as an index indicating a change in the ranking of the products. In use, the processor 1220 compares the ranking change index with a threshold value, matches a specific value stored in a table, or determines the optimal product display update time or period using inference through proportional/inverse relationship or learning derive

The memory 1230 is a storage device that stores instructions related to programs to be executed by the processor 1220 and stores various data required by the processor 1220 . The memory 1230 may store various types of information related to product display and product sales in the shopping mall. The memory 1230 may be implemented as a local memory inside the device or as an external large-capacity database.

The input/output module 1240 includes an information input means such as a keyboard and a mouse and an information output means such as a monitor, a TV, and a touch screen. The input/output module 1240 is used to change various set values (weight, threshold, etc.) related to index calculation and derivation of an update period, and to display related contents.

The system or apparatus described above may be implemented as a hardware component, a software component, and/or a combination of the hardware component and the software component. For example, the systems, devices and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA). ), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions, may be implemented using one or more general purpose or special purpose computers. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, although one processing device is sometimes described as being used, one of ordinary skill in the art will recognize that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that can include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.

The software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or device, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave. The software may be distributed over networked computer systems, and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiments may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software. Examples of the computer readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and carry out program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with reference to the limited embodiments and drawings, various modifications and variations are possible for those skilled in the art from the above description. For example, the described techniques are performed in a different order than the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (23)

Calculating a product ranking change index based on the product ranking by time of at least some of the plurality of products registered in the first shopping mall related to the e-commerce, wherein the product ranking change index is one of the plurality of products at a first time. It is determined by a similarity between an arrangement according to the ranking of at least some of the products and an arrangement according to the ranking of at least some of the plurality of products at a second time point, wherein the second time point is a time point subsequent to the first time point Lim;
calculating a product display update time of a web page for electronic commerce in the first shopping mall by using the product ranking change index; and
Comprising the step of updating the product display of the first shopping mall based on the calculated product display update time,
The product display update period for updating the product display of the first shopping mall is set in proportion to the value of the product ranking change index,
The product ranking change index and the product display update cycle calculated by the product ranking change index are input, and a learning data set is generated using the product sales quantity achieved by the input value as an output value to correspond to the product ranking change index A product display update method of a shopping mall related to e-commerce, which trains a deep learning model to infer the optimal display update cycle. The method of claim 1,
calculating a product display update cycle for updating the product display of the first shopping mall based on the product ranking change index;
comparing the calculated product display update period with the product display update period of a second shopping mall; and
The method further comprising the step of recommending to update the display of the first shopping mall based on the comparison result. delete The method of claim 1,
comparing the product ranking change index of the first shopping mall with a preset reference value; and
and generating a product display update notification generated by the product display update time based on a comparison result. The method of claim 1,
Further comprising the step of calculating the change rate over time of the product ranking change index,
The product display update notification generated by the product display update time is generated in response to a state in which the rate of change over time is higher or lower than a preset reference value. The method of claim 1, wherein the calculating of the product ranking change index comprises:
Based on the product ranking by time of the product existing at at least one of the first time point (t ₁ ) and the second time point (t ₂ ) among the plurality of products, (i) indicating that an upper rank is considered more important than a lower rank The display similarity for each product is calculated using the numerical weight (λ), (ii) the rank at the first time point (Rank(t ₁ )), and (iii) the rank at the second time point (Rank(t _{2 )).} to do; and
and calculating a product ranking change index by using the display similarity for each product. 7. The method of claim 6,
The product display update method of the shopping mall associated with e-commerce, wherein the product ranking by time includes a sales ranking by time of at least some of the plurality of products displayed in the shopping mall or a display ranking of products in an online product page of the shopping mall. 7. The method of claim 6,
The first axis is a time axis, and the second axis further comprises the step of visualizing the product ranking change index over time as a graph of the product ranking change index,
A method for updating a product display of a shopping mall related to e-commerce, in which a plurality of graphs are classified according to the weight value and controlled to be displayed on one screen. delete The method of claim 1,
Keep the value of the product ranking change index and the product display update cycle value corresponding thereto as a table,
and extracting a product display update cycle value corresponding to the calculated product ranking change index value from the table to determine a product display update cycle. The method of claim 1,
calculating a product display update period of a plurality of shopping malls excluding the first shopping mall using a product ranking change index; and
and calculating a product display update period of the first shopping mall by using at least one of statistical values related to the product display update period of the plurality of shopping malls. a storage unit that holds information related to display and sales of products in a plurality of shopping malls related to e-commerce; and
A product ranking change index is calculated based on the hourly product rankings of at least some of the plurality of products registered in the first shopping mall held in the storage unit, and the e-commerce transaction of the first shopping mall is performed using the product ranking change index Comprising a processor for calculating the product display update time of the web page for updating the product display of the first shopping mall based on the calculated product display update time,
The product ranking change index is determined by the similarity between the arrangement according to the ranking of at least some of the products of the plurality of products at the first time and the arrangement according to the ranking of at least some of the products of the plurality of products at the second time. and the second time point is a time point subsequent to the first time point,
The product display update period for updating the product display of the first shopping mall is set in proportion to the value of the product ranking change index,
The product ranking change index and the product display update cycle calculated by the product ranking change index are input, and a learning data set is generated using the product sales quantity achieved by the input value as an output value to correspond to the product ranking change index A product display update device of a shopping mall related to e-commerce, which trains a deep learning model to infer the optimal display update cycle. The method of claim 12, wherein the processor comprises:
A product display update cycle for updating the product display of the first shopping mall is calculated based on the product ranking change index, and the calculated product display update cycle is compared with the product display update cycle of the second shopping mall, and the comparison result is obtained. A product display update device of a shopping mall associated with e-commerce, which recommends updating the display of the first shopping mall based on the information. delete The method of claim 12, wherein the processor comprises:
A product display update device of a shopping mall associated with e-commerce, which compares the product ranking change index of the first shopping mall with a preset reference value and generates a product display update notification generated by the product display update time based on the comparison result. 13. The method of claim 12,
The processor, but calculates the rate of change over time of the product ranking change index,
The product display update notification generated by the product display update time is generated in response to a state in which the rate of change over time is higher or lower than a preset reference value. The method of claim 12, wherein the processor comprises:
Based on the product ranking by time of the product existing at at least one of the first time point (t ₁ ) and the second time point (t ₂ ) among the plurality of products, (i) indicating that an upper rank is considered more important than a lower rank The display similarity for each product is calculated using the numerical weight (λ), (ii) the rank at the first time point (Rank(t ₁ )), and (iii) the rank at the second time point (Rank(t _{2 )).} and calculating a product ranking change index by using the display similarity for each product. 18. The method of claim 17,
The product display update apparatus of the shopping mall associated with e-commerce, wherein the product ranking by time includes a sales ranking by time of at least some of the plurality of products displayed in the shopping mall or a display ranking of products in an online product page of the shopping mall. 19. The method of claim 18,
The first axis is a time axis and the second axis is a graph of the product ranking change index, which visualizes the product ranking change index over time, and controls to be displayed on one screen by dividing a plurality of graphs according to the weight value, related to e-commerce Product display update device in shopping mall. delete 13. The method of claim 12,
The storage unit holds the value of the product ranking change index and the product display update cycle value corresponding thereto as a table,
and the processor extracts a product display update cycle value corresponding to the calculated product ranking change index value from the table to determine a product display update cycle. The method of claim 12, wherein the processor comprises:
The product display update period of the plurality of shopping malls excluding the first shopping mall is calculated using a product ranking change index, and the product of the first shopping mall is calculated using at least one of statistical values related to the product display update period of the plurality of shopping malls. A product display update device of a shopping mall associated with e-commerce, which calculates a display update cycle. a user terminal for transmitting a request related to product display update of a shopping mall related to e-commerce; and
In response to the request related to the product display update from the user terminal, based on the product ranking by time of at least some of the plurality of products registered in the first shopping mall, a product ranking change index is calculated, and the product ranking change index is calculated and a product display update device that calculates a product display update time of a web page for an e-commerce transaction of the first shopping mall by using the product, and updates the product display of the first shopping mall based on the calculated product display update time,
The product ranking change index is determined by the similarity between the arrangement according to the ranking of at least some of the products of the plurality of products at the first time and the arrangement according to the ranking of at least some of the products of the plurality of products at the second time. and the second time point is a time point subsequent to the first time point,
The product display update period for updating the product display of the first shopping mall is set in proportion to the value of the product ranking change index,
The product ranking change index and the product display update cycle calculated by the product ranking change index are input, and a learning data set is generated using the product sales quantity achieved by the input value as an output value to correspond to the product ranking change index A product display update system for shopping malls related to e-commerce, which trains a deep learning model to infer the optimal display update cycle.

Images