KR101083002B1 - Method and Server apparatus for calculating User Conversion Rate - Google Patents

Abstract

According to an exemplary embodiment of the present invention, a method of calculating a purchase conversion probability includes receiving a predetermined search word from a user through the Internet, extracting advertisement items related to the search word; And calculating a user conversion rate (UCR) based on the inverse of the number of clicked line items, for the line item clicked by the user among the advertisement related items for the search word.

Description

Calculation method of purchase conversion probability and server apparatus therefor {Method and Server apparatus for calculating User Conversion Rate}

The present invention relates to a method of calculating the probability of purchase conversion.

Today, with the rapid development of communication networks such as the Internet, information retrieval through communication networks is becoming very common. In other words, by using the Internet, Internet users can easily obtain the necessary information anytime, anywhere, and such information retrieval brings a lot of changes throughout our lives.

Accordingly, various models have been developed for billing methods for providing advertisements between a service provider providing a search service to an Internet user and an advertiser who obtains an advertisement effect by providing information about his website to the Internet user through search results. Is being developed.

As an example of a billing method for providing an advertisement, a method of pre-selecting a location where an advertiser's information is to be displayed on a search result providing screen and assuming a predetermined sales amount for each selected location is sold to the advertiser. This allows advertisers who want to display their information in a particular location to give their first-time advertiser the right to display their information for locations that aren't currently showing their information. There is a problem of waiting until the end of the contract period for the information of the advertiser previously displayed in the location.

That is, the above-described method is a billing method in which the advertiser preempts a specific position where the advertiser's information is to be displayed, and includes a disadvantage that other advertisers cannot provide information during the contract period when the information display position is occupied by the advertiser. In addition, even when waiting until the end of the contract period, there is a problem that the information provided by the waiting advertisers can not be guaranteed because it is a sale method on a first-come, first-served basis. In addition, since the sale amount for the display position is determined by the service provider, there is a problem that the advertiser's position is not considered at all on the sale amount.

In order to solve the above problem of the charging method has been proposed a charging method by PPC (Pay Per Click) method. The PPC method calculates the advertising cost by calculating the number of connection connections (valid connections) that a user connects to an advertiser by clicking on information of a specific advertiser included in a search result. According to the size order of the PPC amount (the charge per click), the advertisements of each advertiser are sequentially arranged within a specific area and provided to the user.

In this PPC billing method, advertisers who provide the highest PPC amount are placed at the top within a specific area. Therefore, advertisers use competitively high PPC amount in order to display their information in a relatively good position. As a result, the advertising costs borne by each advertiser must also continuously increase, resulting in an increase in the number of advertisers who give up advertising.

In addition, as described above, it is determined only by the competition of advertisers such as the placement order or location of the advertisement in the specific area of the screen displaying the search result or the highest PPC amount, such as the user requesting the search. There is also a problem that the advertisement can be placed at the top of the search irrelevant to the requested search, in which case the user can be recognized that the search accuracy is low.

Therefore, recently, researches on a method of providing an advertisement service by analyzing a user's behavioral pattern after providing the line item have been conducted.

The present invention is to solve the above-described problems, it is a technical problem to provide a method for calculating the probability of purchase conversion by analyzing the behavior pattern of the user after providing the line item.

According to an aspect of the present invention, there is provided a method of calculating a probability of purchase conversion according to an aspect of the present invention, the method comprising: receiving a predetermined search word from a user through the Internet, extracting line items associated with the search word, and among advertisement-related items for the search word. Calculating a user conversion rate (UCR) for the line item clicked by the user using the inverse of the number of clicked line items.

According to another aspect of the present invention, a server device for calculating a purchase conversion probability includes: an input unit for receiving a predetermined search word from a user through the Internet, a line item extraction unit for extracting one or more line items associated with the search word, And a UCR determination unit for calculating a user conversion rate (UCR) using an inverse number of the number of clicked item items for the item items clicked by the user among the advertisement related items for the search word.

According to the present invention, the probability of purchase conversion may be calculated by analyzing a user's behavior pattern (eg, a click pattern) after providing the line item.

In addition, according to an embodiment of the present invention, the purchase conversion probability calculated through the user's click pattern analysis may be utilized for various purposes such as determining an advertisement location and determining a charge.

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

One embodiment of the present invention can provide a variety of advertising services through the analysis of the behavior pattern of the user after providing the line item. The advertisement service may include various services such as ranking of an associated advertisement with respect to a specific search word and determining a bill of an advertiser providing the associated advertisement.

In addition, the analysis of the behavioral pattern of the user after providing the line item may include an analysis of the probability of purchase conversion obtained through the analysis of which advertisement the user clicks on, and the analysis of the residence time of the user clicking on the advertisement site. . The purchase conversion probability may be used as one indicator for measuring advertisement effect. In other words, the probability of purchase conversion may be used as an index for probable or indirectly measuring or estimating the purchase conversion rate for the advertiser site. The probability of purchase conversion is an index that considers the number of customers who generate clicks and the number of sites clicked by the customer, not just the sum of clicks. The probability of purchase conversion is that the more clicks on a specific site, the higher the probability that the purchase rate on the corresponding site is higher, and the higher the number of clicked sites, the lower the purchase rate on a specific site is reflected.

1 is a flowchart illustrating a method of calculating a purchase conversion probability according to an embodiment of the present invention.

Referring to FIG. 1, first, a predetermined search word is input from a user through the Internet in step S100. Then, the advertisement items associated with the search word input in step S110 are extracted, and the extracted advertisement item is provided to the user in step S120. Various actions of the user may be performed on the provided line items. For example, the user may click on a specific line item among the provided line items, or input another search word without clicking the line item because the user does not like the provided line item.

In step S130, the user's behavior pattern is analyzed, and in step S140, the purchase conversion probability is calculated through the user's behavior pattern analysis. The purchase conversion probability calculation will be described in more detail below.

2 is an exemplary structural diagram of an advertisement service providing server that provides an advertisement service using a purchase conversion probability obtained in accordance with a purchase conversion probability calculation method according to an embodiment of the present invention.

2, the advertisement service providing server 10 according to the present embodiment may include an input unit 140, a line item extracting unit 110, a UCR determining unit 120, a UCR using unit 130, The result display unit 150 and the database 160 is included.

The input unit 140 according to the present embodiment receives a predetermined search word from a user through the Internet. When a user wants to search for specific information through a communication network, a user may input a search word corresponding to a keyword of information to be obtained and display a advertisement item extracted by the advertisement service providing server 10. Here, the terminal includes a computer, a terminal having a predetermined memory and arithmetic capability, such as a personal digital assistant (PDA), a mobile phone, a smart phone, an electronic dictionary, an MP3 player, and the like.

The advertisement item extractor 110 according to the present embodiment extracts advertisement items associated with a search word input through the input unit 140 using a search engine. The search engine may mean a program or a search server for searching for information corresponding to a predetermined search word input by a user through a search means. For example, a web server at a particular site may be an example.

In response to a user's search request, the search engine may provide brief information about the advertiser (eg, a specific advertisement item) that may provide information requested by the user. The brief information about the advertiser may also include link information for moving to a site operated by the advertiser. The search engine may provide the user with information desired by the user through various search methods such as a search method by inputting a search word and a category search method by information subject. Meanwhile, the search engine may be configured separately from the advertisement service providing server 10 or may be integrated into one.

The UCR determination unit 120 according to the present embodiment calculates a user conversion rate (UCR) for each of one or more advertisement items associated with a predetermined search word. The probability of purchase conversion (UCR) according to the present embodiment may be generated for the line item clicked by the user as a probability that the user will make a purchase conversion in a specific site. For example, the total number of advertisement items clicked by the user among the extracted advertisement items for the search word may be calculated and calculated using the inverse of the value.

search word Click order One 2 3 Line item1 click Line item 2 click Line item 3 Line item 4 click Line item 5

Table 1 shows whether the user clicks on the item 1 to item 5 extracted for the search word. For example, in this case, the probability of purchase conversion (UCR) for the clicked line items according to the present embodiment is 1 /, which is the inverse of 3, that is, the total number of line items clicked by the user among the line items extracted for the search term. Can be three.

The UCR user unit 130 according to the present embodiment calculates a specific factor using the purchase conversion probability. For example, the purchase conversion probability may be used to calculate a Customer Reaction Index (CRI) or to calculate an advertiser's billing amount. The customer response index (CRI) according to this embodiment is indexed to take into account the user's, or customer's, response to ad impressions on the Internet. It may be determined in consideration of the probability that a purchase is made and / or the time a customer stays on the corresponding Internet page.

The result display unit 150 according to the present exemplary embodiment provides an advertisement item associated with a search word to a user who has requested a search for the search word through the Internet.

The database 160 according to the present exemplary embodiment may store various items such as an advertisement item related to a predetermined search word and price information about the advertisement item.

3 is an exemplary flowchart of a purchase conversion probability calculation method according to an embodiment of the present invention.

First, in operation S300, a purchase conversion probability (UCR) is calculated for each of one or more advertisement items associated with a predetermined search word. As described above, the probability of purchase conversion (UCR) is a probability that a user converts a purchase on a specific site, and calculates the total number of line items clicked by the user among the line items extracted for the search term. It can be calculated as the inverse of the value.

In addition to the purchase conversion probability (UCR), the average purchase conversion probability (AVG_UCR) may be calculated for the users who visited the corresponding site in step S310. The purchase conversion probability (UCR) may be calculated by further considering the total number of line items provided to the user for the predetermined search word for each user. For example, a case in which User 1 and User 2 visits an Internet page corresponding to a specific line item will be described.

User 1:
search word Click order One 2 3 Line item1 click Line item 2 Line item 3 click Line item 4 click Line item 5

User 2:
search word Click order One 2 3 Line item1 click Line item 2 Line item 3 click

Table 2 shows whether User 1 clicks on line items 1 to 5 extracted for a query, and Table 3 shows User 2 clicks on line items 1 to 3 extracted for a query. have. For example, in this case, the average purchase conversion probability AVG_UCR for the clicked line items according to the present embodiment is as follows. ) / 2), and 0.333 (= 1/3) for line item 4.

In this case, the total conversion number (listing number) provided to the user for the predetermined search word may be further applied to the purchase conversion probability (UCR). The greater the total number of line items, the lower the likelihood that a particular item is selected. Therefore, more weight is given to clicks generated when there are more line items. For example, in the above case, the total number of line items provided to user 1 is 5 and the total number of line items provided to user 2 is 3. In this case, the average purchase conversion probability AVG_UCR for the clicked line items according to the present embodiment is as follows. About 0.396 (= ((1/3) * 5 + (1/2) * 3) / (5 + 3)) for line item 1 and line item 3, and 0.333 (= 1/3 for line item 4) )

On the other hand, the customer rating determined for each user may be further applied to the purchase conversion probability (UCR).

For example, each customer can be graded using the customer's query set, past purchase history, and sales contribution, and reflected in the UCR decision. For example, assume that the customer rating given to user 1 is 0.9 and the customer rating given to user 2 is 0.7. In this case, the average purchase conversion probability AVG_UCR for the clicked line items according to the present embodiment is as follows. About 0.406 (= ((1/3) * 0.9 + (1/2) * 0.7) / (0.9 + 0.7)) for line item 1 and line item 3, and 0.3 (= (1 / 3) * 0.9).

Of course, it may be calculated in consideration of the total number of line items provided to the user for the predetermined search word for each user and the customer grade determined for each user. In this case, the average purchase conversion probability AVG_UCR for the clicked line items according to the present embodiment is as follows. Line item 1 and line item 3 are approximately 0.386 (= ((1/3) * 5 * 0.9 + (1/2) * 3 * 0.7) / (5 * 0.9 + 3 * 0.7)), and line item 4 In this case, 0.3 (= 1/3 * 0.9).

In operation S320, the normalized purchase conversion probability STD_UCR may be calculated using the purchase conversion probability UCR and the average purchase conversion probability AVG_UCR. This is to compare the high and low effects according to the relative purchase conversion probability (UCR) value of each advertisement list for the search term.

On the other hand, in determining the conversion probability (UCR), average conversion probability (AVG_UCR), and normalized purchase conversion probability (STD_UCR), the concept of temporal addition is added so that recent clicks are purchased at a higher rate than past clicks. It can also influence the calculation of conversion probability.

For example, the weight of time may be reflected in the UCR using the following function.

σ: sample period

τ: reporting period

UCR: user conversion rate

The purchase conversion probability (UCR) calculated by the above method may be used in various ways. For example, the customer response index may be calculated using a purchase conversion probability (UCR), and the advertisement position may be determined based on the calculated customer response index (CRI). The customer response index (CRI) determined according to the present embodiment is reflected in determining the exposure ranking on the Internet when providing line items for a predetermined search word, so that the customer response index (CRI) can directly or indirectly reflect the actual customer response to the search result. In addition, this is expected to increase the reliability of the search for the customer, while being given an objective criterion for the ranking of the advertisement exposure, it is expected to have an effect that can prevent the unnecessary competition of the advertiser and the damage of the price increase.

In addition, it may be used to determine various factors related to Internet advertising, such as determining the advertiser's billing using the purchase conversion probability (UCR).

The above-described purchase conversion probability calculation method may be implemented in the form of program instructions that can be executed by various computer means and recorded in a computer-readable recording medium. In this case, the computer-readable recording medium may include program instructions, data files, data structures, and the like, alone or in combination. On the other hand, the program instructions recorded on the recording medium may be those specially designed and configured for the present invention or may be available to those skilled in the art of computer software.

The computer-readable recording medium includes a magnetic recording medium such as a magnetic medium such as a hard disk, a floppy disk and a magnetic tape, an optical medium such as a CD-ROM and a DVD, a magnetic disk such as a floppy disk, A magneto-optical media, and a hardware device specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. The recording medium may be a transmission medium such as an optical or metal wire, a waveguide, or the like including a carrier wave for transmitting a signal specifying a program command, a data structure, or the like.

In addition, program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

On the other hand, those skilled in the art will understand that the present invention described above can be implemented in other specific forms without changing the technical spirit or essential features.

Therefore, it is to be understood that the embodiments described above are exemplary in all respects and not restrictive. The scope of the present invention is shown by the following claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. do.

1 is a flowchart illustrating a method of calculating a purchase conversion probability according to an embodiment of the present invention.

2 is an exemplary structural diagram of an advertisement service providing server that provides an advertisement service using a purchase conversion probability obtained in accordance with a purchase conversion probability calculation method according to an embodiment of the present invention.

3 is an exemplary flowchart of a method for calculating a purchase conversion probability according to an embodiment of the present invention.

<Description of the code for the main configuration of the drawings>

10: ad service providing server 110: line item extraction unit

120: UCR determination unit 130: UCR utilization unit

140: input unit 150: result display unit

160: database

Claims (10)

In the method of calculating the probability of purchase conversion using a server device for calculating the probability of purchase conversion, Receiving, by the server device, a predetermined search word from a user through the Internet; The server device extracting advertisement items associated with the search term; And Calculating, by the server device, a user conversion rate (UCR) using a reciprocal of the number of clicked line items for the line item clicked by the user among the advertisement related items for the search word, The purchase conversion probability calculating step And calculating, by the server device, the purchase conversion probability by further considering the total number of line items provided to the user for the predetermined search word for each user. delete The method of claim 1, wherein the purchasing conversion probability calculating step Calculating, by the server device, an average purchase conversion probability (AVG_UCR) for each line item as an average value of the purchase conversion probability values calculated for a plurality of users who visited the site for each line item of the search term. ; And Calculating, by the server device, a normalized purchase conversion probability (STD_UCR) using the purchase conversion probability (UCR) and the average purchase conversion probability (AVG_UCR) calculated for the plurality of users. How to calculate the probability of purchase conversion. The method of claim 1, wherein the purchasing conversion probability calculating step And calculating the server device by considering the customer grade determined for each user. The method of claim 4, wherein the purchasing conversion probability calculating step The purchase conversion probability calculation method, characterized in that the recent click on the user's click to determine the conversion probability is greater than the user's click occurred in the past. An input unit for receiving a predetermined search word from a user through the Internet; A line item extracting unit extracting one or more line items associated with the search word; And The UCR determination unit calculates a user conversion rate (UCR) using the inverse of the number of the line item clicked on the line item clicked by the user among the ad-related items for the search term. The UCR determination unit And calculating the purchase conversion probability by considering the total number of line items provided to the user for the predetermined search word for each user. delete The method of claim 6, wherein the UCR determination unit Calculate an average purchase conversion probability (AVG_UCR) for each line item as an average value of the purchase conversion probability (UCR) values calculated for a plurality of users who visited the site for each line item for the search term, A server device for calculating a purchase conversion probability, characterized by calculating a normalized purchase conversion probability (STD_UCR) using the purchase conversion probability (UCR) and the average purchase conversion probability (AVG_UCR) calculated for the plurality of users. . The method of claim 6, wherein the UCR determination unit The server device for calculating the probability of purchase conversion, characterized in that for calculating the purchase conversion probability further considering the customer rating determined for each user. The method of claim 9, wherein the UCR determination unit The server apparatus for calculating the conversion probability of the purchase, characterized in that the recent click on the user's click is determined in a greater proportion than the user's click occurred in the past.

Images