JP2013089080A - Advertisement output device, advertisement output method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To increase possibility that a user takes an action in response to advertisements, when displaying a plurality of advertisements.SOLUTION: A Web server 20 determines a prescribed number of advertisements 2 which are displayed on a screen 1 from among a plurality of advertisements. The Web server 20 comprises: a click log storage part 255 which stores a click log showing that a user has clicked the advertisements 2 in the past; an advertisement selection part 213 which selects one with the largest number of unique users who have clicked the advertisement in the past among combinations of a prescribed number of advertisements; and a Web page generation part 212 which generates screen information including the advertisements selected by the advertisement selection part 213.

Description

  The present invention relates to an advertisement output device, an advertisement output method, and a program.

  Conventionally, when a plurality of advertisements are displayed (listed), the advertisements are listed in descending order of advertisement evaluation by CTR (Click Through Rate), CPM (Cost Per Mille), or the like. The selection of the advertisement according to the evaluation is intended to increase the possibility of the user taking action (clicking in the case of a Web page) according to the advertisement.

2007-272814

  For example, if a user clicks on an ad for a cola and clicks on an ad for a fruit-flavored carbonated drink, the user may be cola-flavored or fruit-flavored. Since it is responding to a beverage, the probability that one of the same carbonated drinks will be clicked when two advertisements for the same carbonated drink are displayed is that the one advertisement is displayed when only one carbonated drink advertisement is displayed. It is assumed that it is not twice the probability of being clicked, but rather smaller than twice. In such a case, there is a possibility that the probability that the advertisement is clicked is higher when the advertisement of the carbonated drink is kept in one and the advertisement of another category is put out.

  Patent Document 1 describes that when a plurality of advertisements are displayed while matching the advertisement category and the content category, the advertisement of the same category is not displayed. It is aimed at increasing the degree of coincidence with, and even if a plurality of advertisements of the same category are not displayed, it is not known whether or not the possibility that the user will take action increases.

  The present invention has been made in view of such a background, and when displaying a plurality of advertisements, an advertisement output device and an advertisement output method that can increase the possibility of a user taking action in accordance with the advertisements And to provide a program.

  A main invention of the present invention for solving the above problem is an apparatus for outputting a predetermined number of advertisements to be displayed from among a plurality of advertisements, wherein the user is associated with the advertisements in the past, The action history storage unit that stores action history information indicating that an action corresponding to the advertisement has occurred and the action history storage unit are referred to, and the action has been caused in the past among the predetermined number of advertisement combinations. An advertisement selection unit that selects a user with the largest number of non-overlapping users and an advertisement output unit that outputs the selected advertisement are provided.

  Moreover, in the advertisement output device of the present invention, the advertisement selection unit calculates the number of actions for each of the advertisements from the action history storage unit, which is the number of users corresponding to the advertisement that are not duplicated. The behavior determination unit, an advertisement determination unit that determines a first advertisement to be displayed from among the plurality of advertisements, and the other advertisements that are different from the first advertisement. A correlation action number calculation unit that calculates a correlation action number that is the number of users corresponding to the other advertisements among the users corresponding to the first advertisement from a history storage unit, and the advertisement determination The unit may further determine a second advertisement to be displayed together with the first advertisement from the plurality of advertisements according to a value obtained by subtracting the number of correlated actions from the number of actions.

  In addition, the advertisement output device of the present invention includes a total number information storage unit that stores total number information for determining the total number of users, and the behavior number calculation unit is a value obtained by dividing the behavior number by the total number. An action rate is calculated, the advertisement determining unit determines the first advertisement according to the action rate, and the correlated action number calculating unit calculates the number of correlated actions for each of the other advertisements. The advertisement determination unit calculates a correlation rate, which is a quotient obtained by dividing the value obtained by dividing the value by the behavior rate for the first advertisement by the behavior rate for the other advertisement. The second advertisement may be determined according to a value obtained by subtracting the correlation rate from the rate.

  The advertisement output device of the present invention further includes a bid amount storage unit that stores a bid amount for each advertisement, and the advertisement determination unit is configured to perform the first according to a value obtained by multiplying the action rate by the bid amount. The second advertisement may be determined according to a value obtained by multiplying the value obtained by subtracting the correlation rate from the activity rate and the bid amount.

  The advertisement output device of the present invention includes a category storage unit that stores a category to which the advertisement belongs for each advertisement, and the action history information includes category specifying information for specifying the category of the advertisement. The advertisement selection unit refers to the behavior history storage unit, and selects the combination of the predetermined number of the categories that has the largest number of non-overlapping users who have caused the behavior in the past, The advertisement corresponding to the selected category may be selected.

  Another aspect of the present invention is an apparatus for outputting a predetermined number of advertisements to be displayed from among a plurality of advertisements, wherein the advertisements are displayed to the user in association with the users and the advertisements. An impression storage unit that stores impression information for calculating the number of impressions, and an action history indicating that the user has previously acted according to the advertisement in association with the user and the advertisement The impression corresponding to the non-overlapping user who has caused the action in the past out of the predetermined number of combinations of advertisements with reference to the action history storage unit that stores information, the action history storage unit, and the impression storage unit An advertisement selection unit that selects the one having the largest total number and an advertisement output unit that outputs the selected advertisement are provided.

  According to another aspect of the present invention, there is provided a method for determining a plurality of advertisements, wherein a computer correlates with a user and the advertisement and indicates that the user has acted in response to the advertisement in the past. The action history information is stored, the action history information is referred to, and the combination of the predetermined number of advertisements that has the largest number of non-overlapping users who have caused the action in the past is selected, and the selection is performed. Will be output.

  Further, another aspect of the present invention is a program for determining a plurality of advertisements, wherein the user has associated with the user and the advertisement, and the user has acted according to the advertisement in the past. And storing the action history information indicating, and referring to the action history information, the combination of the predetermined number of advertisements that has the largest number of non-overlapping users who have caused the action in the past is selected. And a step of outputting the selected advertisement.

  Other problems and solutions to be disclosed by the present application will be made clear by the embodiments of the invention and the drawings.

  ADVANTAGE OF THE INVENTION According to this invention, when displaying a some advertisement, possibility that a user will take action according to an advertisement can be made high.

It is a figure which shows an example of the screen 1 displayed in the advertising system of this embodiment. It is a figure explaining the outline | summary of the selection of the advertisement in this embodiment. It is a figure which shows the structural example of the advertisement display system of this embodiment. 2 is a diagram illustrating a hardware configuration example of a Web server 20. FIG. 3 is a diagram illustrating an example of a software configuration of a Web server 20. FIG. It is a figure which shows the structural example of advertisement information. It is a figure which shows the structural example of user information. It is a figure which shows the structural example of an impression log. It is a figure which shows the structural example of a click log. FIG. 4 is a diagram showing a flow of processing performed by a Web server 20. It is a figure which shows the flow of the selection process of the advertisement in step S408 of FIG. It is a figure which shows the flow of the selection process of the advertisement in the case of calculating a click rate and a correlation rate for every category.

== (1) Summary of Invention ==
Hereinafter, an advertisement display system according to an embodiment of the present invention will be described. In the present embodiment, the advertisement is assumed to be a so-called banner advertisement displayed when browsing a Web page. FIG. 1 is a diagram illustrating an example of a screen 1 displayed in the advertising system according to the present embodiment. A plurality of advertisements 2 are displayed on the screen 1 together with various information.

  The advertisement display system of the present embodiment is intended to increase the possibility that the advertisement 2 is clicked. Specifically, the number of non-overlapping users (unique users) who have clicked on the advertisement in the past is the largest. To choose an ad.

Here, consider that k advertisements (k <n) are selected from n advertisements and displayed on the same Web page. Assuming that the ratio of users who have clicked on the advertisement i (this ratio is referred to as the click rate in this embodiment) is p _i , the click rate for the entire web page is If the probability that the advertisement is not clicked is subtracted from 1, and each click rate is independent, it is almost equal to the sum of all click rates.

Therefore, as shown in FIG. 2 (1), when the click rate of three advertisements is p ₁ > p ₂ > p ₃ , when selecting two ads from these three ads, It is only necessary to select two in order from the advertisement. If the first advertisement and the second advertisement are selected, the click rate of the Web page can be maximized.

However, if the click rate is correlated as shown in FIG. 2 (2), that is, if there are many users who have clicked on both advertisements and other advertisements, the click rate is simply high. Even if selected in order, the click rate of the Web page cannot be maximized. In the example of FIG. 2B, when the ratio of the number of users who have clicked both the advertisement i and the advertisement j (hereinafter referred to as the correlation rate) is q _ij , the following formula (E2) As described above, the click rate for the entire Web page needs to be calculated after subtracting the correlation rate from the sum of the click rates of the two advertisements. In the present embodiment, the approximate calculation is performed using only the correlation rate of two advertisements without considering the correlation rate of three or more advertisements.

When the click rates of the three advertisements are p ₁ > p ₂ > p ₃ , the combination of the first advertisement and the second advertisement is optimal in the total value obtained by simply adding the click rates. since _{q 12} is quite large in the example), a first advertisement than the combination of considering the correlation factor _{p 1 + p 2 -q 12 <} p 1 + p 3 -q 13 , and the first advertisement and second advertisement The combination of the third advertisement has a higher click rate as a whole Web page.

  Therefore, in the advertisement display system of the present embodiment, the advertisement is selected according to the click rate obtained by subtracting the correlation rate so that the possibility that any one of the advertisements is clicked as a whole Web page is increased.

== (2) Overall system configuration ==
FIG. 3 is a diagram illustrating a configuration example of the advertisement display system according to the present embodiment. The advertisement display system according to the present embodiment includes a Web server 20 that is connected to the user terminal 10 via a communication network 30 so as to be communicable.

  The communication network 30 is, for example, the Internet or a LAN (Local Area Network) constructed by a public telephone line network, a dedicated communication line network, a mobile phone line network, a wireless communication network, Ethernet (registered trademark), or the like.

  The Web server 20 is a computer such as a workstation or a personal computer that provides various contents such as images, moving images, and texts according to HTTP (HyperText Transfer Protocol). The user terminal 10 is a computer such as a personal computer, a tablet computer, a mobile phone terminal, or a PDA (Personal Digital Assistant) that operates a browser for browsing various Web pages provided by the Web server 20.

  The user terminal 10 transmits an HTTP request to the Web server 20, receives a Web page described in HTML (HyperText Markup Language) transmitted from the Web server 20 in response to the request, and displays the Web page on the display. Information for specifying the user who is operating the user terminal 10 is set in the request. In this embodiment, it is assumed that user identification information (hereinafter referred to as a user ID) is set in the request. For example, for each session between the user terminal 10 and the Web server 20, the Web server 20 A user ID may be recorded as session information, and information specifying a session may be set in the request.

== (3) Hardware Configuration of Web Server 20 ==
FIG. 4 is a diagram illustrating a hardware configuration example of the Web server 20. The Web server 20 includes a CPU 201, a memory 202, a storage device 203, a communication interface 204, an input device 205, and an output device 206. The storage device 203 is, for example, a hard disk drive, a solid state drive, or a flash memory that stores various data and programs. The communication interface 204 is an interface for connecting to the communication network 30, for example, an adapter for connecting to Ethernet (registered trademark), a modem for connecting to a public telephone line network, and a wireless communication device for performing wireless communication A USB (Universal Serial Bus) connector for serial communication, an RS232C connector, or the like. The input device 205 is, for example, a keyboard, a mouse, a touch panel, a button, or a microphone that inputs data. The output device 206 is, for example, a display, a printer, or a speaker that outputs data.

== (4) Software Configuration of Web Server 20 ==
FIG. 5 is a diagram illustrating a software configuration example of the Web server 20. The web server 20 includes an HTTP processing unit 211, a web page creation unit 212, an advertisement selection unit 213, a log collection unit 214, a web information database 251, an advertisement database 252, a user database 253, an impression log storage unit 254, and a click log storage unit 255. It has.
Note that the functional units of the HTTP processing unit 211, the Web page creation unit 212, the advertisement selection unit 213, and the log collection unit 214 are realized by the CPU 201 included in the Web server 20 executing a program stored in the storage device 202. The Web information database 251, the advertisement database 252, the user database 253, the impression log storage unit 254, and the click log storage unit 255 are storage areas provided by the memory 202 and the storage device 203 provided in the Web server 20. Realized as part.

== (5) Description of Storage Unit ==
The web information database 251 stores information for creating a web page (hereinafter referred to as web information). The Web information is, for example, a file described in HTML, a program file storing a program executed by a mechanism such as CGI (Common Gateway Interface), an image file, or the like. The Web information database 251 can be realized by, for example, a file system directory or folder provided by an operating system operating on the Web server 20.

  The advertisement database 252 (corresponding to the bid amount storage unit and the category storage unit of the present invention) stores information for displaying an advertisement (hereinafter referred to as advertisement information). FIG. 6 is a diagram illustrating a configuration example of advertisement information stored in the advertisement database 252. The advertisement information is associated with the advertisement ID that is the identification information of the advertisement, the advertisement category, the URL (Uniform Resource Locator) related to the advertisement, the title of the advertisement, the bid amount, and the percentage of users who have clicked the advertisement. (Hereinafter referred to as click rate). The bid amount is a unit price paid by the advertiser when the advertisement is clicked, and is determined by the advertiser. The click rate is a ratio obtained by dividing the number of users who have clicked on advertisements in the past by the number of all users.

  In the present embodiment, the character string of the title is displayed as the advertisement 2 on the screen 1. The advertisement 2 is set with a link to a predetermined URL designating the advertisement ID (the Web server 20 is designated as the host name; hereinafter referred to as a redirect URL). When the advertisement 2 is clicked, a request including the advertisement ID is transmitted to the Web server 20, a log indicating that the click is collected by the log collection unit 214 described later, and the advertisement information corresponding to the advertisement ID is collected by the HTTP processing unit 211. (For example, a response specifying the URL as a redirect destination is transmitted). Thereby, the user's action (click) according to the advertisement can be detected.

  In addition, you may make it display an image, a moving image, etc. as an advertisement, or you may make it output an audio | voice, and in that case, URL which shows the data of the image displayed as an advertisement, or a moving image is added to advertising information, for example Can be.

The user database 253 (corresponding to the total number information storage unit of the present invention) stores information about users (hereinafter referred to as user information). FIG. 7 is a diagram illustrating a configuration example of user information stored in the user database 253. The user information includes a user name and a user attribute in association with the user ID. As the user attribute, for example, any attribute related to the user, such as the user's gender, age, residential area, working area, hobby, and object of interest, can be set. The user database 253 stores the number of user information registered in the user database 253 (that is, the number of all users, hereinafter referred to as the total number of users).
Note that the user database 253 may calculate the total number of users by not counting the total number of users but counting the number of user information.

The impression log storage unit 254 (corresponding to the impression storage unit of the present invention) stores information (hereinafter referred to as an impression log) indicating that an advertisement is displayed on the screen viewed by the user (referred to as an impression). To do. FIG. 8 is a diagram illustrating a configuration example of an impression log stored in the impression log storage unit 254. The impression log includes the date and time (delivery date and time) when the advertisement 2 was displayed on the screen 1, the user ID indicating the user who viewed the screen 1, the advertisement ID indicating the advertisement 2 displayed on the screen 1, and the category of the advertisement. It is.
In the example of the screen 1 in FIG. 1, since three advertisements 2 are displayed, when the user views the screen 1 once, three impression logs are registered.

  The click log storage unit 255 (corresponding to the action history storage unit of the present invention) stores information indicating that the user has clicked the advertisement 2 displayed on the screen 1 (hereinafter referred to as a click log). FIG. 9 is a diagram illustrating a configuration example of a click log stored in the click log storage unit 255. The click log includes a user ID indicating the clicked user, an advertisement ID indicating the clicked advertisement 2, a category of the advertisement, and a date and time (click date and time).

== (6) Description of Functional Units ==
The HTTP processing unit 211 performs HTTP processing. The HTTP processing unit 211 receives an HTTP request transmitted from the user terminal 10.
When the redirect URL is specified in the request, the HTTP processing unit 211 calls the log collection unit 214 to register the click log, and then transmits the advertisement information corresponding to the advertisement ID included in the request to the advertisement database 252. And a response specifying the URL included in the read advertisement information as a redirect destination is returned to the user terminal 10.
When a URL indicating a Web page other than the redirect URL is specified in the request, the HTTP processing unit 211 calls information for displaying the Web page by calling the Web page creation unit 212 (hereinafter referred to as screen information). The information is, for example, an HTML file), the image information created by the web page creation unit 212 is responded to the user terminal 10, and the log collection unit 214 is called to register the impression log.

  The web page creation unit 212 (corresponding to the advertisement output unit of the present invention) creates screen information for displaying a web page. The web page creation unit 212 calls the advertisement selection unit 213 to select the advertisement 2 to be displayed on the screen 1, and when the selected advertisement is displayed as the advertisement 2 and the advertisement 2 is clicked, the redirect URL is accessed. For example, a character string in which a link tag is set is inserted into the screen information. In addition, about the process in which the web page preparation part 212 produces various contents other than the advertisement 2, the process by a general web server is used, and description is abbreviate | omitted here.

  The advertisement selection unit 213 selects an advertisement to be displayed on the web page. The advertisement selection unit 213 includes a distribution candidate extraction unit 231, a click rate calculation unit 232, a correlation rate calculation unit 233, and an advertisement determination unit 234. The distribution candidate extraction unit 231, the click rate calculation unit 232, the correlation rate calculation unit 233, and the advertisement determination unit 234 are also realized by the CPU 201 reading out the program stored in the storage device 203 to the memory 202 and executing it. .

  The distribution candidate extraction unit 231 determines advertisement candidates. The distribution candidate extraction unit 231 reads a predetermined number (for example, any number such as 10, 50, 100, etc.) of advertisement information from the advertisement database 252. In this embodiment, the distribution candidate extraction unit 231 reads a predetermined number of pieces of advertisement information in descending order of click rate. For example, the advertisement information may be read according to user attributes, or for each advertisement. In addition, evaluation may be performed according to CTR, CPM, eCPM, etc., and a predetermined number of advertisement information may be read out in descending order of evaluation. Note that a general advertisement selection process can be used for the process in which the distribution candidate extraction unit 231 selects a predetermined number of advertisements from among a number of advertisements.

  The click rate calculating unit 232 (corresponding to the action number calculating unit of the present invention) calculates the click rate of the advertisement. In this embodiment, the click rate calculation unit 232 calculates the click rate and updates the advertisement information when the log collection unit 214 (to be described later) registers the click log. You may make it calculate a click rate about the candidate which the extraction part 231 determined. When the click log is registered in the click log storage unit 255, the click rate calculation unit 232 counts the number of non-overlapping user IDs corresponding to the advertisement ID of the click log in the click log storage unit 255, and the counted number Is divided by the total number of users to calculate the click rate. The click rate calculation unit 232 updates the click rate of the advertisement database 252 corresponding to the advertisement ID of the click log to the calculated click rate.

  The correlation rate calculation unit 233 (corresponding to the correlation action number calculation unit of the present invention) calculates the correlation rate. The correlation value calculation unit 233 calculates the correlation rate by dividing the value obtained by dividing the number of users who have clicked both of the two advertisements by the total number of users by the value obtained by multiplying the click rates of both advertisements.

なお、式（Ｅ６）は、ｑ_ａｂがクリック率ｐ_ａ、ｐ_ｂを超えないように調整するための計算処理である。 The total number of users and U, the number of users who have clicked on an ad a, i.e. the number of duplicate no user ID corresponding to the advertisement a in click log storage unit 255 as u _a, the click rate for advertising a p _a And the number of users who have clicked the advertisement b, that is, the number of non-overlapping user IDs corresponding to the advertisement b in the click log storage unit 255 is u _b, and the click rate for the advertisement b is p _b , The correlation rate _qab is calculated by the following equations (E3) to (E6).

Expression (E6) is a calculation process for adjusting q _ab so that it does not exceed the click rates p _a and p _b .

  The advertisement determination unit 234 determines an advertisement to be displayed from the candidates determined by the distribution candidate extraction unit 231. The advertisement determination unit 234 determines the advertisement so that the value obtained by subtracting the total correlation rate of each pair of advertisements from the total click rate becomes the largest.

In the present embodiment, the advertisement determination unit 234 first determines an advertisement j having the highest click rate from advertisement candidates as an advertisement to be displayed and excludes it from the candidates. The advertisement determination unit 234 calculates the correlation rate q _ij for each remaining candidate advertisement i, subtracts the correlation rate q _ij from the click rate p _i, and subtracts the correlation rate from the click rate p _j Decide on the largest ad to display and exclude it from the candidate. By repeating this until a predetermined number of advertisements are selected, the advertisement is determined such that the value obtained by subtracting the sum of the correlation rates of each pair of advertisements from the total click rate becomes the largest.

  The log collection unit 214 registers an impression log and a click log. When the redirect URL is specified in the HTTP, the log collection unit 214 reads out the category corresponding to the advertisement ID specified in the redirect URL from the advertisement database 252 and the user ID indicating the user who transmitted the HTTP request A click log in which the advertisement ID, the read category, and the current date and time are set is created and registered in the click log storage unit 255. In addition, the log collection unit 214 calls the click rate calculation unit 232 to update the click rate. When a URL other than the redirect URL is specified in HTTP, the log collection unit 214 displays the current date and time, the user ID of the user who transmitted the HTTP request, and the advertisement for each advertisement selected by the advertisement selection unit 213. An impression log including the advertisement ID to be displayed and the category of the advertisement is created and registered in the impression log storage unit 254.

== (7) Explanation of Processing ==
FIG. 10 is a diagram illustrating a flow of processing performed by the Web server 20.
When receiving an HTTP request from the user terminal 10 (S401), the HTTP processing unit 211 determines whether or not an advertisement has been clicked based on whether or not a redirect URL is specified in the request (S402).

  When it is determined that the advertisement is clicked (S402: YES), the log collection unit 214 reads advertisement information corresponding to the advertisement ID set in the redirect URL from the advertisement database 252, and the user ID set in the request, A click log including the advertisement ID, the category of advertisement information, and the current date and time is created and registered in the click log storage unit 255 (S403).

The click rate calculation unit 232 calculates the number of non-overlapping user IDs corresponding to the advertisement ID set in the redirect URL from the click log storage unit 255 as the number of clicks (S404), and the total number of users read from the user database 253 The click rate is calculated by dividing the number of clicks (S405). The click rate calculation unit 232 updates the click rate of the advertisement database 252 corresponding to the advertisement ID set in the redirect URL to the calculated click rate (S406).
The HTTP processing unit 211 transmits an HTTP response to the user terminal 10 so as to redirect to the URL of the advertisement information (S407).

  On the other hand, if it is determined that the advertisement has not been clicked (S402: NO), an advertisement list including advertisement information of the advertisement selected by the advertisement selection unit 213 is created by the process of FIG. 11 (S408). The advertisement information selection process shown in FIG. 11 will be described later.

  The web page creation unit 212 creates screen information such as an HTML file for displaying various contents in response to the request (S409). Note that the screen information creation processing in step S409 can employ processing by a general Web server.

  The web page creation unit 212 creates a redirect URL in which the advertisement ID included in the advertisement information is set as an argument of a predetermined URL for each advertisement information included in the advertisement list (S410). The redirect URL can be in the form of “www.foobar.com/ad?id=xxxx”, for example. For each advertisement information included in the advertisement list, the web page creation unit 212 includes a link tag to a redirect URL including the title included in the advertisement information (for example, “<a href =" http: //www.foobar. com / ad? id = xxxx "> title </a>") is inserted into the screen information (S411).

  The HTTP processing unit 211 transmits the created screen information to the user terminal 10 as a response to the request (S412). The log collection unit 214 creates an impression log including the user ID set in the request, the advertisement ID and category included in the advertisement information, and the current date and time for each advertisement information included in the advertisement list, and the impression log storage unit It is registered in H.254 (S413).

FIG. 11 is a diagram showing the flow of advertisement selection processing in step S408.
The distribution candidate extraction unit 231 extracts a predetermined number m pieces of advertisement information from the advertisement database 252 to obtain a candidate list (S421). Note that what kind of advertisement information is extracted from the advertisement database 252 can be arbitrarily set. For example, m pieces of advertisement information may be extracted at random, or m pieces of advertisement may be extracted in descending order of click rate. In addition, a category in which the user is interested may be stored as an attribute of user information, and m pieces of advertisement information matching the category may be extracted.

  The click rate calculation unit 232 sets the click rate corresponding to each advertisement information included in the candidate list to the variable p (advertisement ID) (S422), and the advertisement determination unit 234 sets 0 to the variable n ( S423).

The advertisement determining unit 234 sets the advertisement ID having the largest click rate p (advertisement ID) to a (S424), deletes the advertisement information corresponding to a from the candidate list, adds it to the advertisement list (S425), and sets n to Increment (S426).
If n is equal to or less than the number k (3 in the example of FIG. 1) of advertisements 2 displayed on screen 1 (S427: YES), the following processing is performed for each advertisement information included in the candidate list.

The correlation value calculation section 233, the advertisement ID of the advertisement information as a b (S428), the click log storage unit 255, extracts without duplication user ID corresponding to a and _{L a} (S429), corresponding to b user extracts the ID without duplication and _{L b} (S430), it extracts the user ID contained in both the _{L a} and _{L b} and _{L ab} (S431). The correlation value calculation section _233, and _{u ab} by counting the number of user ID included in the _{L ab} (S432), the quotient obtained by dividing the _{u ab} in the total number of users acquired from the user database 253 and P (S433). The correlation value calculation unit 233 calculates Q _ab by dividing P by the value obtained by multiplying the click rate p (a) corresponding to a by the click rate p (b) corresponding to b (S434), The correlation rate _qab is calculated by E6) (S435). As a result, q _ab is calculated according to equations (E3) to (E6). Correlation value calculation unit 233 subtracts the correlation factor _{q ab} from CTR corresponds to b p (b) (S436) .

After the above processing is repeated for each advertisement information in the candidate list, the processing from step S424 is repeated until n exceeds k (S427).
In this way, an advertisement combination that maximizes the value obtained by subtracting the correlation rate of each pair of advertisements from the total click rate is registered in the advertisement list.

  As described above, according to the advertisement display system of the present embodiment, the combination of advertisements 2 displayed on the screen 1 is maximized so that the value obtained by subtracting the correlation rate of each advertisement pair from the total click rate is maximized. You can choose. That is, the combination of the advertisements 2 can be determined so that the click rate for the entire Web page is increased. Therefore, the possibility that an advertisement is clicked on the entire screen can be increased. Therefore, it is possible for the advertiser to maximize the advertising effect, and for the publisher of the advertisement (operator of the Web server 20), if an advertising fee is generated by clicking, the profit can be improved. It becomes possible.

  In addition, it is possible to calculate the click rate and the correlation rate for all combinations that extract k advertisements from the candidate items. The click rate for the entire Web page can be calculated by selecting one large advertisement and subtracting the correlation rate with the selected advertisement from the click rate for the other candidates. Thereby, the calculation load concerning calculation of the click rate of the whole Web page can be reduced.

== (8) Modification 1: Category ==
In the present embodiment, the advertisement is directly selected based on the click rate and the correlation rate, but the category may be selected based on the click rate and the correlation rate.
In this case, in formulas (E3) to (E6), u _a and u _b are the number of users who clicked advertisements belonging to categories a and b, respectively, and u _ab is an advertisement belonging to category a and category b. Let it be the number of users who clicked both with the advertisement to which they belong.

  The flow of the advertisement selection process in this case is shown in FIG. The process of FIG. 12 is a modification of the process of FIG. 11 so as to select an advertisement according to the category, and therefore the difference from the process of FIG. 11 will be described. In the process of FIG. 12, in step S421, the distribution candidate extraction unit 231 reads out categories from the advertisement database 252 without duplication, and extracts m categories from them. The category can be selected by an arbitrary process. For example, the selection may be made randomly or in a predetermined order. Moreover, you may make it select according to a user's attribute.

  In step S422, for each category included in the candidate list, the click rate calculation unit 232 counts the number of non-overlapping user IDs corresponding to the category from the click log storage unit 255, and calculates the counted number as the number of users. The click rate p (category) is calculated by dividing.

  In step S424, the advertisement determination unit 234 sets a category having the largest click rate p (category) as a, deletes a from the candidate list, and adds it to the category list. The processing in steps S428 to S436 is performed for each category included in the candidate list. In step S428, the correlation value calculation unit 233 sets the category to be processed as b. When n exceeds k (S427: NO), the advertisement determination unit 234 selects one advertisement information corresponding to the category and adds it to the advertisement list for each category included in the category list (S437). .

  An arbitrary process can be used for the process of determining the advertisement belonging to the category. For example, the advertisement information corresponding to the category may be determined at random, or may be determined according to the user attribute. In this way, it is possible to select an advertisement of a category that is likely to be clicked by the user using the click rate and the correlation rate for each category. When the number of advertisements is large, it is possible to reduce the load related to the calculation of the click rate and the correlation rate by calculating for each category. Thereby, the response time of the Web server 20 can be shortened.

== (9) Modification 2: Number of impressions ==
In this embodiment, the click rate and the correlation rate are calculated from the number of users. However, the click rate and the correlation rate may be calculated from the number of advertisements displayed to the user (the number of impressions).
In this case, in the above formula (E3), U is the total number of impressions of the advertisement, that is, the number of impression logs, u _a is the number of impressions for the user who clicked the advertisement a, and u _b is the user who clicked the advertisement b The number of impressions for.
In the process of FIG. 11, in step S422, the click rate calculation unit 232 sets the total number of impression logs registered in the impression log storage unit 254 as U, and stores the click log for each advertisement information included in the candidate list. The user ID corresponding to the advertisement ID of the advertisement information is extracted from the unit 255 without duplication, and the number of impression logs corresponding to any of the extracted user IDs is counted from the impression log storage unit 254 as u, and u is U The click rate p (advertisement ID) is calculated by dividing. In step S432, the correlation value calculation section _233, and _{u ab} by counting the number of impressions log corresponding to any of the user ID included in the _{L ab} from CPM log storage unit 254. Thereby, a click rate and a correlation rate can be calculated using the number of impressions.

== (10) Modification 3: Number of clicks ==
In the present embodiment, the advertisement is determined based on the click rate, but the number of clicks may be used. In this case, the following formula (E7) is used instead of the above formulas (E3) to (E6).

In the process of FIG. 11, in step S422, the click rate calculation unit 232 counts the number of user IDs corresponding to the advertisement ID from the click log storage unit 255, and sets the number of clicked users as p (advertisement ID). S433 and S434 are omitted, and in step S435, the correlation value calculating unit 233 sets q _{ab as} a quotient obtained by dividing u _ab by a value obtained by multiplying p (a) and p (b).

  Thereby, the advertisement can be selected so that the number of unique users who have clicked the advertisement in the past is maximized. Therefore, it is possible to increase the possibility that an advertisement is clicked by a large number of users, improve the advertising effect for the advertiser, and increase the profit for the publisher.

== (11) Modification 4: Calculation from combination ==
In this embodiment, the overall click rate is obtained by subtracting the correlation rate from the click rate. However, the number of unique users who clicked in the past and the click rate are calculated from the combination of advertisements. You may do it. For example, the correlation value calculation unit 233 determines the number of non-overlapping user IDs corresponding to any one of the k advertisement IDs for all combinations in which k advertisements are selected from the advertisement candidates selected by the distribution candidate extraction unit 231. You may make it count from the click log memory | storage part 255, and may select the combination with the largest number of the counted unique users, or the value which divided this number by the total number of users.

== (12) Modification 5: Consideration of advertising unit price ==
In this embodiment, an advertisement having a large number of unique users is selected. However, the unit price of the advertisement may be considered. In this case, in step S424 in FIG. 11, the advertisement determination unit 234 selects, as a, the advertisement ID that maximizes the value obtained by multiplying the click rate p (advertisement ID) by the bid amount of the advertisement information. As a result, it is possible to maximize profit on the website operator (advertisement publisher) side.

== (13) Other Modifications ==
Moreover, in this embodiment, although it was determined whether the advertisement was clicked, you may use it, if the action according to an advertisement can be detected. For example, the Web server 20 can receive notification from the store computer that the customer has visited the store by looking at the advertisement, and the click log can be registered accordingly.

  In the present embodiment, no condition is added to the aggregation period of the impression log and the click log. However, the click rate and the correlation rate may be calculated based on the impression log and the click log in the past predetermined period. .

  Although the present embodiment has been described above, the above embodiment is intended to facilitate understanding of the present invention and is not intended to limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and the present invention includes equivalents thereof.

  For example, in this embodiment, the advertisement display system is configured by one Web server 20, but is not limited thereto, and may be configured by a plurality of computers. For example, a web server that provides a web page, an application server that creates a web page, and a database server that manages various data can be used.

  In addition, the advertisement is displayed on a part of the web page. However, the present invention is not limited to this, and a screen on which only the advertisement is displayed may be displayed. May be displayed.

1 screen 2 advertisement 10 user terminal 20 web server 30 communication network 211 HTTP processing unit 212 web page creation unit 213 advertisement selection unit 214 log collection unit 231 distribution candidate extraction unit 232 click rate calculation unit 233 correlation rate calculation unit 234 advertisement determination unit 251 Web information database 252 Advertising database 253 User database 254 Impression log storage unit 255 Click log storage unit

Claims (8)

A device that outputs a predetermined number of advertisements to be displayed from a plurality of advertisements,
An action history storage unit that stores action history information indicating that the user has taken action in accordance with the advertisement in the past in association with the user and the advertisement;
With reference to the behavior history storage unit, an advertisement selection unit that selects, from the predetermined number of advertisement combinations, the one having the largest number of non-overlapping users who have caused the behavior in the past;
An advertisement output unit for outputting the selected advertisement;
An advertisement output device comprising: The advertisement output device according to claim 1,
The advertisement selection unit
For each of the advertisements, an action number calculating unit that calculates an action number that is the number of the users corresponding to the advertisement from the action history storage unit;
An advertisement determining unit that determines a first advertisement to be displayed from the plurality of advertisements according to the number of actions;
For each of the other advertisements different from the first advertisement, a correlation that is the number of users corresponding to the other advertisements among the users corresponding to the first advertisement from the action history storage unit. A correlation action number calculation unit for calculating the number of actions;
With
The advertisement determining unit further determines a second advertisement to be displayed together with the first advertisement from the plurality of advertisements according to a value obtained by subtracting the number of correlated actions from the number of actions.
Advertising output device characterized by The advertisement output device according to claim 2,
A total number information storage unit for storing total number information for determining the total number of users,
The action number calculation unit calculates an action rate that is a value obtained by dividing the action number by the total number,
The advertisement determination unit determines the first advertisement according to the action rate,
For each of the other advertisements, the correlation action number calculation unit calculates a value obtained by dividing the number of correlation actions by the total number, and sets the action rate for the other advertisement to the action rate for the first advertisement. Calculate the correlation rate, which is the quotient divided by the multiplied value,
The advertisement determination unit determines the second advertisement according to a value obtained by subtracting the correlation rate from the activity rate;
An advertisement output device characterized by the above. The advertisement output device according to claim 3,
A bid amount storage unit for storing a bid amount for each advertisement;
The advertisement determination unit
The first advertisement is determined according to a value obtained by multiplying the behavior rate by the bid amount,
Determining the second advertisement according to a value obtained by multiplying a value obtained by subtracting the correlation rate from the activity rate and the bid amount;
An advertisement output device characterized by the above. The advertisement output device according to claim 1,
A category storage unit for storing a category to which the advertisement belongs for each advertisement;
The action history information includes category specifying information for specifying the category of the advertisement,
The advertisement selection unit refers to the behavior history storage unit, selects a combination of the predetermined number of the categories that has the largest number of non-overlapping users who have caused the behavior in the past, Selecting the advertisement corresponding to the selected category;
An advertisement output device characterized by the above. A device that outputs a predetermined number of advertisements to be displayed from a plurality of advertisements,
An impression storage unit that stores impression information for calculating the number of impressions corresponding to the user and the advertisement, and the number of impressions displayed for the user;
In association with the user and the advertisement, an action history storage unit that stores action history information indicating that the user has taken action in accordance with the advertisement in the past;
Referring to the action history storage unit and the impression storage unit, select the combination of the predetermined number of advertisements that has the largest total of the number of impressions corresponding to the non-overlapping user who has caused the action in the past An ad selector to
An advertisement output unit for outputting the selected advertisement;
An advertisement output device comprising: A method of outputting multiple advertisements,
Computer
In association with the user and the advertisement, storing action history information indicating that the user has taken an action corresponding to the advertisement in the past,
Referring to the action history information, select the one having the largest number of non-overlapping users who have caused the action in the past from among the predetermined number of advertisement combinations,
Outputting the selected advertisement;
A method for determining advertisements. A program for outputting a plurality of advertisements,
On the computer,
Storing action history information indicating that the user has taken action in accordance with the advertisement in the past in association with the user and the advertisement;
Referring to the action history information, selecting a combination of the predetermined number of advertisements that has the largest number of non-overlapping users who have caused the action in the past;
Outputting the selected advertisement;
A program for running

Images