JP4948573B2 - Advertisement billing apparatus and method - Google Patents

Description

  The present invention relates to billing calculation for web advertisements.

  In the field of so-called web advertisements such as the Internet, selectively display advertisements corresponding to attributes (hereinafter referred to as “registered attributes”) such as gender, age, and residential area input by registered users during ID registration. There is a service that is called targeting advertising. In the targeting advertisement, the advertisement is displayed for a predetermined number of times (for example, the corresponding advertisement is displayed when the user identified by the login ID or the like corresponds to the attribute of the display target person for each advertisement. (Thousand times) It is determined in advance in the form of the unit price per display and the unit price per click.

  In addition, even without user registration or the like, a technique has been proposed in which an attribute is estimated from an action history of a accessing user, that is, a viewed page or a searched keyword, and an advertisement suitable for the attribute is selected and displayed (for example, (See Patent Document 1 and Non-Patent Document 1).

JP 2003-150844 A

Search Terrier Co., Ltd., “Attribute Inference Targeting”, [online], [Search June 28, 2009], Internet <URL: http://www.searchteria.co.jp/news/pr20090106.html>

  However, the estimated attribute (hereinafter referred to as “estimated attribute”) has characteristics such as the degree of certainty each time it is displayed, depending on from which point in the action history it is estimated, compared to the registered attribute. Because it is different, there is a problem that it is not suitable for calculation of advertising fee based on a uniform unit price, especially when a large number of impressions such as tens of thousands of times or a combination of multiple attributes to narrow down the target audience Due to the complexity, actual implementation was difficult.

  The present invention is based on the above-mentioned problem. The purpose of the present invention is to provide an objectively valid advertisement in a targeting advertisement using an estimated attribute, including a complicated and large-scale case, by calculation in proportion to the corresponding ratio of the estimated attribute. It is to realize billing.

  Based on the above object, according to one aspect of the present invention, in an advertising billing apparatus in which a computer calculates a billing amount for an advertisement distributed to a user terminal accessed via a communication network, the user terminal is identified and A plurality of pattern data that associates an estimated attribute value and the corresponding rate with an identification recording means for storing an action history in access in a predetermined history storage means and an element value included in the action history or a combination thereof in advance. Storing pattern data storage means; estimation means for estimating an attribute value by comparing the action history corresponding to the identified user terminal with each pattern data; contents of each advertisement; The advertisement storage means storing the attribute value of the target person who becomes the condition and the advertising unit price in association with each other, and the advertisement corresponding to the estimated attribute value The advertisement distribution means that searches from the advertisement storage means and distributes to the user terminal, and the advertisement distributed based on the estimated attribute value are associated with the attribute value in the pattern data used for the estimation of the attribute value. And a monetary amount calculating means for calculating a charge amount by multiplying the corresponding rate by the advertising unit price corresponding to the advertisement.

  Another aspect of the present invention captures the above aspect from the viewpoint of a method, and stores the contents of each advertisement, the attribute value of the target person as a delivery condition, and the advertising unit price in association with each other. An advertisement storage means, and a pattern data storage means for storing in advance a plurality of pattern data in which estimated attribute values and corresponding rates are associated with element values included in the action history or a combination thereof. An advertisement billing method in an advertisement billing apparatus in which a computer calculates a billing amount of an advertisement distributed to a user terminal that is accessed via a communication network, wherein the user terminal is identified and an action history in access for each user terminal is predetermined. The attribute record value is obtained by collating the identification record process to be stored in the history storage means, the action history corresponding to the identified user terminal and the pattern data. An estimation process to be determined, an advertisement distribution process for retrieving an advertisement corresponding to the estimated attribute value from the advertisement storage means and distributing it to the user terminal, and an estimation of the attribute value of the advertisement distributed based on the estimated attribute value And an amount calculation process for calculating a charge amount by multiplying the corresponding rate associated with the attribute value in the pattern data used in the above by the advertising unit price corresponding to the advertisement.

  In this way, for each ad delivery based on the estimated attribute, the corresponding rate corresponding to the estimated attribute value is multiplied by the advertising unit price of the registered attribute to calculate the charge amount, and according to the certainty of the attribute The advertiser pays the advertising fee, and objectively reasonable advertising billing can be realized even when multiple attributes are combined or the number of cases is large.

  According to another aspect of the present invention, in any one of the above aspects, the advertisement storage unit stores a plurality of the attribute values for one advertisement. , And other attribute values are stored in association with the premium rate or premium when applicable, and the amount calculation means (process) uses the estimated attribute value for the other attribute value. In this case, the corresponding rate associated with the attribute value in the pattern data used for estimating the attribute value is multiplied by the premium rate or the premium amount and added to the unit price of the one attribute value. It is characterized by.

  In this way, in search-linked advertising, etc., an additional attribute that adds a surcharge if it is applicable to the required keyword specification is specified, and if the additional attribute is an estimated attribute, the corresponding rate is also multiplied by the corresponding rate Thus, even in a complicated case where an estimated attribute is combined with an essential attribute value, objectively valid advertisement billing can be easily realized.

  It should be noted that other categories such as methods or programs corresponding to the above aspects and more specific aspects described below are also included in the present invention.

  As described above, according to the present invention, objectively appropriate advertising billing is realized in a targeting advertisement using an estimated attribute, including a complicated large-scale case, by calculation in proportion to a corresponding ratio of the estimated attribute. It becomes possible.

The functional block diagram which shows the structure of embodiment of this invention. The figure which illustrates each information (data) in the embodiment of the present invention. The flowchart which shows the process sequence in embodiment of this invention.

  Next, modes for carrying out the present invention (referred to as “embodiments”) will be described with reference to the drawings. It should be noted that assumptions common to those already described in the background art and problems are omitted as appropriate.

[1. Constitution〕
In the present embodiment, as shown in FIG. 1 (configuration diagram), an advertising billing device (hereinafter referred to as “this device”) in which a computer calculates a billing amount of an advertisement distributed to a user terminal T2 accessed via a communication network. 1). As an actual configuration, the entire apparatus 1 can be realized by a plurality of server devices, a single means can be realized by a plurality of server devices, a plurality of means can be realized by a single server device, The system including T (T1, T2, etc.) may be grasped as a system that is an object of the present invention. Also, some functions may be realized by calling an external platform or the like with an API (Application Program Interface) or network computing.

  In other words, FIG. 1 is a simple configuration example for explanation. In this example, the apparatus 1 includes at least an arithmetic control unit 5 such as a CPU and an external storage device (HDD) as a general computer configuration. Etc.) and a storage device 6 such as a main memory, and communication means 7 (such as a LAN adapter) for communication with a communication network (such as the Internet or a cellular phone network) (not shown). Each terminal T (T1, T2, etc.) accesses the apparatus 1 with a web browser or the like using a personal computer or a mobile phone terminal device.

  In the present apparatus 1, a predetermined computer program (not shown) stored (installed) in advance in the storage device 6 controls the arithmetic control unit 5 so that elements (10, 15,...) Shown in FIG. Realize. Among these elements, each means other than the storage means is a processing means that realizes / executes an information processing function / action described later in the section of “action”.

  The information storage means can be realized in the storage device 6 in any format such as various databases (also referred to as “DB”), variables such as files and arrays, and system setting values. Among them, the pattern data storage means 20 has an estimated attribute value for the element value included in the action history or a combination thereof, and a corresponding rate that means the probability, certainty, reliability, certainty, etc. of the attribute value. Are stored in advance.

  Here, FIG. 2 (1) shows an example of pattern data. Each row has an attribute value with respect to a combination pattern of a browsing history and a search keyword that is an element value of an action history. Are the corresponding rates. For example, the pattern data of the pattern ID “P51” has a browsing history of a web page whose theme is “independent” (independent opening) in a certain website (“1”), and the search engine uses “martial arts” as a keyword history. Yes (“1”), but without “embroidery” (“0”), the user has the attribute value “male” with the corresponding rate of “80%” (if the reverse, the attribute with the corresponding rate of “20%”) The value “female”), where “baby” is a theme that is not included in the pattern of the ID (“−”).

  How to obtain such pattern data for attribute estimation is up to you, but as an example, first, based on the registered attributes of a large number of registered users who have registered attributes, A combination of element values that is considered to have a large number of persons with a specific attribute is created by a data mining technique such as statistical analysis. Then, by extracting the registered users that actually correspond to the combination of element values, and summing up the proportions that correspond to the attributes, the corresponding rates can be obtained, so if combined with the corresponding combination of measures, Pattern data.

[2. Action)
The apparatus 1 configured as described above operates as follows.
[2-1. (Overview of action)
First, as an overview, the identification recording means 10 identifies an unspecified number of user terminals T2, stores an action history in access for each user terminal T2 in a predetermined history storage means 15, and the estimation means 30 For T2, the attribute value is estimated by comparing the action history with the pattern data. The advertisement distribution unit 40 searches the advertisement storage unit 35 for an advertisement corresponding to the estimated attribute value (also referred to as “estimated attribute value”) and distributes the advertisement to the user terminal T2. For the advertisement distributed based on the estimated attribute value, the monetary amount calculation means 50 determines the corresponding rate associated with the attribute value in the pattern data used for the estimation of the attribute value as the advertisement corresponding to the advertisement. Multiply the unit price to calculate the billing amount.

[2-2. Concrete example〕
For a more specific explanation, the following simple advertising product is assumed. In other words, the advertiser submits the advertisement content, the attribute value of the target person (for example, male or female, age, etc., also referred to as “target attribute value”) as a condition for displaying the advertisement, A bid price that is a kind of bid is specified, an advertisement with a high bid price is preferentially displayed, and the advertiser is charged for the bid price each time an advertisement is displayed or clicked. An actual advertisement product involves a combination different from this or a more complicated calculation method, but this simple example is used for explanation.

  In this case, in the present apparatus 1, when the submission accepting means 8 accepts the submission of a targeting advertisement that is a web advertisement from the advertiser terminal T1, that is, when receiving an order, for example, uploading an image file, for example, At the same time, it accepts the input of the URL of the destination web page, such as a hyperlink when the image is clicked, and also accepts the input of the target attribute value and the bid price as described above, and associates them with each other As illustrated in FIG. 2B, the advertisement storage unit 35 stores the advertisement.

  Here, the flowchart of FIG. 3 records the action history for the access from the user terminal and periodically performs the attribute estimation based on it, and accesses the attribute estimated in combination with the registered attribute. The processing procedure for realizing the cycle of displaying the advertisement on the spot is shown.

  That is, each time the identification recording means 10 receives a page request for requesting the display of a web page with an advertisement display frame as an access from an unspecified number of user terminals T2 (step S10), the identification recording means 10 designates the user terminal T2 as a cookie of the web browser. (Step S12). Then, the advertisement distribution means 40 attempts to retrieve and read the registered attribute from the registered attribute storage means 13 and to retrieve and read the estimated attribute from the estimated attribute storage means 32 as attributes corresponding to the identified user terminal T2. An advertisement corresponding to the obtained attribute (that is, an advertisement in which at least one of the registered attribute or the estimated attribute corresponding to the user terminal T2 matches the target attribute value) is retrieved from the advertisement storage unit 35 and accessed. At the same time, the advertisement is distributed to the user terminal T2 (step S14).

  Here, with respect to the registration attribute, the registration receiving unit 12 receives the registration from each user terminal T2 and stores it in the registration attribute storage unit 13 at the time of user registration, and the advertisement distribution unit 40 refers to it. The estimated attribute is estimated by the estimating means 30 from the action history (for example, FIG. 2 (3)) stored in the history storage means 15 by the estimation recording means 10 and stored in the estimated attribute storage means 32 as will be described later. The advertisement distribution means 40 refers to it. Note that the process of selecting a specific advertisement based on some criteria for access is also referred to as “reserve”.

  For example, for the user terminal T2 that corresponds to the pattern (ID: “P51”) in FIG. 2A and is estimated to be “80%” “male”, the advertisement (ID: “X87” in FIG. 2B). Since the target attribute value of “−02”) is “male”, it is conceivable to display this advertisement.

  Then, for the advertisement distributed based on the registration attribute (step S16: “NO”), the amount calculation unit 50 stores the bid price (for example, “100 yen”) as it is in the advertisement fee storage unit 55 (step S16). S18) For the advertisement distributed based on the estimated attribute (step S16: “YES”), the corresponding rate associated with the attribute value in the pattern data used for estimating the attribute value corresponds to the advertisement. A charge amount is calculated by multiplying the bid price and recorded in the advertisement fee storage means 55 (step S20).

  For example, when the advertisement (ID: “X87-02”) in FIG. 2 (2) is displayed based on the estimated attribute “male” based on the pattern (ID: “P51”) in FIG. 2 (1), FIG. As shown in 4), the charge rate “¥ 80” is calculated by multiplying the corresponding rate “80%” by the bid price “100 yen” and recorded in the advertisement fee storage means 55. Based on the billing amount information (for example, FIG. 2 (4)) recorded in the advertisement fee storage means 55, the billing processing means 60 sends the advertiser terminal T1 to the advertiser terminal T1 at a predetermined timing such as a monthly eve. Then, processing related to billing for advertisements, such as accounting of billing amounts, handling of details and responding to inquiries, is performed.

  Furthermore, for each access from the user terminal T2, the identification recording means 10 records an action history (for example, FIG. 2 (3)) in the current access in the predetermined history storage means 15 as an action history (step S22). Then, the estimation unit 30 performs an action history (for example, FIG. 5) stored in the history storage unit 15 as described above for each user terminal T2 at each predetermined timing for attribute estimation such as night batch processing (step S24). 2 (3)) and each pattern data (for example, FIG. 2 (1)) stored in the pattern data storage means 20, the attribute value is estimated, and the result is stored in the estimated attribute storage means 32. It is recorded by appending or updating (step S26), and used for subsequent advertisement allocation.

[3. effect〕
As described above, in this embodiment, for each advertisement delivery based on the estimated attribute, the attribute rate is calculated by a simple process of calculating the charge amount by multiplying the corresponding rate corresponding to the estimated attribute value by the advertising unit price of the registered attribute. The advertiser pays the advertising fee according to the certainty, and even if a plurality of attributes are combined or the number of cases is large, objectively appropriate advertising billing can be realized.

[4. Other embodiments]
In addition, said each embodiment is only an illustration, and this invention includes what is illustrated below and other embodiment other than that. For example, in the present invention, elements such as each means may be realized not only by an arithmetic control unit of a computer but also by an electronic circuit based on wired logic or the like, or a non-Neumann type information processing mechanism that will appear in the future. Further, each configuration diagram, data diagram, flowchart diagram, and the like are merely examples, and the presence / absence of each element, its order, specific contents, and the like can be appropriately changed.

  For example, in the above-described embodiment, an example in which the estimated attribute is used in combination with the registered attribute is shown. However, for the registered attribute, the combined use in the allocation of advertisement is not essential, and the concept is exclusively related to the standard advertisement unit price. Also good. In the above embodiment, the bid price is exemplified as the advertisement unit price. However, since the application target of the present invention is not limited to the bid-type advertisement such as the search-linked advertisement, the advertisement unit price is determined between the price list and the advertiser. It may be a fixed unit price that is agreed upon in a contract or the like, and, of course, it is not limited to click billing, and billing standards such as display number (impression) billing and contracted sales (conversion) billing after access are free.

  In the above embodiment, the advertiser specifies only basic attributes such as gender age (so-called “demographic”), and in the case of attribute estimation, the original bid is multiplied by the corresponding rate. Since this is merely an illustrative example, the present invention is not limited to this, and can be applied to advertisements of other patterns.

  For example, the present invention can be applied to a complicated case where an estimated attribute value is combined with an essential attribute value. Specifically, for example, when an essential keyword (for example, “real estate”) designated by an advertisement is input by a search engine, the advertisement is displayed together with a search result, and when the advertisement is clicked, the advertiser's Based on products that charge the advertiser for a click bid of 100 yen bid, if the user falls under an additional specified attribute (for example, “male”), an additional rate of 20% is given to the advertiser. It can also be applied to products such as billing.

  In this case, the submission accepting means stores a plurality of the attribute values “real estate” and “male” for one advertisement in the advertisement storage means, and one attribute value “real estate” is bid as an indispensable attribute value. For “male”, which is associated with the unit price “100 yen” and is another attribute value (referred to as “additional attribute value”), an additional rate of 20% when applicable (an additional amount of 20 yen may be used). Are stored in association with each other.

  In this case, the present apparatus 1 operated by the advertising company gives priority to a user who matches the additional attribute value “male” among the users who searched for the keyword “real estate” even if the estimated attribute value. The advertisement will be displayed. Then, when the estimated attribute is used for the additional attribute value “male”, the amount calculation means uses the pattern data (for example, the pattern (ID: “ID” of FIG. 2A) for estimating the estimated attribute value “male”). P51 ”) is obtained by multiplying the corresponding rate“ 80% ”associated with the attribute value“ male ”by the premium rate 20% (or the premium amount), and 16% is obtained as the one attribute value“ real estate ”. The amount of charge is calculated to be 116 yen by adding to the bid price of 100 yen.

  In this way, in search-linked advertising, etc., an additional attribute that adds a surcharge if it is applicable to the required keyword specification is specified, and if the additional attribute is an estimated attribute, the corresponding rate is also multiplied by the corresponding rate Thus, even in a complicated case where an estimated attribute is combined with an essential attribute value, objectively valid advertisement billing can be easily realized.

  In addition, attribute estimation methods are free, for example, a camera provided in so-called digital signage estimates the attributes such as the gender and age of a user nearby, displays advertisements on that signage, and the user has The mobile phone's radio waves are received by a small antenna base station provided in the signage to identify the terminal ID, and an advertisement is displayed on the mobile phone based on the terminal ID, or the user can The case where an advertisement is displayed on a web page based on the terminal ID when the website is accessed is also included in the scope of the present invention. The action history in this case is in contact (access) to the entire information system including predetermined digital signage.

T (T1, T2) Terminal 1 Advertisement billing device (this device)
5 Arithmetic Control Unit 6 Storage Device 7 Communication Unit 8 Submission Accepting Unit 10 Identification Recording Unit 12 Registration Accepting Unit 13 Registration Attribute Storage Unit 15 History Storage Unit 20 Pattern Data Storage Unit 30 Estimation Unit 32 Estimation Attribute Storage Unit 35 Advertisement Storage Unit 40 Advertising distribution means 50 Amount calculation means 55 Advertising fee storage means 60 Claim processing means

Claims (3)

In an advertising billing apparatus that calculates a billing amount of an advertisement distributed to a user terminal accessed by a computer via a communication network,
Identification recording means for identifying a user terminal and storing an action history in access for each user terminal in a predetermined history storage means;
Pattern data storage means for storing in advance a plurality of pattern data in which estimated attribute values and corresponding rates are associated with element values included in the action history or combinations thereof;
Estimating means for estimating an attribute value by comparing the action history corresponding to the identified user terminal with the pattern data;
An ad storage means for storing the contents of each advertisement, the attribute value of the target person as a distribution condition, and the advertising unit price in association with each other;
An advertisement distribution means for retrieving an advertisement corresponding to the estimated attribute value from the advertisement storage means and distributing the advertisement to the user terminal;
For an advertisement distributed based on the estimated attribute value, a charge amount is obtained by multiplying the corresponding rate associated with the attribute value in the pattern data used for estimating the attribute value by the advertising unit price corresponding to the advertisement. An amount calculation means for calculating
An advertising billing apparatus characterized by comprising: The advertisement storage means stores a plurality of the attribute values for one advertisement, and associates an advertising unit price as an indispensable attribute value for one attribute value, and an additional rate when other attribute values are applicable Or, the premium is associated and stored,
In the case where the estimated attribute value is used for the other attribute value, the amount calculation means increases the corresponding rate associated with the attribute value in the pattern data used for the estimation of the attribute value. The advertisement billing apparatus according to claim 1, wherein the advertisement billing device is multiplied by a rate or an additional amount and added to an advertisement unit price of the one attribute value. An ad storage means for storing the contents of each advertisement, the attribute value of the target person as a distribution condition, and the advertising unit price in association with each other;
Pattern data storage means for storing in advance a plurality of pattern data in which estimated attribute values and corresponding rates are associated with element values included in the action history or combinations thereof;
An advertising billing method in an advertising billing apparatus for calculating a billing amount of an advertisement distributed to a user terminal that is accessed via a communication network,
An identification recording process for identifying a user terminal and storing an action history in access for each user terminal in a predetermined history storage unit;
An estimation process for estimating an attribute value by comparing the action history corresponding to the identified user terminal with the pattern data;
An advertisement distribution process for retrieving an advertisement corresponding to the estimated attribute value from the advertisement storage means and distributing it to a user terminal;
For an advertisement distributed based on the estimated attribute value, a charge amount is obtained by multiplying the corresponding rate associated with the attribute value in the pattern data used for estimating the attribute value by the advertising unit price corresponding to the advertisement. Amount calculation processing to calculate
An advertising billing method comprising:

Images