JP2010277472A - Advertising placement volume estimation apparatus and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To estimate an accurate placement volume based on data on the page of advertising insertion, an advertising menu master and others in Internet advertising. <P>SOLUTION: An advertising placement volume estimation apparatus includes a crawling part 101 for crawling through an advertising-inserted URL by reference to an advertising-inserted URL master table to obtain webpage information including a linked URL, an advertisement extraction part 102 for determining whether or not the linked URL matches a domain relevant to advertising by reference to an advertisement determination master table, and extracting advertisement information that is advertising information at the linked URL according to the webpage information, and a placement volume estimation part 103 for selecting an advertising menu matching the advertisement information by matching based on the page of insertion by reference to an advertising menu master table, and estimating a placement volume of the advertisement information from price information in the selected advertising menu and the actual number of days of placement of the advertisement information. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an advertisement amount estimation apparatus and program for estimating the amount of advertisement in an Internet advertisement.

  An advertisement providing method for posting a banner advertisement or the like on an Internet WEB site has been realized. For example, in an advertiser / advertisement agency, an advertisement campaign for sales promotion of goods and services is drafted, and an advertisement creative composed of electronic data such as a copy, a still image banner, and a moving image is produced. When the advertisement creative is submitted to each medium company, it is posted on various media such as a WEB site in the campaign.

  In addition, an Internet placement survey system has already been deployed.

JP 2002-163548 A

  The conventional Internet volume survey system only surveys banner advertisements on the Internet. Internet advertisements other than banner advertisements, such as text advertisements and banner text advertisements, are excluded from the volume estimation. There is a problem that it is not possible to conduct a survey on the amount of traffic.

  SUMMARY OF THE INVENTION An object of the present invention is to provide an advertisement amount estimation device and a program that can accurately estimate the amount of advertisement based on the data of the advertisement insertion surface in the Internet advertisement and the advertisement menu master.

  In order to solve the above-described problems, an advertisement placement amount estimation device according to an aspect of the present invention includes an advertisement placement URL master that holds a list of advertisement placement URLs that are URLs for identifying web sites on which advertisements are placed. Refer to the table, refer to the crawling processing unit that acquires the WEB page information including the link destination URL by performing the crawling process on the advertisement posting URL, and the ad determination master table that holds a list of domains, An ad extraction processing unit that determines whether a link destination URL matches any domain in the list of domains, and extracts ad information that is advertisement information of the link destination URL based on the WEB page information; Referring to the advertisement menu master table that holds at least a list of advertisement posting pages, A selection of an advertisement menu that matches the ad information by a matching process based on a face, and a quantity of advertisement to be estimated based on price information of the selected advertisement menu and an actual number of days of the advertisement information to be published And a processing unit.

  ADVANTAGE OF THE INVENTION According to this invention, the advertisement amount estimation apparatus and program which estimate the amount of advertisement accurately based on the data of the advertisement insertion surface in an internet advertisement, an advertisement menu master, etc. can be provided.

1 is a block diagram illustrating an advertisement placement amount estimation device according to an embodiment of the present invention. A diagram showing advertisements that are subject to estimation by the conventional Internet volume survey system in the classification of advertisements on the Internet The figure which shows the advertisement which the advertisement amount estimation apparatus which concerns on one Embodiment of this invention makes the object of the amount of advertisement estimation in the classification | category of the advertisement on the internet The figure which shows the crawling process part which concerns on the same embodiment The figure which shows the ad extraction process part which concerns on the same embodiment The figure which shows the layout of the ad determination master table which concerns on the same embodiment The flowchart which shows the procedure of the ad extraction process which concerns on the same embodiment The flowchart which shows the procedure of the ad extraction process which concerns on the same embodiment The flowchart which shows the procedure of the ad extraction process which concerns on the same embodiment The figure which shows the layout of the campaign table which concerns on the same embodiment The figure which shows the layout of the ad material master table which concerns on the same embodiment The figure which shows the amount estimation processing part of the amount based on the embodiment The flowchart which shows the procedure of the amount estimation processing of the amount which concerns on the embodiment The flowchart which shows the procedure of the amount estimation processing of the amount which concerns on the embodiment The figure which shows the layout of the ad appearance frequency table classified by URL concerning the embodiment The figure which shows the layout of the network ad master table which concerns on the embodiment The figure which shows the layout of the network menu master table which concerns on the embodiment The figure which shows the layout of the menu information summary work file classified by ad concerning the embodiment The figure which shows the layout of the posting result file according to the placement side concerning the embodiment

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. The main terms used in the description of the embodiment are as follows.

  “Advertising URL” refers to a URL (Uniform Resource Locator) where an Internet advertising placement frame exists. Examples of URLs include “http://www.Acompany.co.jp/” (top page of the Acompany site), “http://travel.Acompany.co.jp/” (Acompany travel site) ) Etc.

  “Link URL” is a link included in html information (WEB page). The html information of the advertisement posting URL includes a link that transitions to content, a link that transitions to external content, a link that transitions to an advertisement campaign, and the like. For example, the link to the content is a link from the Acompanyy site top page to the top page of the travel site of Acompany, and the link to the external content is, for example, from the Acompanyy site news article to the B company site (external content). The link to the link or the advertisement campaign is, for example, a link from the Acompany site top page to a certain automobile company's advertisement campaign page.

“Final URL” refers to the URL of the campaign page that finally arrives from the link destination URL of the advertisement.

“Network menu” refers to an advertisement menu placed in the ad network. The network menu is published for a plurality of media.

“Distribution amount” is the amount of advertisement menu distribution, and is synonymous with the number of impressions (the number of imps).

-“Period guaranteed type menu” refers to a sales form of a menu that guarantees an advertisement placement period. Delivery volume is often not guaranteed.

“Imp guaranteed menu (impression guaranteed)” refers to a sales form of a menu that guarantees the amount of advertisement delivery. The posting period is often not guaranteed.

“Rotation menu” refers to an advertisement menu in which the number of times an advertiser displays is specified and rotation is posted by multiple companies in the same space (same frame). There are both a period guaranteed menu and an imp guaranteed menu.

-“Paste menu” refers to an advertisement menu where advertisers post the same space in one company. It is often a period-guaranteed menu.

  FIG. 1 is a block diagram illustrating an advertisement placement amount estimation device according to an embodiment of the present invention. As shown in the figure, the advertisement amount estimation device 100 is connected to, for example, the Internet N, and investigates the amount of electronic advertisement (Internet advertisement) posted on a website or the like on the Internet N. For example, the crawling processing unit 101, the ad extraction processing unit 102, and the output amount estimation processing unit 103 are the main components. Is output. Note that, in the following embodiment, a case of estimating the performance of posting for each placement surface will be described, but the axis for analyzing the amount of advertisement placement is not limited to this. For example, the present invention can be used for analysis by client, industry, site, product category, and advertisement menu.

  The advertisement placement amount estimation device 100 includes various master tables and files for storing information necessary for processing by each processing unit or storing intermediate information generated by the processing. Specifically, the advertisement posting URL master table 200, the html information 201, the ad determination master table 202, the ad information 203, the non-ad target link table 204, the URL-specific ad appearance count table 205, the ad-specific menu information summary work file 206, And a placement-by-placement result file 207. Also, an advertisement menu master table 300, a network ad master table 301, a network menu master table 302, and the like are provided. These master tables and the like are roughly classified into those that are manually maintained by humans and those that are automatically updated by the system. As for a master table or the like that is manually maintained by a person, the advertisement placement amount estimation device 100 according to the present embodiment reads and uses data when necessary information is prepared by maintenance.

  Further, the advertisement menu master table 300, the network ad master table 301, the network menu master table 302, and the like do not necessarily have to be provided in the advertisement amount estimation device 100 according to the present embodiment, and are different from the present device. Or what is provided in a system etc. is good also as a structure accessed via a network.

  As a rough processing procedure of the advertisement amount estimation device according to the present embodiment configured as described above, (1) First, crawling is performed on the advertisement posting URL on the Internet N, and the html information 201 is acquired. (2) Next, in the html information 201, a link destination that matches the ad determination master table 202 is extracted as ad information, and (3) the acquired ad information 203 and the advertisement menu master table 300 are matched, Is specified, and the amount of money to be put out is estimated and output as a placement result file 207. In addition, since the final URL and the advertisement creative are obtained for each campaign as the ad information 203, a person determines whether or not the advertisement client, the product category, and the house ad (house advertisement) based on the information. Can be attached. Eventually, it will be possible to conduct more detailed and multifaceted analysis of the amount of advertisements posted using this information.

  FIG. 2 is a diagram showing advertisements that are subject to estimation of the amount of advertisements in the conventional Internet advertisement amount survey system in the classification of advertisements on the Internet. Advertising menus placed on the entire Internet include (1) billing based on the posting period (period guarantee) and billing based on impression (imp guarantee), and (2) billing based on performance compensation type (clicks and conversions). It is roughly divided into what to do.

  In addition, (1) menus for billing according to the posting period (period guarantee) / impressions for billing (imp guarantee) are classified into menus placed on a single site and menus placed across multiple sites. .

  As described above, the conventional system only surveys banner advertisements ADp on the Internet, and Internet advertisements other than banner advertisements ADp, such as text advertisements and banner text advertisements, are excluded from the estimation of the output amount. Therefore, there is a problem that the survey of the output amount cannot be performed accurately. Moreover, in the conventional system, the performance-reward type advertisement cannot be excluded. In addition, it does not identify a menu for placing a plurality of sites (network menu), and treats them as a menu for placing a single site. These are factors that significantly reduce the accuracy and reliability of the estimated amount of output.

  On the other hand, as shown in FIG. 3, the advertisement amount estimation device according to the embodiment of the present invention, except for the result reward type menu, (1) charges based on the posting period (period guarantee) and charges based on impression (imp guarantee). The menu to be performed is the target of the estimated amount of output. Further, as the types of advertisements, not only banner advertisements but also banner text advertisements and text advertisements are targeted for estimation of the amount of advertisements. Further, the estimation is performed by determining a menu (network menu) for posting to a plurality of sites.

  Hereinafter, a specific processing procedure performed by the advertisement amount estimation device according to the present embodiment will be described in detail.

[Crawling process]
FIG. 4 is a diagram illustrating the crawling processing unit 101 according to the embodiment. The crawling processing unit 101 does not perform crawling for the entire Internet N, but performs crawling on an advertisement insertion URL where an advertisement is posted, and acquires html information 201. For this reason, the crawling processing unit 101 refers to the advertisement posting URL master table 200 and performs crawling on the URL described in the table. We do not crawl URLs where ads are not posted.

  The html information 201 is, for example, html format electronic data displayed on a medium such as a PC (personal computer), for example, various contents, link URLs to the inside and outside of the site, and information regarding electronic advertisements (link destination URL). Etc. are included in the inside. All such html information 201 including the image file (.swf / .gif / .jpeg) is acquired. Note that the information acquired by the crawling process is not limited to the html format.

  In order to obtain the rotation menu without omission, it is preferable to set the number of crawling in consideration of the number of rotations held in the advertisement menu master table 300. The advertisement menu master table 300 holds information related to the advertisement menu in detail. Specifically, for each advertisement menu, the URL of the placement posting surface (linked to the advertisement posting URL master table 200), size, capacity, and material type ( Flash / GIF / JPEG) and the like are retained. For example, it is information for identifying an advertisement menu such as a banner regular size (350 × 80) and a banner double size (350 × 160).

  For a site that requires login, the advertisement quantity estimation device 100 according to the present embodiment logs into the site and acquires an advertisement. In consideration of the case where advertisement distribution is controlled according to user attributes (for example, different advertisements are distributed by gender and age), it is preferable to log in with a plurality of attributes and acquire advertisements for such sites. .

[Ad extraction processing]
FIG. 5 is a diagram illustrating the ad extraction processing unit 102 according to the embodiment. The ad extraction processing unit 102 refers to the ad determination master table 202 and extracts an ad from the html information 201 acquired by the crawling processing unit 101. “Ad” refers to advertisement information, specifically a banner advertisement and a text advertisement. When an ad is extracted by the ad extraction processing unit 102, an ad ID is assigned to the ad and stored in the storage unit as ad information 203. Further, with regard to the data of the ad ID, the appearance rate information is stored in the URL-specific ad appearance count table 205.

  Since the html information 201 includes information that is not subject to the advertisement, only the advertisement is extracted based on the following two criteria.

(Criteria 1) When at least a part of the link destination URL in the html information 201 matches the domain in the ad determination master table 202, the link destination URL is regarded as an ad. However, ads that match the ad server domain for performance-reward-type advertisements are excluded from the estimated amount of money to be placed.

(Criteria 2) When the link destination URL in the html information 201 is linked to a domain different from the advertisement posting URL, the link destination URL is regarded as an ad. However, when the link destination URL matches the non-add target link table 204, the link is not added.

  The link destination URL that is not recognized as an add is stored in the add-untargeted link table 204. The non-add target link table 204 holds a list of link destination URLs not corresponding to the advertisement.

  With the above mechanism, a link to a news article, a link to a blog, a link to a middle content, etc. can be efficiently eliminated.

  FIG. 6 is a diagram showing a layout of the ad determination master table according to the embodiment. The ad determination master table 202 stores, for example, a domain that is periodically updated by a person and manually updated, and determines whether the ad is an ad or not for each site. The domain of the ad distribution server for each site is regularly checked and mastered. Depending on the site, the ad distribution server may be used only for some advertisements, so an ad server domain is registered for each material type (image or text) for which a link destination URL is set. Whether the link destination URL included in the html information 201 of the advertisement posting URL matches these domains is determined whether it is an ad or not.

  In the ad determination master table 202, when the non-advertisement target flag is “0”, it means a domain regarded as an ad, and when the flag is “1” or “2”, it means a domain regarded as non-ad target.

  When the flag is “3”, it means a domain that is regarded as an ad when the material (link destination content) is text, and when the flag is “4”, it is regarded as an ad when the material is only an image. This means a domain. When the flag is “5”, it means a domain in which both the image and text are regarded as ads.

  In the above, the domain considered as an ad is subject to the estimation of the amount of advertisements. An ad whose material is composed only of text is a text advertisement, an ad whose material is composed only of an image is a banner advertisement, and an ad whose material is composed of text and an image is a banner text advertisement.

  The link table 204 not to be added is a table that stores link data other than advertisements that are automatically updated by the system and associated with the advertisement insertion URL (site ID). Note that link data determined not to be advertised for each crawl is also stored in the non-advertised link table 204.

  7A-7B and FIG. 8 are flowcharts showing the procedure of the ad extraction process according to the embodiment. First, the processing of the following steps S1 to S10 shown in FIGS. 7A and 7B is repeated by sequentially analyzing the link destination URLs acquired from the html information 201 of the advertisement posting URL.

  (Step S <b> 1): Regarding the site ID of the advertisement insertion URL, it is checked whether the ad server domain targeted for text advertisement exists in the ad determination master table 202. That is, the ad determination master table 202 is searched with the site ID of the advertisement posting URL and the common ID (XXXX) of all sites, and there is an advertisement nonexistence flag that is one of “0”, “3”, and “5”. Check if.

  (Step S2): When it exists (corresponds) in the process of Step S1, it is checked whether the link destination URL (the material is text) matches the ad server domain in the forward direction. If there is a forward match, it is picked up as an ad survey target link. If there is no forward match, the link destination URL is stored in the non-added link table 204.

  (Step S <b> 3): If it does not exist (corresponds) in the process of Step S <b> 1, it is checked whether the ad server domain targeted for image advertisement exists in the ad determination master table 202 for the site ID of the advertisement posting URL. That is, the ad determination master table 202 is searched with the site ID of the advertisement posting URL and the all-site common ID (XXXX), and there is an advertisement nonexistence flag that is one of “0”, “4”, and “5”. Check.

  (Step S4): If it exists (corresponds) in the process of Step S3, it is checked whether or not the link destination URL (material is an image) matches the ad server domain in the forward direction. If there is a forward match, it is picked up as an ad survey target link. If there is no forward match, the link destination URL is stored in the non-added link table 204.

  (Step S <b> 5): If not present (corresponding) in the process of Step S <b> 3, it is checked whether or not a non-add target domain exists in the ad determination master table 202 for the site ID of the advertisement posting URL. In other words, the ad determination master table 202 is searched with the site ID of the advertisement posting URL and the common ID (XXXX) of all sites, and it is checked whether there is an advertisement non-target flag of “1” or “2”. .

  (Step S6): When it exists (corresponds) in the process of Step S5, it is checked whether or not the link destination URL matches the non-add target domain. If there is a forward match, the link destination URL is stored in the non-added link table 204. If there is no forward match, the process proceeds to step S7.

  (Step S7): If it does not exist (corresponds) in the process of Step S5, or if the forward match does not occur in the process of Step S6, it is checked whether the link destination URL matches the non-added link table 204. If there is a forward match, the link destination URL is stored in the non-added link table 204. If there is no forward match, the process proceeds to step S8.

  (Step S8): If it does not exist (corresponds) in the process of Step S7, it is checked whether the final URL reached by following the link destination URL is different from the domain of the advertisement posting URL. If it is in a different domain, it will be picked up as a link to be added. If it is not a different domain, it is stored in the non-target link table 204.

  (Step S9): With respect to the add target link extracted in the processes of Steps S1 to S8, the final destination URL is acquired by following the link destination URL, and the campaign information is acquired. The campaign information is obtained by analyzing the html information of the final arrival URL. A campaign ID is issued so as to be unique for each final URL, and is stored in the campaign table (ad information 203).

  FIG. 9 is a diagram showing a layout of the campaign table according to the embodiment. The campaign table is a table that stores advertisement campaign information, and stores the html information of the final arrival URL with the campaign ID as a key item (system automatic update). In FIG. 9 and subsequent figures, an asterisk means that the item is a key item.

  (Step S10): The advertisement material (image / text) associated with the link destination URL is stored in the ad material master table (ad information 203) regarding the link to be added extracted in the processes of steps S1 to S8. At this time, an ad ID that is unique between the campaign ID and the binary hash value of the ad material is issued.

  FIG. 10 is a view showing a layout of the ad material master table according to the embodiment. The ad material master table is a table that stores information on advertisement material recognized as an ad with the ad ID as a key item (system automatic update).

[Transfer amount estimation processing]
FIG. 11 is a diagram showing the submitted amount estimation processing unit 103 according to the embodiment. The submitted amount estimation processing unit 103 includes a menu candidate extraction unit 105, a menu candidate selection unit 106, a price selection unit 107, and a campaign unit recounting unit 108.

<Menu candidate extraction>
The menu candidate extraction unit 105 matches the advertisement acquired by crawling with the advertisement menu held in the advertisement menu master table 300, and extracts an advertisement menu that is an ad candidate acquired by crawling. The ad acquired by crawling is information held as the ad information 203 shown in FIG. As keys for matching, items such as a placement page, an ad size (width, height), a file capacity, and a material are used. Information regarding the extracted advertisement menu candidates is held in the menu information summary work file 206 for each ad.

<Selecting menu candidates>
The menu candidate selection unit 106 selects an advertisement menu from the menu candidates extracted by the menu candidate extraction unit 105 and held in the ad-specific menu information summary work file 206 according to an appropriate rule. In selecting a menu candidate, as will be described later, a banner text advertisement, a plurality of line ads, and a network menu are identified and processed.

<Price selection>
The price selection unit 107 estimates the advertisement placement amount for each placement based on the price of the advertisement menu selected by the menu candidate selection unit 106 and the amount of placement. The estimated amount of advertisements to be estimated is, for example, the amount of advertisements and the number of impressions to be posted.

  In the case of a period guaranteed menu, the price selection unit 107 estimates the price and amount of advertisement menu based on the posting period, and in the case of an impression guaranteed menu, the price selection unit 107 estimates the price and amount of advertisement menu based on the appearance rate.

<Recalculation of campaign unit>
Multiple materials may rotate within one advertisement menu due to material replacement. In order to cope with this case, the campaign unit recalculation unit 108 recalculates the estimation results for the same campaign, the same period, and the same menu. For example, if there are three ads that are regarded as the same menu within the same campaign and the same period, the amount of money and the number of imps that are posted on the menu are divided by “3” as the number of ads.

  Hereinafter, the procedure for the estimated amount of output according to the embodiment will be described in detail with reference to FIGS. 12A-12B.

<Menu candidate extraction unit 105>
(Step S1): The menu candidate extraction unit 105 reads the URL-specific ad appearance count table 205 in units of ad, and checks whether or not it has been put on a plurality of sites. FIG. 13 shows the layout of the URL-specific ad appearance count table 205. The URL-specific ad appearance count table 205 is a table that indicates how often an ad has appeared with respect to a crawled URL with the site ID and the URLID as key items. When the menu candidate extraction unit 105 reads the same ad ID from the URL-specific ad appearance count table 205, if a plurality of different site IDs are obtained, the ad corresponds to the advertisement network menu. Is determined.

  (Step S2): When the processing in Step S1 is applicable, the material storage destination held in the ad information is matched with the material storage destination domain information in the network ad master table 301. If there is a forward match, it is determined that the menu is a network menu, and the amount of money to be posted and the amount to be posted imp of the network menu master are assigned. After the assignment, the process proceeds to step S6, and when there is no coincidence, the process proceeds to step S3. The layout of the network ad master table 301 is shown in FIG. The network ad master table 301 is a table that stores ad network ID and ad network serial number as key items, and stores the storage location of the ad material in the network menu, and is checked and updated by a person as needed. FIG. 15 shows the layout of the network menu master table 302. The network menu master table 302 is a table that stores the charge and the amount of advertisement menu for each ad network using the ad network menu ID as a key item. The network menu master table 302 is automatically updated based on the information in the network fork ad master table 301 and the advertisement menu master table 300. In this case, banner advertisements and text advertisements of each site where the ad network menu is published are averaged to calculate the advertisement menu charge and amount.

  (Step S3): If not applicable in the process of Step S1, ads that have been published in the same campaign, the same placement surface, and the same placement period are grouped in order to match the advertisement menu master table 300. This makes it possible to recognize “banner text ads” and “multiple line ads”, which are advertisement menus in which a plurality of creatives are put out as a set. For example, if a banner and text advertisement of “160 × 100” are put on 3 / 2-3 / 9 on the top page of the company A site, these two ads are grouped.

  (Step S4): The ad information 203 and the advertisement menu master table 300 are matched with the matching key, and a plurality of advertisement menu candidates are extracted. In the case of a banner text advertisement, matching is performed using the size of the banner advertisement. The result is stored in the ad-specific menu information summary work file 206. The matching key includes a placement / publishing side of the advertisement menu master table 300, an actual placement / posting side of the ad information 203, a banner size, a material type (Flash / GIF / JPEG), and a file capacity.

  The layout of the menu information summary work file 206 by ad is shown in FIG. The ad-specific menu information summary work file 206 is a table in which an ad ID, a site code, and a menu code are key items, and a plurality of menu candidates are associated with the ad identified by the ad ID and stored.

<Selecting menu candidates>
(Step S5): The menu candidate selection unit 106 selects an appropriate advertisement menu from the menu candidates extracted by the menu candidate extraction unit 105 according to a rule. For example, the priority is determined from the following viewpoints and the advertisement menu is selected. In other words, the degree of match between the placement posting surface of the advertisement menu master table 300 and the actual placement posting surface of the ad information 203, presence / absence of fee information, presence / absence of publication amount information, agency limited, not an advertiser specific menu, Use conditions such as large or large amount of publication as a rule. Furthermore, the period guarantee may be given priority over the imp guarantee.

  If there is only one menu candidate, the advertisement menu is selected.

<Price selection>
(Step S6): The price selection unit 107 estimates the advertising amount and the posting imp based on the price information and the amount of advertisement held in the advertisement menu selected in Step S5. Here, it is checked whether the guaranteed form of the selected advertisement menu is a period guarantee or an imp guarantee. If it is a period guarantee menu, the process proceeds to step S7. If it is an imp guarantee menu, the process proceeds to step S8. Move.

  (Step S7): The price information is specified (estimated) by comparing the result determined in step S6 as the period guaranteed menu, the advertisement posting period, and the posting period of the advertisement menu master table 300.

  The procedure for estimating the assumed period of the menu is, for example, as follows.

・ Ad placement period ≤ 7 days
A one-week menu in the advertisement menu master table 300 is adopted. When there is no corresponding advertisement menu, the default period of the advertisement menu is adopted.

Ad placement period> 7 days and ad placement period ≦ 14 days
A two-week menu in the advertisement menu master table 300 is adopted. When there is no corresponding advertisement menu, the default period of the advertisement menu is adopted.

・ Ad placement period> 14 days and ad placement period ≦ 21 days
A three-week menu in the advertisement menu master table 300 is adopted. If there is no applicable advertisement menu, the default period of the advertisement menu is adopted. Ad advertisement period> 21 days and ad advertisement period ≦ 28 days
The 4-week menu in the advertisement menu master table 300 is adopted. When there is no corresponding advertisement menu, the default period of the advertisement menu is adopted.

・ Ad placement period> 28 days and ad placement period ≦ 58 days
The menu for one month in the advertisement menu master table 300 is adopted. When there is no corresponding advertisement menu, the default period of the advertisement menu is adopted.

・ Ad placement period> 58 days and ad placement period ≦ 84 days
A menu for two months in the advertisement menu master table 300 is adopted. When there is no corresponding advertisement menu, the default period of the advertisement menu is adopted.

・ Ad placement period> 84 days and ad placement period ≦ 168 days
The menu for three months in the advertisement menu master table 300 is adopted. When there is no corresponding advertisement menu, the default period of the advertisement menu is adopted.

・ Ad placement period> 168 days and ad placement period ≦ 320 days
The menu for 6 months in the advertisement menu master table 300 is adopted. When there is no corresponding advertisement menu, the default period of the advertisement menu is adopted.

・ Ad placement period> 320 days
A one-year menu in the advertisement menu master table 300 is employed. When there is no corresponding advertisement menu, the default period of the advertisement menu is adopted.

Therefore, for example, in the case where the ad posting period is 3 / 2-3 / 25 (24 days) and the price master is as shown in Table 1 below, the default period is 1 week, 1 million yen, 1 million imp. Is adopted.

  However, there is always one default period on the price master.

  (Step S8): When it is determined that the menu is the imp guaranteed menu in the process of Step S6, it is estimated how much the advertisement has been published from the appearance rate of the ad information. The appearance rate of the ad information can be obtained from the number of appearances (the number of times the ad has appeared) and the number of accesses (the total number of times the crawl URL has been crawled) in the URL-specific ad appearance number table 205 shown in FIG. it can.

  When there are n price masters for each number of guaranteed imps, the appearance rate is allocated from the number of guaranteed imps that is smaller by (100 ÷ n)%. Since the ad appearance rate is stored in the URL-specific ad appearance count table 205, it is used.

For example, if the ad appearance rate is 26.7% and the price master is as shown in Table 2 below, there are four price masters for each guaranteed imp number, so the appearance rate is distributed by 25% from the smaller guaranteed imp number. To do. An ad with an appearance rate of 26.7% is considered to have been delivered 2 million imp. That is, No. 2 in Table 2. Adopt price information of 2.

  (Step S9): By the processing up to Step S7 or Step S8, the advertisement menu and its “price”, “publishing amount (number of imps)”, and “publishing period” are specified. From these, the amount of money to be posted per day and the number of imps to be posted per day are calculated. For example, the amount paid per day = the price of the advertisement menu / the posting period of the advertisement menu, and the number of postings per day = the assumed / guaranteed posting amount of the advertisement menu / the posting period of the advertisement menu.

  Next, by multiplying the actual number of days that the ad information has, the amount of money to be posted and the number of impressions to be posted are estimated. For example, the amount of ad placement = the amount of money posted per day × the number of days of posting ad information, and the number of ad posting imps = the number of postings per day × the number of days of posting ad information. This estimation result is stored for each page in the page-by-page result file 207.

  It should be noted that the actual number of days that the ad information has can be determined based on the ad information 203.

  FIG. 17 is a diagram showing the layout of the placement-by-placement posting result file 207. The placement result file 207 by placement surface stores placement site code, placement surface number, ad ID, and date of appearance as key items, and stores the amount of money to be placed and the placement imp for each placement surface, placement surface, ad unit, and appearance date. It is a table.

<Recalculation of campaign unit>
(Step S10): The campaign unit recounting unit 108 checks whether there is an ad that has the same campaign, the same posting period, and the same menu. If it exists, the process proceeds to step S11.

  (Step S11): If there is a corresponding ad in the process of Step S10, it is regarded as an ad whose material has been replaced, and the amount of money to be submitted, the number of ads included in the same campaign, the same posting period, and the same menu, Rebate the submission imp. As a result, compression is performed in units of campaigns.

For example, in the case of Table 3, there are four ad IDs in the same campaign, the same posting period, and the same menu. These are regarded as four replacement materials being rotated, and the amount of money and the amount imp of the four ads are each multiplied by ¼. As a result, as shown in Table 4, the total value of the ads matches one menu amount.

  According to the embodiment of the present invention described above, the following operational effects can be obtained. First, regarding the acquisition of advertisement information used for estimating the amount of advertisements, in the crawl process, all link destination URLs included in the acquired html information 201 are frequently acquired (for example, every day) and matched with the advertisement menu master table 300. Thus, the latest URL where the advertisement is put out can be obtained. In this way, it is preferable to crawl the posting surface (“news surface, auction surface, etc.”) that frequently generates a new URL without omission.

  And according to this embodiment using the ad determination master table 202 representing the domain of the ad server, the advertisement information can be efficiently acquired from the html information 201 acquired by the crawling process. In particular, a text advertisement can be determined when the advertisement information is acquired. For example, for a link included in the html information 201 acquired by the crawl process, it is possible to determine whether the linked content is a link text to another content, a middle surface, or a text advertisement.

  In addition, it is possible to determine performance-reward-type advertisements (listing advertisements, affiliate advertisements, content match advertisements, interest match advertisements, etc.), and exclude them from the target for estimating the amount of advertisements. In this case, the ad server domain for the performance-reward-type advertisement is managed in the ad determination master table 202, and the ad that matches the ad-server domain for the performance-reward-type advertisement may be excluded from the estimated amount of submission money. There are banner advertisements, text advertisements, etc. in performance-reward-type advertisements. To accurately grasp the amount of money placed in the advertisement menu (period guarantee / imp guarantee), it is precisely which ad is a result-reward-type advertisement. It is necessary to judge. Incidentally, there is a click charge for performance-rewarded advertising, and it is necessary to estimate the number of clicks by predicting the number of clicks.

  Next, in the advertisement menu determination and the amount of money to be submitted / improvement of the impression imp, information related to the advertisement menu is held in detail as the advertisement menu master table 300. Specifically, the advertisement posting surface (URL), size, and capacity for each advertisement menu are stored. The material type (Flash / GIF / JPEG) is stored. By using such an advertisement menu master table 300, it is possible to comprehensively determine which advertisement menu the ad acquired by crawling corresponds to. Further, it is possible to determine not only the advertisement menu placed on a single placement surface (for example, the advertisement menu placed on the top page) but also the advertisement menu placed on a plurality of placement surfaces. Also, according to the difference in delivery capacity (creative capacity), the difference in the menu price is reflected in the draft estimation (the menu price will be higher if the delivery is large capacity). In addition, “banner text advertisement” and “multiple line ad” which are advertisement menus that are a set of a plurality of creatives can be determined as one advertisement menu. The content ratio of these advertisement menus is high in the Internet advertisement, and is very important for accurately estimating the amount of money to be posted and the price to be posted. In addition, the network menu can be determined. That is, regarding a network-type menu placed across a plurality of sites, it is possible to determine which network menu is based on the storage location of the advertisement material.

  Therefore, according to the embodiment of the present invention, it is possible to provide an advertisement output amount estimation device and a program that accurately estimate the amount of advertisement based on the data of the advertisement insertion surface in the Internet advertisement and the advertisement menu master.

  Note that the present invention may be implemented as a program for causing a general-purpose computer to function as the above-described placement performance estimation apparatus. A computer-readable recording medium in which the program is recorded may be read from the recording medium reading device and executed by the CPU.

  Further, the present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

N ... Internet;
100 ... Advertising amount estimation device;
101 ... Crawling processing unit;
102: Ad extraction processing unit;
103 ... Amount estimation processing unit;
200 ... Advertising URL master table;
201 ... html information;
202 ... Ad determination master table;
203 ... Ad information (campaign table, ad material master table);
204 ... Link table not to be added;
205 ... Ad appearance frequency table by URL;
206 ... Menu information summary work file by ad;
207… Publication result file by page

Claims (6)

By referring to the advertisement posting URL master table that holds a list of advertisement posting URLs that are URLs for identifying the WEB sites on which advertisements are posted, a crawling process is performed on the advertisement posting URLs, so that the link destination URL is changed. Crawling processing unit for acquiring WEB page information including;
Refer to an ad determination master table holding a list of domains, determine whether the link destination URL matches any domain in the list of domains, and based on the WEB page information, the link destination URL An ad extraction processing unit for extracting ad information which is advertisement information of
Refer to an advertisement menu master table that holds at least a list of advertisement posting pages, select an advertisement menu that matches the advertisement information by matching processing based on the advertisement posting surface, and select price information of the selected advertisement menu And an advertisement amount estimation processing unit that estimates the amount of advertisement information to be posted based on the actual number of advertisement information days. A non-added link table that holds a list of link destination URLs not corresponding to the advertisement;
The ad extraction processing unit is configured such that a link destination URL included in the WEB page information is a link to a domain different from a domain indicated by the advertisement insertion URL in the advertisement insertion URL master table, and the link that is not an add target The advertisement amount estimation device according to claim 1, wherein the advertisement information is extracted when the advertisement information is not included in the table. The ad determination master table is
The domain that corresponds to the ad for which you want to estimate
Domains that correspond to ads that are not subject to trafficking estimates,
Domains that correspond to text ads consisting only of text,
Domains corresponding to banner ads consisting only of images,
3. The advertisement amount estimation device according to claim 1, further comprising flag information for identifying a domain corresponding to a banner text advertisement composed of text and an image.   2. The output amount estimation processing unit groups the ad information for those having the same placement surface, and selects an advertisement menu that matches the grouped ad information in the matching processing. 4. The advertisement placement amount estimation device according to any one of 1 to 3. For each site ID for identifying the WEB site, the total number of crawls performed on the WEB site, the ad ID for identifying the ad information, and the number of ad appearances by URL that holds the number of appearances of the ad information Table,
A network menu master table for holding material information and price information for each network menu;
The submission amount estimation processing unit
It is determined whether the ad information extracted by the ad extraction processing unit is a network menu placed across a plurality of sites based on the URL-specific ad appearance count table,
If the network menu is determined, the corresponding network menu is obtained by matching the material information associated with the ad information with the material information of the network menu master table, and the price information of the network menu is used. The advertisement amount estimation apparatus according to claim 1, wherein the amount of advertisement is estimated. Computer
By referring to the advertisement posting URL master table that holds a list of advertisement posting URLs that are URLs for identifying the WEB sites on which advertisements are posted, a crawling process is performed on the advertisement posting URLs, so that the link destination URL is changed. Crawling processing means for acquiring WEB page information including,
Refer to an ad determination master table holding a list of domains, determine whether the link destination URL matches any domain in the list of domains, and based on the WEB page information, the link destination URL Ad extraction processing means for extracting ad information which is advertisement information of
Refer to an advertisement menu master table that holds at least a list of advertisement posting pages, select an advertisement menu that matches the advertisement information by matching processing based on the advertisement posting surface, and select price information of the selected advertisement menu A program for functioning as an output amount estimation means for estimating the output amount of the ad information based on the actual number of days of posting of the ad information.

Images