JP2004504674A - System and method for selecting another advertising inventory instead of a sold-out advertising inventory - Google Patents

Abstract

No summary

Description

[0001]
[Prior art]
The use of the Internet by the general public is definitely prevalent. More and more people are accessing the Internet and obtaining the vast amount of information that the Internet provides. With the proliferation of Internet users, advertising on the Internet has become an important priority for many advertisers.
[0002]
This results in a significant amount of revenue from displaying the advertising banner of the advertiser on the displayed website or web page for web portals and ISPs. For example, 1 to 100,000 users visit Yahoo a day! For such prominent portals, displaying advertiser ads on websites or web pages can generate significant revenue.
[0003]
Advertisers typically pay a fee for each advertisement viewed by a web user. Contracts for displaying advertisements are usually made weeks / months before the advertisement is delivered. The contract period is between one day and several years. There are typically various types of contracts, including regular contracts, proprietary contracts, and unlimited contracts. For a regular contract, the advertiser purchases a predetermined number of ad views in the selected area (web page). For exclusive contracts, the advertiser purchases all views in the selected area. For unlimited contracts, the advertiser purchases all remaining advertising views in this area after all other regular contracts associated with the selected area have been concluded.
[0004]
Ideally, ads should be strategically located, both physically and temporally, to maximize the effectiveness of the ad and to be as close to the target audience as possible. There is. Therefore, it is important to identify strategic locations when planning your advertising efforts. Determining strategic locations for advertising purposes often uses artificial statistics. It is important to bring demographic data to a system-wide basis for a web portal or ISP that provides a large number of web sites or web pages. So Yahoo! It is desirable to develop methods and systems that can derive demographic data from Internet user behavior on a collective basis for web portals such as.
[0005]
In addition, the use of aggregate demographic data to decompose and use specific websites based on demographic groups or other characteristics, which may be tailored to the specific purpose of a particular ad, Or it would be useful to be able to identify a web page.
[0006]
In addition, advertisements on the Internet, like advertisements made via more traditional media such as television or print, are also subject to physical limitations. For obvious reasons, it is natural and often the best choice for an advertiser to request an advertisement view on a web portal or ISP's homepage. However, due to the limited physical area of a web page, the demand for an advertising area or view on a popular web page often far exceeds supply. Thus, a significant amount of advertising is not always placed on the most favorable web pages.
[0007]
To maximize profits, ideally, excess demand should be diverted to other advertising areas or views available on other web pages. However, before convincing advertisers to place ads on these other web pages, they justify their confidence that their ads will produce the same benefit when placed on those web pages. Need to be done. Thus, a method and system capable of providing demographic data derived from the behavior of Internet users associated with web pages to direct excess advertising demand to other equally effective web pages It is desirable to develop.
[0008]
Yahoo! as background technology! A web page is usually configured in a tree structure. Figure 1 is Yahoo! FIG. 2 is a simplified network diagram showing the structure of a network. The top of this tree is Yahoo! The whole network. Below this section, Yahoo! Shopping, Yahoo! Sports, Yahoo! Yellow page, Yahoo! There are various clauses such as search. Under each of these clauses are various derived clauses, and each of these derived clauses may further have various derived clauses. For example, Yahoo! Below the sports section, there are sections such as NFL, NHL, NBA, etc., and below the NBA section there are sections such as standing, statistics, games, and the like. The search result page is also included as a part of the tree. For example, Yahoo! Below the search section, Yahoo! There are all results pages from the search terms entered on the front page.
[0009]
Summary of the Invention
The present invention relates to systems and methods for collecting and extracting Internet user behavior data. Furthermore, the present invention allows for the collection and withdrawal of historical and demographic data based on Internet user's activity to select an alternative advertising inventory instead of an out-of-stock advertising inventory. Is what you do.
[0010]
In an exemplary embodiment, the present invention has multiple ad (ad) records. An ad record is generated for each ad that appears on the viewed web page. Preferably, the additional record generation routine first generates one or more additional ad records based on one or more original ad records. This is done to allow the calculation of historical and demographic data for the tree of web pages (all nodes are derived from a single node), as well as for the entire web site.
[0011]
The filter routine then processes all ad records to generate a first record group and a second record group. The first record group contains only registered user records, and the second record group contains records for all registered and unregistered users. For registered users, the record includes P cookies, L cookies, B cookies, and space IDs, while for unregistered users, the record does not include P cookies or L cookies, Only space ID is included. The first group of records is created by examining whether a record has a P cookie and / or an L cookie, each record also including an associated B cookie and a space ID. The second record group is generated by extracting the related B cookie and space ID from each user's (registered or unregistered) record. These two record groups are thus divided into a first group containing only registered user records, each record having an additional P and L cookie, and a group of all users. It is a second group including records about (registered and unregistered).
[0012]
The B cookie provides identification information about the user's particular browser. The P cookie provides demographic information about the registered user, such as age, gender, occupation, and the like. The L cookie provides the username of the registered user. Finally, the space ID provides identification information about the web page (for which a record is generated).
[0013]
Each record group is then processed differently. For a group of registered user records, the first sorting routine sorts (sorts, sorts) these records based on L cookies, ie by user name. The index routine then generates an index using the Pcookie for each of these records. Since the P cookie contains demographic information, the generated index essentially represents a demographic distribution. In a preferred embodiment, the index is represented by a bitmap, where each bit of the bitmap represents a demographic characteristic.
[0014]
The second sort routine then sorts the first record group based on the space ID. By sorting the space IDs, the records are then grouped by web page. Thus, demographic information from records related to the same web page can be obtained.
[0015]
The third sorting routine also sorts the first group of records based on the index associated with these records. Since these indices represent a demographic distribution, records having the same demographic distribution are then grouped together. The collective space ID information can then be examined from these grouped records to derive information about the web pages most frequently viewed by users with the same demographic distribution. This information can then be used to identify alternative advertising inventory among multiple web pages.
[0016]
The second group of records for all users is handled differently. The first terry (record, calculate) routine first calculates each total number of visits to each of a number of web pages identified from the space ID of the second record group.
[0017]
The second terry routine then determines the number of affinity relations. One similarity relationship is determined for each pair of different web pages among the web pages identified from the space ID of the second record group. This second terry routine also maintains a running total for each of the determined similarities. By using the total number of each visit and each cumulative total, information about the likelihood that the user visits one web page and also visits other web pages can be determined. This information can be used to identify alternative advertising inventory as well.
[0018]
Reference to the remaining portions of the specification, including the drawings and claims, will realize other features and advantages of the present invention. Other features and advantages of the present invention, as well as the structure and operation of various embodiments of the present invention, are described in detail below with reference to the accompanying drawings, in which like reference numbers indicate the same or functionally similar elements. It is described in.
[0019]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, the present invention will be described. 2A-E are simplified flow diagrams illustrating the operation of one embodiment of the present invention. Referring to FIG. 2A, advertisement log 20 is a collective inventory that provides a record of all advertisements displayed to a user during a predetermined time interval. The predetermined time interval can vary based on several factors, such as processing needs, system storage constraints, and the like. Each time a user views a particular home page, information about all advertisements appearing on that particular web page is recorded in the advertisement log 20. Preferably, an advertisement record is generated for each advertisement displayed on each web page viewed. For example, if the viewed web page displays ten advertisements, one advertisement record is generated in the advertisement log 20 for each advertisement. In the exemplary embodiment, each entry or advertisement record in the advertisement log 20 includes various information about the user who saw the advertisement, including, for example, B cookies, P cookies, L cookies, and space ID.
[0020]
B cookies contain identification information about the browser used by a particular user. Such information typically includes, for example, a serial number assigned to the browser.
[0021]
The P cookie contains certain demographic information about the user, such as gender, birthday, zip code, country, occupation, industry, and interest. The information represented by the P cookie is generally obtained from the user when the user registers with a web portal or ISP. Therefore, P cookies are typically applicable only to registered users. Similarly, the L cookie contains the registered user's username.
[0022]
Finally, the space ID includes identification information indicating the particular web page visited by the user. In other words, the space ID gives information about what page was viewed by the user.
[0023]
In the exemplary embodiment, additional record generation process 22 examines each log record in advertisement log 20 to determine whether additional advertisement records need to be generated. This is done to calculate a tree of web pages and / or data for all web sites to reflect a more complete distribution of demographic data. As mentioned above, Yahoo! Some web portals organize their web pages into a tree structure with nodes. In a preferred embodiment, for each advertisement record for a particular web page (or node), the same record with the appropriate space ID for each node that is an ancestor of that particular node will have such a record. Unless the same record already exists in the advertising log, it will be generated. In other words, additional advertising records can be generated for each node above that particular node.
[0024]
The above can be explained, for example, as follows. As shown in FIG. 1, when a user visits the home page for San Francisco Forty Nine Nations 16 directly, that is, arrives at the web page by entering the appropriate URL, the web page is displayed. An ad record for the ad displayed above is generated in the ad log. In addition, for each of these advertising records, the node above it, NFL web page 14, Yahoo! Sports web page 12, and Yahoo! Additional advertising records are also generated for each of the network web pages 10, respectively. Since each of these web pages is separate from each other, their respective advertising records necessarily reflect the space ID corresponding to themselves, but otherwise have the same demographics as the starting record. Information. However, if a user enters Yahoo! on the San Francisco Forty-Niners Web page 16! If arriving from the network page 10 via a less direct route of successively clicking on the appropriate hyperlink start, no additional advertising records will be generated. This is a good record for Yahoo! This is because the user visits each web page starting from the network page 10 and has already been added to the advertisement log. The newly created additional advertisement record is added to the departure advertisement record. By adding these additional advertising records, information about the tree of web pages and the entire web site can be collected and made available for subsequent use.
[0025]
All records are then filtered by a filter process 14 that uses different criteria to generate two groups of records. More specifically, a first group of records is created for registered users, and a second group of records is created for all users who are registered or not. Since only the registered user's record contains the L cookie, this is achieved by checking whether the L cookie is present in the record. Alternatively, the P cookie can be used to generate a first group of records. Thus, after the filter process 24, the respective records for the two groups of users, registered users and all users, are identified and grouped together.
[0026]
In addition to identifying these two groups of records, the records for the two groups are further stored based on L cookies and B cookies, respectively. Referring to FIGS. 2B and 2E, the two groups of records are stored separately, for example, in databases 26,40. As described below, the records of these two groups undergo different processing.
[0027]
The processing of the first group of records, the records of registered users, is described below. Registered user records are sorted (sorted) based on L cookies. As mentioned above, the L cookie contains information about the registered user's username. Thus, by classifying L cookies, all of the records and information relating to a particular registered user are grouped together.
[0028]
Referring to FIG. After the registered user's records have been categorized, each of these records, particularly the P cookie, is then processed by the indexing process 28 to generate an index for the particular record. The index is a bitmap representing the demographic data in each record. Each bit in the bitmap represents a particular demographic feature. For example, the first bit of the bitmap may represent the gender of the user, eg, male. Similarly, other bits can be used to represent demographic features such as ages 30-39, over $ 100,000 in annual income, engineering jobs, and the like. By selecting the values "1" and "0" for each bit, each record indicates whether a particular demographic is present or absent.
[0029]
In another embodiment, additional bits in the bitmap are generated based on web pages visited (obtained via the space ID of each record) by the user. These bits reflect the user's history and interest that add value to the user's demographic characteristics.
[0030]
In this regard, by generating corresponding indices for these records, different processes can be used to process these records and the associated indices. For example, records can be categorized either by space ID or bitmap, and then processed as described further below.
[0031]
In an operation 30 shown in FIG. 2C, according to an exemplary embodiment, these records are categorized by space ID. As mentioned above, the space ID provides identification information used to indicate the particular web page visited by the user. By classifying space IDs, all records originating from the same web page are grouped together. In particular, this means that all demographic information relating to each web page has been collected and is available for further processing.
[0032]
Once classified by space ID, the record is then summarized at 32. During this summarization process 32, information from the records is collected and stored. More specifically, for each space ID, the corresponding bitmap or index information for all records having the space ID is currently available and stored in database 34 for future access. In other words, demographic information for all users who have visited a particular web page is accessible from database 34.
[0033]
Further, as part of the summarization process 32 for each space ID, a bit count is calculated for each bit of the bitmap. By calculating the bit count for each bit, the total number of users with a particular demographic characteristic who have visited a particular web page is determined. For example, the total number of men who saw a particular web page is determined.
[0034]
Additional demographic assessments within the identified demographic groups can be further achieved by using bitmaps. Continuing with the embodiment described above, using the record identified as having male users, the bits representing between ages 20-29 are further selected to determine the total number of males in a particular age group. I can do it. Additional subdivisions can likewise be achieved by using additional bits of the bitmap. Here, various demographic features can be determined for a particular web page by examining individual bits in the bitmap, or a combination thereof.
[0035]
In another process 36 shown in FIG. 2D, records for registered users are sorted by bitmap, according to an exemplary embodiment. By classifying records based on the bitmap, records having the same bitmap are grouped together. Since each bit of the bitmap represents a demographic feature and the bitmap represents a collection of demographic features (profiles), users with the same set of demographic features are then grouped together. Divided.
[0036]
Additional information can be obtained from records grouped based on a collection of demographic features. This information is then formatted and stored in database 38 for later processing. For example, for each group of records having a particular set of demographic characteristics, space ID information is derived from each record in the group. By examining the collection space ID information, the web pages can identify web pages that are most frequently visited by users having a particular set of demographic characteristics. Thus, information on web pages often visited by each demographic group is available. By utilizing the information in this manner, advertisers can be properly advised regarding the placement of their advertisements. Ads can be strategically adapted and placed to maximize the presentation and efficiency of the ads with respect to the target demographic group.
[0037]
The processing of the second group of records, ie the records of all users, will now be described. The records for all users referring to FIG. 2E are categorized based on B cookies and stored in database 40. As mentioned above, B cookies can be used to identify a particular browser of a user. As a result, by classifying the B cookies, all behavior data originating from the same browser is grouped together. In certain circumstances, such as when the Internet is available and computers are generally accessible to the public, or such computers are used by various family members, multiple users may use the same browser at different times. It will be appreciated that using and may produce results that represent many users. From a statistical point of view, behavior data collected from a browser remains available as long as the same browser is used consistently by the same general group of users.
[0038]
Records for all users are handled differently. Note that the processing of the first group of records and the processing of the second group of records are independent of each other. As mentioned above, the records of all users are sorted by the B cookie. Sorting by B cookies is preferred because unregistered users do not have pre-given registration or demographic information. Therefore, records for unregistered users do not include any P cookies. However, these records include Space IDs. This is because spatial IDs information is captured during a browsing session, regardless of the status of the user.
[0039]
The sorted records of all users go through an affinity and tallying process 42. More specifically, for each B cookie, ie, browser, a cumulative sum is tallyed and kept for each individual unique spatial ID. That is, an individual tally of each web page visited by the browser is maintained. Collective, cumulative sums for various unique spatial IDs from all browsers are then combined and stored for future use.
[0040]
In addition, various types of information are segmented using the collective spatial ID information collected from each browser. For example, affinity relationships are determined for each set of different spatial IDs, and the corresponding affinity count for that set is incremented. By determining an affinity relationship among all web pages visited by the same browser, a user (or multiple users) uses that browser with respect to viewing one web page in relation to other web pages Information may be obtained to predict user trends and usage behavior. Information about affinity relationships in web pages visited by all browsers is then collected. It should be understood that affinity relationships can be created to associate more than one set of different web pages. Such a relationship may include, for example, three or more different web pages. Those skilled in the art will know how to achieve such an affinity relationship.
[0041]
The above description can be explained by way of examples. Assume that web pages A, B and C are visited by a particular browser. The corresponding cumulative total for tracking the number of visits to web pages A, B and C is incremented by one. Further, since the browser has visited three web pages A, B and C, affinity relationships AB, AC and BC are created and the corresponding affinity count is incremented.
[0042]
After all user records have been processed, there is a cumulative total for each web page that has been visited collectively by the browser, there is an affinity count for each set of different web pages, 44. By using collective sums and affinity counts, advertisers are better advised on the placement of their ads.
[0043]
Using the above example, after processing the records for all users, it was determined that the collective sums for web pages A, B, and C were 100, 200, and 300, respectively, and the affinity relationships AB, A- Assume that the affinity counts for C and BC are determined to be 50, 100 and 150, respectively. Using the above figures, the percentage of users who also visit web page B visits web page A is 50%, and the percentage of users who also visit web page A visits web page B is 25%. Something is further calculated. This percentage is calculated as follows: Since the collective sum for a web page is 100 and the affinity count for AB is 50, this means that 100 users visited page A, and only 50 of the 100 users who visited page A had a web page B also visits, thus implying a 50% numerical result. Similarly, since the collective total for web page B is 200 and the affinity count for AB is 50, only 50 out of 200 users visiting web page B see that web page A also visited. Means 25% numerical result.
[0044]
Thus, according to these three percentages, advertisers can be properly advised of options if their initial choice is sold out. For example, using the 50% visit figure described above, the advertiser may find that if all of the ad views on web page A were sold out, the ads intended for viewers viewing web page A would If the ad were placed on web page B instead, it would know that it had a 50% probability of being viewed by the same audience. This is because 50% of the users who watch web page A also usually visit web page B. Similarly, using a 25% number, an advertisement intended for a viewer viewing web page B may only be viewed 25% by the same viewer if the advertisement was placed on page A. We can judge that there is not. Thus, by using the derived information, the advertiser is advised to place the ad on an alternative web page that maximizes the intended exposure aimed at the target audience. obtain. By selling more advertising images on their pages, web portals and ISPs can generate additional revenue.
[0045]
While the invention described herein is used in the context of identifying alternative advertising inventory in multiple web pages, the invention is used to facilitate searching for related web pages. It should be understood that it can be easily implemented for applications in other areas, such as identifying web pages that fit a particular demographic profile. For example, assume that among the results obtained from the search, the user is particularly interested in a particular web page. Using the present invention, other related web pages similar to the web page of interest to the user can be identified and shown to the user. One skilled in the art will know other ways and ways to apply the invention.
[0046]
The invention described herein may be implemented using both hardware and / or software or a combination thereof. In the preferred embodiment, the database used to store advertising logs and other processing records is implemented using sorted flat files techniques. Various processes such as filtering, sorting, generation of indices or bitmaps, etc., are preferably implemented using computer software such as C, C ++, etc. One skilled in the art would know other ways, methods, and techniques for implementing the present invention.
[0047]
The examples and embodiments described herein are for illustrative purposes only, and are not to be construed as being included in the spirit and scope of the present application and the scope of the appended claims. It should be understood that various modifications or changes will be suggested to one skilled in the art. All publications, patents and patent applications cited herein are hereby incorporated by reference in their entirety and for all purposes.
[Brief description of the drawings]
FIG.
Yahoo! FIG. 2 is a simplified network diagram illustrating a network structure.
FIG. 2A
FIG. 4 is a simplified flow diagram of an exemplary embodiment of the present invention.
FIG. 2B
FIG. 4 is a simplified flow diagram of an exemplary embodiment of the present invention.
FIG. 2C
FIG. 4 is a simplified flow diagram of an exemplary embodiment of the present invention.
FIG. 2D
FIG. 4 is a simplified flow diagram of an exemplary embodiment of the present invention.
FIG. 2E
FIG. 4 is a simplified flow diagram of an exemplary embodiment of the present invention.

Claims (54)

A system for collecting and deriving biographical and demographic data based on user behavior,
A first plurality of user records each having a user name, demographic information, web page identification information, and browser identification information;
An index mechanism configured to form an index for each of the first plurality of user records based on the demographic information contained therein;
A first classification mechanism configured to classify the first plurality of user records based on the web page identification information;
A second classification mechanism configured to classify the first plurality of user records based on the index formed for the first plurality of user records;
With the system. The system of claim 1, wherein the index comprises a bitmap having a plurality of bits, and each bit of the bitmap represents a demographic characteristic. The system of claim 1, wherein the first plurality of user records comprises a group of user records for a particular web page after being classified by the first classification mechanism. 4. The system of claim 3, wherein analyzing the group of user records derives demographic information about the particular web page. The system of claim 1, wherein the first plurality of user records comprises a group of user records relating to a demographic profile after being classified by the second classification mechanism. 6. The system of claim 5, wherein web page identification information from the group of user records for the demographic profile can be used to identify another advertising space. 6. The web page identification from the group of user records for the demographic profile can be used to identify one or more such web pages that match particular search criteria or user preferences. System. A system for collecting and deriving biographical and demographic data based on user behavior,
A first plurality of user records each having web page identification information and browser identification information;
A first tally mechanism configured to calculate each total number of visits to each of a plurality of web pages using the web page identification information from the first plurality of user records;
A second tally mechanism configured to determine a plurality of kinships and track each cumulative sum for each of the plurality of kinships, wherein each of the plurality of webpages comprises a different one of the plurality of webpages. A system in which one relative is determined for a pair. 9. The system of claim 8, wherein each of the total number of visits and each of the cumulative totals for each of the plurality of kinship relationships can be used to identify another advertising space. 9. The system of claim 8, wherein the total number of visits to each of the plurality of web pages and the cumulative total for each of the plurality of kinships can be used to identify another advertising space. The respective total number of visits to each of the plurality of web pages and the respective cumulative total for each of the plurality of relationships identify one or more such web pages that meet particular search criteria or user preferences. 9. The system according to claim 8, which can be used for: A system for collecting and deriving demographic data based on user interaction for a first plurality of web pages,
A plurality of user records such that at least one user record is generated each time one of the first plurality of web pages is viewed;
A filtering process configured to generate a first group of user records and a second group of user records based on the plurality of user records;
An indexing process configured to form an index for each of the first group of user records;
A first classification process configured to classify the first group of user records based on a second plurality of web pages identified from the first group of user records, the first classification process comprising: A first classification process that is started after the index process of forming the index for each group is completed;
A second classification process configured to classify the first group of user records based on the index formed for the first group of user records;
A first tally process configured to calculate each total number of visits to each of the third plurality of web pages identified from the second group of user records;
A second tally process configured to determine a plurality of kinships and maintain a running total for each of the plurality of kinships, wherein a plurality of webpages from the second group of user records are provided. A system wherein a group is identified and one kinship is determined for each pair of different web pages within each of said plurality of groups of web pages. 13. The system of claim 12, wherein said first group of user records comprises a record of registered users, and said second group of user records comprises a record of all users. Each of the first group of user records includes user name, demographic information, web page identification information and browser identification information, and each of the second group of user records includes web page identification information and browser identification information. 13. The system of claim 12, including information. The system of claim 14, wherein the index for each of the first group of user records is formed using the demographic information. A record forming process configured to form one or more additional user records based on one or more of the plurality of user records;
15. The system of claim 14, wherein the plurality of user records include the one or more additional user records, and wherein the record forming process is performed before the filtering process. 13. The system of claim 12, wherein each of the total visits and each of the cumulative totals for each of the plurality of kinship relationships can be used to identify another advertising space. 13. The system of claim 12, wherein the index comprises a bitmap having a plurality of bits, and each bit of the bitmap represents a demographic characteristic. 13. The system of claim 12, wherein the first group of user records comprises a sub-group of user records for demographic profiles after being classified by the second classification process. 20. The system of claim 19, wherein web page identification information from the sub-group of user records for the demographic profile can be used to identify another advertising space. 13. The method of claim 12, wherein each of the total visits and each of the cumulative totals for each of the plurality of kinship relationships can be used to identify one or more such web pages that meet particular search criteria or user preferences. System. 20. The method of claim 19, wherein web page identification information from the sub-group of user records for the demographic profile can be used to identify one or more such web pages that meet particular search criteria or user preferences. The described system. In a system that identifies different advertising inventory in web pages,
Comprising a plurality of ad records, an ad record is generated for each ad appearing on the viewed web page, and each ad record includes a B cookie and a space ID;
The system further comprises an ad record formation routine for forming one or more additional ad records based on one or more of the plurality of ad records, wherein the plurality of ad records include the one or more additional ad records. Including records,
A filtering routine for generating a first recording group and a second recording group based on the plurality of ad recordings, wherein each of the first recording groups further includes a P cookie and an L cookie;
A first classification routine for classifying the first record group based on the L cookie;
An indexing routine for forming an index for each of the first group of records using the P cookie;
A second classification routine for classifying the first recording group based on the space ID;
A third classification routine for classifying the first recording group based on the index formed for the first recording group;
A first tally routine for calculating a total number of visits to each of the plurality of web pages identified from the space ID of each of the second record groups;
A second tally routine for determining a plurality of kinships and maintaining a running total for each of the plurality of kinships, wherein the second tally routine further comprises: A system wherein a plurality of groups are identified and one relative is determined for each pair of different web pages in each of the plurality of groups of web pages. The B cookie contains browser identification information,
The P cookie contains demographic information,
24. The system of claim 23, wherein the L cookie includes a user name and the space ID includes web page information indicating which web page is associated with the ad record. 24. The system of claim 23, wherein the index comprises a bit matrix having a plurality of bits, and each bit of the bitmap represents a demographic characteristic. 24. The system of claim 23, wherein the first recording group includes a sub-group of recordings for a particular web page after being classified by the second classification routine. 27. The system of claim 26, wherein analyzing a sub-group of the records derives demographic information about the particular web page. 24. The system of claim 23, wherein the first group of records comprises a subgroup of records relating to demographic profiles after being classified by the third classification routine. 29. The system of claim 28, wherein the space ID from the subgroup of records for the demographic profile can be used to identify the different advertising inventory. 24. The system of claim 23, wherein each of the total visits and each of the cumulative totals can be used to identify the different advertising inventory. 29. The method of claim 28, wherein the space ID from the sub-group of the record for the demographic profile can be used to identify one or more such web pages that match particular search criteria or user preferences. system. 24. The system of claim 23, wherein each of the total visits and each of the cumulative totals can be used to identify one or more such web pages that meet a particular search criterion or user preference. A method for collecting and deriving biographical and demographic data based on user behavior with respect to a plurality of web pages, comprising:
Retrieving a first plurality of user records each having a user name, demographic information, web page identification information, and browser identification information;
Forming an index for each of the plurality of user records based on the demographic information contained therein;
Classifying the plurality of user records based on the web page identification information, and classifying the plurality of user records based on the index formed for the plurality of user records;
A method including the step of: 34. The method of claim 33, wherein the index comprises a bitmap having a plurality of bits, and each bit of the bitmap represents a demographic characteristic. Identifying a group of user records for a particular web page from the first plurality of user records after the first plurality of user records have been classified based on the web page identification information; and Deriving demographic information about the particular webpage from
34. The method of claim 33, further comprising the step of: Identifying a group of user records belonging to a demographic profile from the first plurality of user records after the first plurality of user records have been classified based on the respective indexes; and Identify another advertising inventory using the above web page identification from
34. The method of claim 33, further comprising the step of: 37. The method of claim 36, further comprising using the web page identification information from the group of user records to identify one or more such web pages that match particular search criteria or user preferences. In a method for collecting and deriving biographical and demographic data based on user behavior,
Searching for a plurality of user records each having web page identification information and browser identification information;
Calculating the total number of visits to each of the plurality of web pages using the web page identification information from the plurality of user records;
Identifying a plurality of groups of web pages based on the browser identification of the plurality of user records;
A plurality of kinships are determined such that one kinship is determined for each pair of different webpages in each of the plurality of groups of webpages, and each kinship is determined for each of the plurality of kinships. Track the running total,
A method including the step of: 39. The method of claim 38, further comprising deriving another advertising inventory using each of the total visits and each of the cumulative totals. 39. The method of claim 38, further comprising using each of the total number of visits and each of the cumulative totals to identify one or more such web pages that meet particular search criteria or user preferences. In the method of identifying different advertising inventory in a web page,
An ad record is generated for each ad that appears on the viewed web page, and a plurality of ad records are generated such that each ad record includes browser information and web page identification information;
Filtering the plurality of ad records into a first record group and a second record group, each of the first record groups further including demographic information and a username;
Forming an index for each of the first record groups using the demographic information contained therein;
Classifying the first recording group based on the web page identification information;
Classifying the first recording group based on the index formed for the first recording group;
Calculating a total number of visits to each of the plurality of web pages identified from the web page identification information of the second recording group;
Identifying a plurality of groups of web pages based on the browser information of the second recording group;
A plurality of kinships are determined, such that one kinship is determined for each pair of different webpages in the plurality of groups of webpages, and each accumulation for each of the plurality of kinships is determined. Calculate the sum,
A method including the step of: After the filtering step, categorizing the first record group based on the user name contained therein and forming one or more additional records based on one or more of the plurality of ad records. 42. The method of claim 41, wherein the plurality of ad records include the one or more additional records. Identifying a sub-group of records for a particular web page from the first recording group after the first recording group has been categorized based on the web page identification information; and identifying the particular web from the sub-group of recordings. Derive demographic data about the page,
42. The method of claim 41, further comprising the step of: Identifying the sub-group of records belonging to the demographic profile from the first record group after the first record group has been classified based on each of the indexes; and identifying the web page from the sub-group of records. Use the information to identify different advertising inventory,
42. The method of claim 41, further comprising the step of: The method of claim 44, further comprising using the web page identification information from the user record group to identify one or more such web pages that meet certain search criteria or user preferences. 42. The method of claim 41, wherein the total number of visits and the cumulative total are used to identify the different advertising inventory. 42. The method of claim 41, further comprising using the respective total number of visits and the cumulative total to identify one or more such web pages that meet a particular search criterion or user preference. In the method of identifying different advertising inventory in a web page,
Means for generating a plurality of ad records, wherein an ad record is generated for each ad appearing on the viewed web page, and wherein each ad record includes browser information and web page identification information;
Means for filtering the plurality of ad records into a first record group and a second record group, each of the first record groups further comprising demographic information and a username.
Means for forming an index for each of the first record groups using the demographic information contained therein;
Means for classifying the first recording group based on the web page identification information;
Means for classifying the first recording group based on the index formed for the first recording group;
Means for calculating a total number of visits to each of the plurality of web pages identified from the web page identification information of the second recording group;
Means for identifying a plurality of groups of web pages based on the browser information of the second recording group;
Means for determining a plurality of kinships such that one kinship is determined for each pair of different webpages in the plurality of groups of webpages;
Means for calculating each cumulative sum for each of the plurality of relatives;
With the system. Means for classifying the first recording group based on the user name contained therein after the filtering step;
Means for forming one or more additional records based on one or more of the plurality of ad records, wherein the plurality of ad records includes the one or more additional records. Item 49. The system according to Item 48. Means for identifying a sub-group of records for a particular web page from the first recording group after the first recording group has been classified based on the web page identification information;
Means for deriving demographic data for the particular web page from the sub-group of records;
49. The system of claim 48, further comprising: Means for identifying a subgroup of records belonging to a demographic profile from the first recording group after the first recording group has been classified based on each of the indices;
Means for identifying another advertising inventory using the web page identification information from the sub-group of records;
49. The system of claim 48, further comprising: 52. The system of claim 51, further comprising means for using the web page identification information from the user record sub-group to identify one or more such web pages that match particular search criteria or user preferences. . 49. The system of claim 48, further comprising means for using the respective total visits and the respective cumulative totals to identify the different advertising inventory. 49. The system of claim 48, further comprising means for using the respective total number of visits and the cumulative total to identify one or more such web pages that match a particular search criterion or user preference.

Images