JP2002531897A - Consumer research and ad selection system - Google Patents

Abstract

(57) Summary A consumer survey and advertisement selection system (500) is presented that can characterize consumers (100) based on their purchasing and viewing habits. The result of this process is a consumer characterization vector (562) that describes the probabilistic demographics and product preferences of the subscriber or viewer (100). An advertising characterization vector (548) can be determined that describes a real or virtual market for a product or desired audience. Advertising features, including advertising statistics vector (548), advertising product category (552), and advertising product preference vector (554) are transmitted along with the consumer ID (512). The consumer ID (512) is used to search for a consumer characterization vector (562) that correlates with the advertisement characterization vector (548) to determine the suitability of the advertisement for the consumer (100). Is done. The price to place the advertisement can be determined based on the results of the correlation of the advertisement features with the consumer characterization vector (562). The system can be used to increase the effectiveness and cost-efficiency of advertisements, as well as determine prices for transmitting or displaying advertisements based on the collection of advertisements using consumer profiles.

Description

DETAILED DESCRIPTION OF THE INVENTION

TECHNICAL BACKGROUND The arrival of the Internet will create the ability to communicate data instantaneously throughout the world and will address a number of new applications that enhance the lives of consumers. One potential enhancement is the ability of consumers to receive advertisements related to their lifestyle, rather than the advertising flow determined by the programs they are watching. Such "targeted advertising" can potentially reduce the amount of unnecessary information consumers receive in the mail, between television programs, and when using the Internet. Examples of editing goal settings can be found on the World Wide Web where banners are served based on page content. 1
Http://www.worldwidewebsite.com on June 19, 998
. doubleclick. Double-click (D
doubleClick) Dynamic Ad Reporting and Goal Setting (Dynamic A
Dvertizing Reporting and Targeting (D
ART) "discloses a double-click advertising solution to match the advertiser's selected targeted profile with the personal user profile and deliver the appropriate banner. . Users and advertisements are matched based on keywords related to geographical location or page content. On June 30, 1998, the World Wide Web site http: // www. star
pt. The product reference "Ad Force" from Imgis, printed from the com / core, discloses an advertisement management system for targeting users and delivering their advertisements. Users are targeted based on the type of content or keywords they are watching.

BACKGROUND OF THE INVENTION [0002] From an advertiser's perspective, the ability to target advertising is annoying because it is determined that their advertisement is at least relevant to the consumer, and thus the advertisement does not fit into a lifestyle. Advertisers may find it useful to have some confidence that they will not be considered. Various systems have been proposed for matching consumer profiles to advertisements, such as U.S. Patent No. 5,774,170, which discloses a system for delivering targeted advertisements to consumers.

[0003] Other systems propose methods for delivering programs tailored to a subscriber's profile. U.S. Pat. No. 5,446,919 discloses a communication system that can target a demographically or psychographically defined audience. Demographic and psychographic information about the audience members is downloaded and stored in the audience member receiver. The media messages are transmitted to the viewer members along with a selection profile command that details the demographic / psychographic profile of the viewer member that will receive each media message. Audience members corresponding to the group specified by the selection profile command are presented with a media message.

US Pat. No. 5,223,924 discloses a system and method for automatically correlating user preferences to a television program information database. The system includes a processor that performs a "free text" search technique to correlate the downloaded television program information with viewer preferences. U.S. Patent No. 5,410,344 discloses a method for selecting an audio-video program based on viewer preferences, wherein each of the audio-video programs has a plurality of program attributes and a corresponding representation of the program attributes. Content code. The method comprises the steps of storing a viewer preference file including an attribute rating indicating a degree of impact of the program attributes on the viewer in response to a comparison between the program content code and the viewer preference file, wherein the program includes Selected for.

[0005] To determine the applicability of an ad to consumers, it is necessary to know something about its lifestyle, and especially to understand its demographics (age, family size, and income) . In some cases, it is useful to know that particular purchasing habit. Purchasing habits are used by e-commerce to look up their visitors. As an example, on June 30, 1989, the World Wide Web site http: // www. aptex. com / product
s-selectcast-commerce. The product reference "SelectCast for Commerce Server" from Aptex software Inc., printed from http://www.aptexsoftware.com, discloses the product SelectCast for commerce servers. The product personalizes online shopping based on observed user attitudes. User interests are learned based on the content they browse, the promotions they click on, and the products they purchase.

[0006] Knowledge of a consumer's purchasing habits is that soup vendors want to know which consumers are buying their competitor's soup and, as a result, switch brands to consumers. May be beneficial to product vendors in that they can set advertising goals for their consumers in the task of persuading them.
In the case of a new product referral, the strategy may be to convince the loyal customer to try the new product, but the vendor probably does not want to target the loyal customer right. In both cases, it is very useful for the vendor to be able to determine which brand of product the consumer is currently purchasing.

[0007] There are several problems associated with collecting, processing, and storing consumer data. First, it may be difficult to collect consumer data and determine consumer demographic parameters. Research can be done and consumers will be glad to have access to family data, usually personal, including family size. In these situations, there is usually a need for consumer agreement on how the data will be used. If the consumer does not provide this data directly, the information must be "looked up" from a variety of information collected about the consumer, typically from special purchases.

[0008] A relatively intrusive method for collecting consumer information is described in US Pat. No. 4,546,382, which discloses a television and market research data collection system and method. A data collection device with memory can be used to see which of the multiple TV modes are being used, as well as which TV channels are being used, as well as input from the appropriate optical scanning device to collect consumer product purchases. Stores data on whether there is

Once data is collected, usually from a single source, some processing
It can be performed to determine a particular aspect of a consumer's life. As an example, the process is:
It can be performed on credit data to determine which consumers have good credit risk and have recently applied for a credit. The resulting list of consumers can typically be invited by direct mail. Information such as credit history is stored in multiple databases, but storage of other information, such as details of food purchases, is typically not performed. If a detailed list of each individual food purchase was recorded, the information would be of little use since it would be nothing more than an open shopping list.

[0010] Privacy concerns are also an important factor in using consumer purchase information.
Consumers will usually find that advertising and other information are tailored to their interests, but will not allow indiscriminate access to their demographic profiles and purchase records.

[0011] The Internet has created the concept of "negatively priced information" that consumers are paid to accept advertisements. Consumers paying to view advertisements can be accomplished interactively on the Internet, allowing consumers to view advertisements at a certain price. A. Nathanael Goldhab
ber and Gary Fitts are the inventor, "Attention
A previously proposed scheme, such as the scheme described in US Pat. No. 5,794,210 entitled "On Brokerage," provides a method in which a consumer is presented with a list of advertisements and their corresponding payments. Such a system will be described. Consumers are compensated by choosing from the list and seeing the advertisement. The system also uses software agents that represent consumers that match the consumer's interest profile with the advertisement. Matching is performed using "relevance indexing" based on a hierarchical tree structure. The system requires real-time interactivity in which the viewer must select an advertisement from a list of options being presented.

[0012] The ability to place advertisements to consumers and compensate consumers for viewing advertisements opens up many possibilities for new models of advertisements. However, it is important to understand consumer demographics and product preferences to be able to determine whether an advertisement is appropriate.

Although it is possible to gather statistics about the consumers of a particular product and compare its profile to individual demographic data points, such a methodology is based on existing customers of the same or similar products. Only address potential consumer choices based on demographics.

[0014] John B. "A system and method for selectively distributing commercial messages in a communication network" (System and me
thing for selective distributing co
US Patent No. 5,515,0 entitled "Mercial messages)".
No. 98 describes how actual consumer target home data for a product is compared against subscriber home data to determine the applicability of a commercial to a home. The target home of the product or service is characterized by comparing the profile of the customer home to the profile of all homes or correlating them. A rating is established for each household for each category of goods / services. As defined by the rating, homes within the predefined subscriber's percentile are targeted by the product or service advertiser.

[0015] Also, in many cases, the goal is to advertise to markets with arbitrary characteristics, and rather than relying on existing consumer characteristics of the product, probabilistic or decision of the consumer / subscriber It would be desirable to obtain a measure of the correlation of any of these features with theoretical data. Such a correlation would probably be based on both demographic characteristics and product preferences.

R. Dedrick is an inventor of "Methods and Apparatus for Providing Electronic Advertising to End Users at Best Pricing for Consumers (Method and
Apparatus for providing electronica
diversements to end users in a cons
Another previously proposed system described in U.S. Pat. No. 5,724,521 entitled "umer best-fit pricing manner"
Leverage the consumer scale as a mechanism to determine to which group an ad is intended. The consumer scale matching process compares a set of features stored in the user profile database to a consumer scale associated with the electronic advertisement. The fee charged to the advertiser is determined by where the set of features falls on the consumer scale. Such systems require the specification of a large number of parameters and weighting factors, and require access to certain non-statistical personal profile information.

For the above reasons, there is a need for a consumer survey system that can examine consumers, provide access to consumer profiles in a secure manner, and return a measure of the potential applicability of an advertisement. . There is also a need for an ad selection system that can tailor advertisements to any target market characteristics and do so in a manner that protects the privacy of consumer data and characteristics.

DISCLOSURE OF THE INVENTION The present invention supports receiving consumer purchase information where the consumer characterization vector is updated based on the product characterization information. Consumer characterization vectors are consumer demographic vectors that provide a probable measure of consumer demographics, and products that typically describe products that consumers have purchased in the past and therefore likely to buy in the future Contains preference vector. Product characterization information is a probabilistic determination of the demographics of the purchaser of an item, heuristic rules that can be applied to probabilistically describe the demographics of the consumer based on the purchase, and the purchase itself. Represents the vector notation.

In a preferred embodiment, a detailed computer readable purchase record is received with a unique consumer identifier. Demographic characterization vectors corresponding to consumers can be retrieved. If there is no existing demographic characterization vector for the consumer, a new demographic characterization vector can be created. In a preferred embodiment, the new demographic characterization vector contains no information. A set of heuristic rules is retrieved, including probabilistic criteria for the demographic characteristics of a typical purchaser of an item. A new demographic characterization vector is calculated based on purchases, existing demographic characterization vectors, and heuristic rules.

In a preferred embodiment, the calculation of the demographic characterization vector is performed by calculating a weighted average of the product demographic vector and the existing demographic characterization vector. A weighting factor is used, where the weighting factor is determined based on the ratio of the current product purchase to the cumulative product purchase. Cumulative product purchases can be measured as the amount spent on a particular category of item (eg, food, clothing, accessories) for a specified time period, such as one month or one year.

In a preferred embodiment, the heuristic rules take the form of a product demographic vector that describes the demographics of known purchasers of an item. Each product may have an associated product demographic vector.

The present invention is used to create consumer product preference descriptions that describe the brand and size products that consumers purchase and provide a probable interpretation of the products they are likely to purchase in the future. be able to. Product preference descriptions can be generated by creating a weighted average of existing product preference vectors that describe a consumer's historical product preferences (product type, brand and size) and characteristics of recent purchases.

The present invention can be implemented as a data processing system or computer program that processes consumer purchase records and updates its demographic and product preference profiles based on the use of product characterization information. The data processing system
It can also be used to receive information about advertisements and to perform a correlation between advertisements and consumer demographics and product preferences.

The present invention can be realized as software resident on one or more computers. The system can be implemented as individual computers that receive information about consumer purchases or can be implemented in a network of computers where parts of the system reside on various computers.

One advantage of the present invention is that it enables a consumer profile to be automatically updated based on its purchase and forms a description about the consumer, including demographic features and product preferences. is there. This description can be used by advertisers to determine the suitability of the advertisement for the consumer. Consumers benefit from the system in order to receive advertisements that appear more applicable to them.

The present invention can be used to examine consumers, support advertisement characterization vectors associated with advertisements with consumer characterization vectors, and determine applicability of advertisements to consumers.

Another feature of the present invention is that based on the degree of correlation of an advertisement with its profile,
The ability to price access to consumers. If the advertisement turns out to be very strongly correlated with the consumer demographics and product preferences, a relatively high price can be charged to send the advertisement to the consumer. From the consumer's point of view, if the correlation between the advertisement and the consumer's demographics or product preferences is high, the consumer may be charged less to see the advertisement, as it may be important.

The present invention is based on the reception of a characterization vector and the use of a unique consumer ID,
A system for determining the applicability of an advertisement to a consumer is also described. The consumer ID is
Used to search for consumer characterization vectors, and the correlation between the consumer characterization vector and the advertisement characterization vector is used to determine the applicability of the advertisement to the consumer. The price paid for the presentation of the advertisement can be determined based on the degree of correlation.

The price for presenting an advertisement may increase with correlation, as is typical when content / opportunity providers are also surveying entities. Pricing is when consumers are profilers who are interested in seeing ads that are highly correlated to their demographics, lifestyle, and product preferences and are willing to charge less for seeing ads. Decreases with correlation.

The present invention can be used to specify the purchaser of a particular product.
In a preferred embodiment, the advertisement characterization vector includes an indication of the target product, ie, a description of the target market, including the buyer, brand, and product size for a particular product type. The advertisement characterization vector correlates to the consumer characterization vector retrieved based on the unique consumer ID. A correlation factor is determined and indicates whether the consumer is a purchaser of the product to be advertised. This feature can be used to identify buyers of a particular brand and can be used to target ads to invite those consumers to leave their current product provider. In a similar manner, this feature can be used to introduce loyal consumers to new products within the product family, or products of different sizes.

One advantage of the present invention is that it can specify any target market parameters and does not necessarily correspond to an existing market, but reflects various market segments to be advertised. The point is that you can do it. Market segments can be indicated by demographic characteristics or by product preferences.

Another advantage of the present invention is that a demographic sample of the buyer of the product is not required to define the target market.

The present invention can be used to determine the applicability of an advertisement to consumers based on demographics, product preferences, or a combination of both.

In a preferred embodiment of the invention, the correlation is calculated as a scalar product of the advertisement characterization vector and the consumer characterization vector. The advertising characterization vector and the consumer characterization vector can be composed of demographic features, product purchase features, or a combination of both.

In a preferred embodiment, the pricing of the display of the advertisement is created based on the result of the correlation between the advertisement characterization vector and the consumer characterization vector. In the first embodiment, the price increases as a function of the correlation. This embodiment may represent a situation where the party determining the correlation also controls the ability to display the advertisement.

In an alternative embodiment, the price for displaying an advertisement is reduced as a function of the degree of correlation. This embodiment may represent a situation in which consumers control access to consumer characterization vectors and charge less to see ads that are highly correlated with their interests and demographics. A feature of this embodiment is the ability of the consumer to reduce the number of unnecessary advertisements by charging higher prices to view less important advertisements.

One advantage of the present invention is that it allows ads to be directed to new markets by setting certain parameters in the ad characterization vector, and specific statistical information about existing customers of similar products. It does not require knowledge. Another advantage is that the system allows ads to be directed to consumers of competing brands, or to specifically target loyal customers. This feature may be useful for introducing new products to an existing customer base.

Another advantage of the present invention is that correlation can be performed by calculating a simple scalar (dot) product of the advertisement characterization vector and the consumer characterization vector. No weighted sum or other statistical analysis is needed to determine the applicability of the advertisement.

The present invention can be realized as a data processing system and as a computer program. The invention can be implemented on individual computers or using distributed computers with portions of the system running on a variety of computers.

An advantage of the present invention is the ability to direct advertisements to consumers who will find advertisements of interest. This eliminates unnecessary advertising. Another advantage is the ability to target a particular group of potential advertiser customers.

[0041] These and other features and objects of the invention will be more fully understood from the following detailed description of preferred embodiments which should be read in view of the accompanying drawings.

The accompanying drawings, which are incorporated in and form a part of the specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

DETAILED DESCRIPTION In describing the preferred embodiments of the invention shown in the figures, certain terms will be used for clarity. However, the invention is not intended to be limited to the specific terms so selected, and each specific term is intended to include all technical equivalents that operate in the same manner to achieve a similar purpose. It should be understood to include

Referring generally to the figures, and particularly to FIGS. 1-10, the methods and apparatus of the present invention are disclosed.

FIG. 1A is a user relationship diagram showing the relationship between the consumer research system and various entities. As can be seen in FIG.
May be received from a consumer personal computer (PC) 104, or may receive an ad 182 mailed to it.

The advertisements and information displayed on the consumer PC 104 or the television 108 can be received on the Internet 150 or can be received on a combination of the Internet 150 and another telecommunications access system. Telecommunications access systems include cable television delivery systems, switched digital video access systems operating over telephone lines, microwave telecommunications systems, or consumers.
0 and any other media that provides connectivity between the content server 162 and the advertising server 146, but is not so limited.

The content / opportunity provider 160 maintains a content server 162 that can transmit content including broadcast programs over a network such as the Internet 150. Other methods of data transport can be used, including private data networks, but the content server 160 can be connected to a device owned by the consumer 100 through an access system.

The content / opportunity provider 160 allows the consumer 100 to
It is so called because the content / opportunity provider can insert an advertisement when it receives a transmission from 2. For video programming, the content / opportunity provider is typically the cable network operator, or source of entertainment material, and the opportunity refers to the ability to send advertisements during commercial breaks.

The majority of the content being transmitted today is thus performed in broadcast formats such as broadcast television programs, broadcast radio, and newspapers (broadcast by satellite and over cable television networks). You. While the interconnectivity provided by the Internet allows consumers to send special programs, there is still a large amount of broadcast material where advertisements can be partially sponsored. Ability to insert advertisements into the broadcast stream (video, audio, or mail)
Is an opportunity for the advertiser 144. Content can also be broadcast over the Internet and combined with existing video services, in which case there will be opportunities for advertisement insertion.

FIG. 1A shows the content / opportunity provider 160 and the content server 162 connected independently to the Internet 150, and the consumer device is also directly connected to the Internet 150, 160 may also control access to subscribers. This can occur when the content / opportunity provider is also a cable operator or a telephone company. In such an example, a cable operator or telephone company may provide content to consumers 100 on a cable operator / telco access network. As an example, if the cable operator controls the content being sent to the consumer 100 and programs the number of times for insertion of the advertisement, the cable operator will allow the cable operator to run the advertisement at the time of the commercial break. The cable operator is considered a content / opportunity provider 160 because the insertion can provide the advertiser with an opportunity to access the consumer 100.

In a preferred embodiment of the present invention, a pricing policy can be defined.
Content / opportunity provider 160 may charge advertiser 144 for access to consumer 100 during the opportunity. In a preferred embodiment, the price charged for access to the consumer 100 by the content / opportunity provider varies as a function of the applicability of the advertisement to the consumer 100. In an alternative embodiment, the consumer 100 retains control over access to profiles and fees for viewing advertisements.

The content provider may be a mailing company or a printing company that creates print information for the consumer 100. As an example, the content server 162 is connected to the consumer 100 to the printer 164 that creates the mailed advertisement 182. Alternatively, the printer 164 can create an advertisement to enter the newspaper delivered to the consumer 100. Other prints are created by the printer 162 and delivered to the consumer 100 in a variety of ways.

Advertiser 144 maintains an advertisement server 146 that includes a wide variety of advertisements in the form of printable still video, video advertisements, audio advertisements, or a combination thereof.

The profiler 140 maintains a consumer profile server 130 that contains the characteristics of the consumer 100. The consumer survey system is a consumer profile server 1
30 or connected to a consumer profile server to look up consumers,
Operated by a profiler 140 that can use another computing device.

Data for performing a consumer survey is received from the point of purchase 110. The point of purchase 110 may be a grocery store, department store, or other retail outlet, or a website or other location where a purchase request is received and processed. In the preferred embodiment, data from the point of purchase is transferred over a public or private network 120, such as a local area network within the store or a wide area network connecting many department stores or grocery stores. In an alternative embodiment,
The data from the point of purchase 110 is transmitted over the Internet 150 to the profiler 140
Sent to.

The profiler 140 collects data from the store, but may also be a third party who contracts with the consumer 100 and the retailer to receive purchase point data, and a profile consumer 100. Consumers 100 can make such arrangements based on the increased convenience provided by targeted advertising or through reimbursement equipment that they pay regularly for revealing their particular purchase records. You may agree.

The consumer profile server 130 is a consumer television 108 or a consumer PC
It may include a consumer profile determined from viewing habits observations at 104. Determining demographic and product preference information based on the consumer's use of services such as cable television and Internet access monitors the subscriber's choice of programming and is associated with the subscriber selection and the program being watched. This can be accomplished by determining household demographics based on the information obtained.

In one embodiment, the channel selection is recorded and based on the time of day in which the program is watched and the viewing period, heuristic rules are applied and, as shown in FIG. 2A, age, gender, Make probabilistic decisions on household demographics, including household size, and income. This can be achieved by applying a heuristic rule that associates the program with known and assumed features of the viewer of the program. As an example, there is a high probability that the viewers of the morning manga are in the age group of 3 to 8 years old.
It is known that there is a high probability that one or more viewers in the age group are included.

In one embodiment, information about the program is extracted from an electronic program guide (EPG) that contains information about the scheduled program. In another embodiment, information about the program is retrieved from a closed caption channel transmitted in a broadcast signal.

The volume at which a program is viewed is also stored, forming an additional basis for characterization of the subscriber, where mute of a channel indicates limited interest in a particular program or advertisement. In the case of ads, the silencing of the ad depends on the effectiveness (or ineffectiveness) of the ad
And may serve as part of the basis for subscriber characterization. The muting of the program, as well as the duration during which the program is being watched, can also be used to determine the subscriber characterization vector.

By processing the recorded viewing habits with program-related information and rules similar to those shown in FIG. 7, but related to programs rather than purchases, a probable demographic profile of the home It is possible to construct a subscriber characterization vector containing

[0062] As used herein, the term consumer characterization vector also refers to the subscriber characterization vector described above. Both the consumer characterization vector and the subscriber characterization vector include demographic information and product preference information for the consumer 100.

FIG. 1B shows an alternative embodiment of the present invention in which the consumer 100 is also a profiler 140. Consumer 100 may directly or through consumer PC 104 or set-top 106 connect to a consumer profile server 13 connected to a network.
Maintain 0. The consumer profile server 130 may include a consumer survey system, or the survey may be performed in connection with the consumer PC 104 or the set top 106. A subscriber characterization system that monitors the viewing habits of the consumer 100 can be used in conjunction with a consumer survey system to create more accurate consumer profiles.

When the consumer 100 is also a profiler 140, as shown in FIG. 1B, access to consumer demographics and product preference characterization may be due to cash compensation instead of receiving increased advertising accuracy. Or consumer 1 who will grant access to profiles instead of discounts or coupons for goods and services
It is controlled exclusively by 00.

FIG. 2A shows an example of a probabilistic demographic characterization vector. A demographic characterization vector is a representation of the probability that a consumer falls into a certain demographic category, such as age group, gender, family size, or income range.

In a preferred embodiment, the demographic characterization vector includes a category of interest. The categories of interest may be organized according to a wide range of fields such as music, travel, and restaurants. Examples of music interest categories include country music, rock, classical, and folk. Examples of travel categories include "travel to another state more than once a year" and "travel more than twice a year by plane."

FIG. 2B shows a deterministic demographic characterization vector. The resulting demographic characterization vector is a representation of the consumer profile as determined from deterministic rather than probabilistic data. As an example, if the consumer 100 agrees to answer certain questions regarding age, gender, family size, income and interests, the data contained in the consumer characterization vector will be deterministic.

As with probabilistic demographic characterization vectors, deterministic demographic characterization vectors may include categories of interest. In a preferred embodiment, consumer 1
00 answers specific questions of surveys generated and managed by profiler 140 and created and managed by telephone, in writing, or via Internet 150 and consumer PC 104. The survey questions either correspond directly to elements in the probabilistic demographic characterization vector or obtain deterministic results due to storage in the demographic characterization vector.

FIG. 2C shows a product preference vector. Product preferences represent the average of consumer preferences from past purchases. As an example, buying a breakfast cereal manufactured by Kellogg under the trademark CORN FLAKES is about twice as much as buying a breakfast cereal manufactured by Post under the trademark ALPHABITS. A consumer who never purchases a breakfast cereal manufactured by General Mills at WHEATIES will have a product taste characterization such as the characterization shown in FIG. 2C. As shown in FIG. 2C, the preferred size of a consumer purchase of a particular product type can also be represented by a product preference vector.

FIG. 2D may be comprised of a consumer ID field 237, a deterministic demographic data field 239, a probabilistic demographic data field 241, and one or more product preference data fields 243. FIG. 4 represents a data structure for storing a consumer profile. As shown in FIG. 2D, the product preference data field 243 may be composed of a plurality of fields arranged by product category 253.

Depending on the data structure used to store the information contained in the vector,
Any of the aforementioned vectors may take the form of a table, record, linked table, series of records, or software object in a relational database.

The consumer ID 512 may be an arbitrary identification value uniquely associated with the consumer 100. In the preferred embodiment, consumer ID 512 is a telephone number, but in an alternative embodiment, consumer ID 512 is a credit card number. Other unique identifiers include consumer names, consumer addresses, and social security numbers that include the initials of the intermediate names or unique alphanumeric sequences.

The vectors described and represented in FIGS. 2A-2C form consumer characterization vectors of varying lengths and dimensions, and portions of the characterization vectors can be used individually. The vectors can also be concatenated or aggregated to produce a longer vector that provides a more detailed profile of the consumer 100.
A matrix notation of a vector that indexes a particular element, such as product category 253, can be used. Hierarchical structures can be used to organize vectors, make hierarchical search algorithms available, and locate special parts of vectors.

FIGS. 3A and 3B show an advertising demographic vector and an advertising product preference vector, respectively. Ad demographic vectors that are similar in structure to the demographic characterization vector are used to target ads by setting demographic parameters in the ad demographic vector to correspond to the demographic group of interest Is done. As an example, the advertisement is in the age group of 18 to 24 years and in the age group of 24 to 32 years, there is no gender bias, a typical household size is 2 to 5, and income is typical If an ad is created for a market that is in the range of $ 20,000 to $ 50,000, the ad demographic vector will be similar to that shown in Figure 3A. Ad demographic vectors provide a statistical estimate of who the ad is for, based on the advertiser's belief that the ad will be beneficial to the manufacturer when the ad is viewed by individuals in those groups. Represent. The advantages are
Typically, this will take the form of increased product sales or brand awareness. “Image ads,” which merely show artistic work as an example, but do not sell products directly, can be very effective for young people, but annoying to older individuals. The advertising demographic vector can be used to establish criteria for directing ads to demographic groups between the ages of 18 and 24.

FIG. 3B shows an advertisement product preference vector. The advertising product preference vector is used to select consumers with a particular product preference. In the example shown in FIG. 3B, the advertisement product preference vector is set so that the advertisement can be directed to the purchasers of AlphaBits and Whites instead of the purchasers of cornflakes. This particular setting will be effective when advertisers represent Kellogg and are billed for increased cornflakes sales. By targeting current buyers of AlphaBits and Whiteies, advertisers can try to move those buyers to the Kellogg brand and specifically convince them to buy cornflakes. Given that there is a payment required to present an advertisement to the content / opportunity provider 160 or to the customer 100, the advertiser 144 sets an advertising goal, thereby We want to be cost effective.

If the advertiser 144 wishes to reach only Kellogg's cornflake buyers, the category would be set to a high value, and in the example shown to one. As shown in FIG. 3B, the size of the product can also be specified. If there is no preference in the size category, the values can be set to be equal. In a preferred embodiment, the value of each feature, including brand and size, is normalized separately.

Ad characterization because ads can be targeted based on a set of demographic and product preference considerations that may not represent any particular group of current consumers of the product The vector can be set to identify a number of demographic groups that would normally be considered uncorrelated. Since the advertisement characterization vector may have a target profile that does not represent the actual consumer of the product, the advertisement characterization vector may be considered to have any element. As used herein, any term refers to the selection of target market features that need not represent the actual existing market or a single purchase segment.

In a preferred embodiment, the consumer characterization vectors shown in FIGS. 2A-2C and the advertisement characterization vectors shown in FIGS. 3A and 3B have a standardized format, where Is where each demographic feature and product preference is identified by an indexed location. In this embodiment, the single value is 1
Represents two probabilistic or deterministic values (eg, the probability that a consumer is in the age group of 18 to 24 years, or weighting of an advertisement's age group)
.

In an alternative embodiment, groups of demographic and product features form individual vectors. As an example, an age category can be viewed as a vector, where each component of the vector represents the probability that the consumer is in that age group.
In this embodiment, each vector can be considered as a base vector for a consumer description or target advertisement. Consumer characterization, or advertisement characterization, consists of a finite set of vectors in a vector space that describes the consumer or advertisement.

FIG. 4 shows a block diagram of a computer system for realizing a consumer survey system. System bus 422 transfers data between CPU 203, RAM 204, read-only memory-basic input / output system (ROM-BIOS) 406, and other components. The CPU 203 accesses the hard drive 400 through the disk control device 402. The standard input / output device is connected to the system bus 422 through the I / O control device 201. Keyboard is keyboard port 41
6, the monitor is attached to the I / O controller 201, and the monitor is connected to the monitor port 418.
Connected through. The serial port device uses the serial port 420 to communicate with the I / O controller 201. Industry standard architecture (ISA
) Expansion Slot 408 and Peripheral Component Interconnect (PCI) Expansion Slot 41
0 allows additional cards to be inserted into the computer. In a preferred embodiment, a network card can be used to connect a local area network, a wide area network or other networks. The computer system shown in FIG. 4 may be part of the consumer profile server 130 or may be a processor residing in another element of the network.

FIG. 5 shows a context diagram of the present invention. Context diagrams are useful in showing the relationships between the system and external entities. Although the use of context diagrams does not limit implementations of the invention to a particular programming language, context diagrams may be particularly useful in developing an object-oriented implementation of a system. The present invention relates to C, C ++, Smalltalk,
It can be implemented in a wide variety of programming languages, including but not limited to Java and Perl, and can be developed as part of a relational database. Other languages and data structures can be utilized to implement the present invention and are known to those skilled in the art.

Referring to FIG. 5, in a preferred embodiment, consumer survey system 500 resides on consumer profile server 130. Purchase point record 510 is purchase point 11
0 and stored in the consumer profile server 130. Heuristic rule 5
30, pricing policy 570, and consumer profile 560 are similarly stored on consumer profile server 130. In a preferred embodiment, the advertising records 540 are stored on the advertising server 146, and the connectivity between the advertising records 540 and the consumer survey system 500 is via the Internet or other network.

In an alternative embodiment, the entities represented in FIG. 5 are located on servers interconnected via the Internet or other networks.

Consumer survey system 500 receives purchase information from purchase points, as represented by purchase point record 510. Information included in the purchase points of the purchase record 510 includes a consumer ID 512 and a product ID 514 of the purchased product. In the preferred embodiment, the date and time of purchase 520 are transmitted to the consumer survey system 500 via the purchase point record 510.

The consumer survey system 500 can access the consumer profile 560 to update the profiles contained therein. Consumer survey system 500
Retrieves the consumer demographic characterization vector 562 and the product preference vector 564. Following the search, one or more data processing algorithms are applied to update the vector. The algorithm for the update is shown in the flowchart of FIG. 8A. The updated vectors, now named as new demographic characterization vectors 566 and new product preferences 568, are returned to the consumer profile 560 for storage.

Consumer survey system 500 can determine probable consumer demographic characteristics based on product purchases by applying heuristic rules 519. Consumer survey system 500 provides product ID 514 to heuristic rule record 530 and receives the heuristic rules associated with the product. An example of a low discovery rule is shown in FIG.

In a preferred embodiment of the present invention, the consumer survey system 500 can determine the applicability of the advertisement to the consumer 100. For determining applicability of the advertisement, a correlation request 546 is received by the consumer survey system 500 from the advertisement record 540 along with the consumer ID 512. The advertising record 540 also provides advertising features including an advertising demographic vector 548, an advertising product category 552, and an advertising product preference vector 554.

The application of the cross-correlation process leads to a demographic correlation 556 and a product correlation 558 that can be returned to the advertisement record 540, as described according to FIG. 8B. In a preferred embodiment, advertiser 144 uses product correlation 558 and demographic correlation 556 to determine the applicability of the advertisement and to determine whether it is worth buying the opportunity. In a preferred embodiment, the pricing policy 570 is used by the consumer survey system 500 to store the ad record 54
It is used to determine an advertising price 572 that can be sent to zero.

The pricing policy 570 is for the consumer survey system 50 to obtain the advertisement price 572.
Accessed by 0. Pricing policy 570 is based on consumer survey system 500.
Takes into account the correlation results provided by An example of a pricing scheme is shown in FIG.

FIGS. 6A and 6B show pseudo code for the update process and for the correlation operation, respectively. The update process stores the consumer 100 in the form of a new demographic characterization vector 562 and a new product preference vector 568.
Includes leveraging purchase information along with heuristic rules to obtain a more accurate notation of.

As shown in the pseudo code in FIG. 6A, purchase point data is read and product purchases are integrated into the update process. Consumer survey system 500 uses heuristic rule 5
The product demographic vector obtained from the set of 19 is searched, and the consumer profile 5 is searched.
From 60, apply the product demographic vector to demographic characterization vector 562 and product preference vector 564.

The update process, as shown by the pseudo code in FIG. 6A, utilizes a weighting factor that determines the importance of purchasing that product for all members purchased in a particular product category. In a preferred embodiment, the weight is identified by the product ID 514 purchased by the consumer 100 identified by the consumer ID 512 and a certain identification purchased at that time relative to the total amount of the product purchased during the extended time period. Is calculated as the ratio of the total of products having the product ID 514. In a preferred embodiment, the extended time period is one year.

In a preferred embodiment, the product category total purchase is determined from a record that includes the number of times the consumer 100 has purchased the product identified by a particular product ID.

In alternative embodiments, other types of weighting factors, running averages, and statistical filtering techniques can be used to update the demographic characterization vector using the purchase data. The system can also reset to a clear past demographic characterization vector and product preference vector.

The new demographic characterization vector 566 is obtained as a weighted sum of the product demographic vector and the demographic characterization vector 562. The same procedure is performed to obtain a new product preference vector 568. Before storing these new vectors, normalization is performed on the same new vectors. As used herein, the term product characterization information refers to product demographic vectors, product purchase vectors, or heuristic rules, all of which can be used in the update process. What is a product purchase vector?
Points to a vector representing the purchase of the item represented by the product ID. As an example, 32
The product purchase vector for the purchase of an ounce sized Kellogg's cornflakes contains the identity value of the Kellogg's cornflakes and has a 32 ounce sized product purchase vector. In the renewal process, the total amount of purchase, as represented by the product purchase vector, is added to the product preference vector to update the product preference vector, and an estimate that the consumer will purchase a 32 oz size corn flake of Kellogg's Increase the probability of being done.

In FIG. 6B, the pseudo code of the correlation process is shown. After receiving the product characteristics and the consumer ID 512 from the advertisement record, the consumer survey system 500 retrieves the consumer demographic characterization vector 562 and its product preference vector 564. Demographic correlation is the correlation between the demographic characterization vector 562 and the advertising demographic vector. The product correlation is defined as an advertisement product preference vector 554.
And the product preference vector 564.

In a preferred embodiment, the correlation process involves the calculation of dot products between vectors. The resulting scalar is the correlation between the two vectors.

In an alternative embodiment, as shown in FIG. 10, basis vectors describing consumer aspects can be used to calculate the projection of advertising vectors on those basis vectors. In this embodiment, the result of the advertisement correlation may itself be in the form of a vector whose components indicate the degree of correlation with each consumer demographic feature or product preference feature of the advertisement. As shown in FIG. 10, the base vectors are consumer age 1021, consumer income 1001, and consumer family size 103.
It is one. The advertisement characterization vector 1500 represents the desired characteristics of the target audience and may include product preferences as well as demographic characteristics.

In this embodiment, the degree of orthogonality of the basis vectors will determine the uniqueness of the answer. The projections on the basis vectors form a set of data representing the corresponding values of the parameters measured on the basis vectors. As an example, if the household income is one basic vector, then the projection of the advertisement characterization vector onto the household income basic vector will return a result indicating the target household income of the advertisement.

Since the base vector cannot be easily created from some product preference categories (eg, cereal preferences), the product preference vector is shown in FIG. 2C, which represents the statistical average of cereal purchases in growing size containers. Alternative notations can be used for the ones shown. This vector can be interpreted as an average basis for cereals purchased by the consumer within a specified time period.

The individual measures of correlation represented by the cross-correlation vector can be used to determine the applicability of the advertisement to the subscriber, or the total correlation can be used to determine the total applicability of the advertisement. Can be generated to represent.

In a preferred embodiment, the individual measures of correlation, or the projection of the advertising feature vector to the consumer base vector, are not enabled to protect consumer privacy, only the absolute sum It has been reported. In geometric terms, this can be interpreted as a disclosure of the total length of the projection, rather than the actual projection itself.

In an alternative embodiment, the demographic and product preference parameters are used to form a set of scored pairs of elements of the consumer characterization vector paired with corresponding elements of the advertisement characterization vector. being classified. Correlation coefficients such as Pearson's product-moment correlation can be calculated. Other methods of correlation are available and are well known to those skilled in the art.

When the consumer and advertising characterization vectors are not in a standardized format, transformations can be performed to standardize the order of demographics and product preferences, or the data can be age, income or family size. Etc. can be decomposed into a set of basic vectors indicating a particular creation.

FIG. 7 shows an example of a heuristic rule that includes rules for defining a product demographic vector. Probable decisions on household demographics can be made from product characteristics. Similarly, the monthly amount purchased can be used to estimate the size of a household. The heuristic rules shown in FIG. 7 serve as examples of the types of heuristic rules that can be used to better characterize the consumer 100 as a result of their purchase. Heuristic rules are logical tests, statistical estimates,
Or it may include a market study that provides a basis for further estimating the demographics of the consumer 100 based on its purchase.

FIG. 8A depicts a flowchart for updating a consumer characterization vector. The system receives data from the purchase point at purchase information receiving point step 800. The system performs a test to see if the deterministic demographic characterization vector is available in the deterministic demographic information available step 810, and if not, to update the demographic features Proceed to.

Referring to FIG. 8A, in the purchase ID read step 820, the product ID 514 is read, and in the consumer demographic characterization vector update step 830, the purchase ID is read in FIG.
An algorithm, such as the algorithm represented in, is applied to obtain a new demographic characterization vector 566 that is stored in the consumer profile 560 in the update demographic characterization vector storage step 840.

The end test step 850 loops back to purchase ID information read 820 if all purchased products have not yet been processed for update and continue to branch to update product preference vector 564. be able to. In this branch, the purchased product is specified in the purchase ID information reading step 820. Product preference vector 5
An algorithm such as that shown in FIG. 6A is applied in a product preference vector update step 870 to update 64. The updated vector is stored in the consumer profile 560 in a store product preference vector step 880. This process is performed until all purchased items are integrated in the update process.

FIG. 8B shows a flowchart of the correlation process. At step 900, the advertising features described above in accordance with FIG. 5 along with the consumer ID are received by the consumer survey system 500. At step 910, demographic correlation 556 is calculated, and at step 920 product preference correlation 558 is calculated. An illustrative example of a correlation algorithm is presented in FIG. 6b. Prior to terminating the procedure at termination step 950, the system may provide demographic correlation 556 and product preference correlation 55.
8 back to the advertising record 540.

FIG. 9 shows two pricing schemes, one for content / opportunity provider 160 based pricing 970 that shows increasing costs as a function of correlation. In this low price scheme, the higher the correlation, the greater the amount charged by the content / opportunity provider 160 to broadcast the advertisement.

FIG. 9 illustrates consumer-based pricing 96 that allows consumers to charge less to receive ads that are more highly correlated to their demographics and interests.
0 is also indicated.

As an example of the industrial applicability of the present invention, the consumer 100 can use the consumer survey system 500 to purchase items at a grocery store that also serves as a profiler 140. The purchase record is used by the profiler to update the probabilistic notation of the consumer 100, both in terms of its demographics, as well as its product preferences. For each item purchased by the consumer 100, product characterization information in the form of a product demographic vector and a product purchase vector is used to update the consumer 100 demographic characterization vector and product preference vector. .

The content / opportunity provider 160 determines whether there is an opportunity to present an advertisement to the consumer 100 or thereafter. Content / opportunity provider 160 can announce this opportunity to advertiser 144 by sending details about the opportunity and consumer ID 512. The advertiser 144 can then query the profiler 140 by sending the consumer ID 512 along with the ad special removal information including the correlation request 546 and the ad demographic vector 548. The consumer survey system 500 performs a correlation to determine the degree to which the advertising target market is correlated with the estimated demographics and product preferences of the consumer 100. Based on this decision, advertiser 144 can decide whether to purchase the opportunity.

Although the invention has been described with respect to particular embodiments, it will be apparent to those skilled in the art that various deviations and modifications may be made that fall expressly within the scope of the invention. The invention is intended to be broadly protected within the spirit and scope of the appended claims.

[Brief description of the drawings]

FIG. 1A shows a user relationship diagram of the present invention.

FIG. 1B shows a user relationship diagram of the present invention.

FIG. 2A shows a consumer demographic characterization vector.

FIG. 2B shows a deterministic consumer demographic characterization vector.

FIG. 2C shows a consumer product preference characterization vector.

FIG. 2D shows a storage structure for consumer characterization vectors.

FIG. 3A shows an advertising demographic characterization vector.

FIG. 3B shows an advertising product preference characterization vector.

FIG. 4 shows a computer system capable of realizing the present invention.

FIG. 5 shows a context diagram of the present invention.

FIG. 6A shows a pseudo code update of a feature vector.

FIG. 6B shows a correlation operation.

FIG. 7 shows heuristic rules.

FIG. 8A shows a flowchart for updating a consumer characterization vector.

FIG. 8B shows a flowchart for updating a correlation operation.

FIG. 9 shows pricing as a function of correlation.

FIG. 10 shows the notation of consumer characterization as a set of base vectors and advertisement characterization vectors.

[Procedural Amendment] Submission of translation of Article 34 Amendment of the Patent Cooperation Treaty

[Submission date] June 30, 2000 (2000.6.30)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 17/30 340 G06F 17/30 340A H04N 7/173 610 H04N 7/173 610Z ZEC ZEC (31) Priority Claim No. 09 / 268,519 (32) Priority Date March 12, 1999 (March 12, 1999) (33) Priority Claimed States United States (US) (81) Designated States EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE), OA (BF, BJ, CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, SD, SL, SZ, TZ, UG, ZW), EA (A , AZ, BY, KG, KZ, MD, RU, TJ, TM), AE, AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, CN, CR, CU , CZ, DE, DK, DM, EE, ES, FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS, JP, KE, KG, KP, KR, KZ, LC, LK, LR, LS, LT, LU, LV, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, SI, SK, SL , TJ, TM, TR, TT, TZ, UA, UG, US, UZ, VN, YU, ZA, ZWF term (reference) 5B075 KK07 KK13 KK33 ND20 ND23 ND34 NS01 NS02 PQ05 PQ14 PR08 QP01 UU08 UU40 5C064 BC04 BC18 BC23 BC18 BD08 BD13 [Continued Summary] Can be used to increase performance and cost efficiency You.

Claims (37)

[Claims] 1. A method for surveying consumers based on consumer purchases, comprising:
(A) receiving the consumer's computer-readable detail purchase record, wherein the computer-readable detail purchase record has an associated unique consumer identification; and (b) product characterization. Retrieving information; (c) calculating a consumer characterization vector from the computer readable detail purchase record and the product characterization information; and (d) the consumer characterization on a computer readable medium. Storing the vector. 2. The method of claim 1, wherein the consumer characterization vector includes a demographic characterization of the consumer. 3. The method of claim 1, wherein the consumer characterization vector includes the consumer's product preference characterization. 4. A method for surveying consumer demographics based on consumer purchases, comprising: (a) receiving a computer readable detailed purchase record of a consumer, wherein the computer read A possible detail purchase record having an associated unique consumer identification, wherein the computer readable detail purchase record includes information regarding the purchase of an item; and (b) retrieving a demographic characterization vector. Wherein the demographic characterization vector is retrieved based on the associated unique consumer identification, wherein the demographic characterization vector includes probabilistic criteria for the consumer demographic feature. (C) retrieving a set of heuristic rules, wherein the set of heuristic rules is a demographic characteristic of a purchaser of the item. (D) calculating a new demographic characterization vector from the demographic characterization vector and the set of heuristic rules; and (e) the new demographic characterization vector. Storing on a computer readable medium. 5. The method of claim 4, wherein the set of heuristic rules is in the form of a product demographic vector. 6. The method of claim 5, wherein the new demographic characterization vector of step (d) is calculated as a weighted average of the product demographic vector and the demographic characterization vector. 7. The method of claim 6, wherein the weighted average is determined based on a weighting factor, wherein the weighting factor is determined as a ratio of a current product purchase to a cumulative product purchase. 8. A method for investigating consumer preferences based on a consumer purchase, comprising: (a) receiving a consumer's computer-readable detailed purchase record, wherein the consumer's computer-readable purchase record is provided. A detail purchase record having a unique consumer identification associated therewith, wherein the computer readable detail purchase record includes information regarding the purchased item; and (b) retrieving a product preference vector. Searching for the product preference vector based on the associated unique consumer identification, wherein the product preference vector includes criteria of conventional purchase characteristics by the consumer; and (c) the computer readable. Creating a product purchase vector from the detailed purchase record, wherein the product purchase vector comprises at least one of the purchased items. (D) calculating a new product preference vector from the product preference vector and the product purchase vector; and (e) storing the new product preference vector on a computer-readable medium. The method comprising: 9. The method of claim 8, wherein the new product preference vector of step (d) is calculated as an average of the product preference vector and the product purchase vector. 10. The method of claim 9, wherein the weighted average is determined based on a weighting factor, wherein the weighting factor is determined as a ratio of a current product purchase to a cumulative product purchase. 11. A data processing system for surveying consumers, said data processing system comprising: (a) computer processing means for processing data; and (b) storing data on a storage medium. Storage means for: (c) first means for retrieving a consumer characterization vector; and (d) second means for receiving a computer readable detail purchase record of the consumer, Second means having a unique consumer identification with which the detailed purchase record is associated; and (e) the consumer characterization based on creating a weighted sum of the product characterization information with the consumer characterization vector. A third means for updating the vector,
A third means in which the updated consumer characterization vector is stored on the storage device medium. 12. The data processing system of claim 11, wherein the consumer characterization vector includes a demographic characterization of the consumer. 13. The data processing system of claim 11, wherein the consumer characterization vector includes a product preference characterization of the consumer. 14. A computer program embodied in a computer readable medium for examining a consumer based on a purchase of at least one item, the computer program comprising:
(A) a consumer purchase source code segment for receiving a detailed consumer purchase record having a unique consumer identification with which the detailed consumer purchase record is associated; and (b) a consumer characterization vector. An update code segment for updating the consumer characterization vector based on creating a weighted sum of products of the product characterization information with the consumer characterization vector; and (c) the consumer A storage source code segment for storing the characterization vector on a computer-readable medium. 15. The method of claim 14, wherein the consumer purchase source code segment for receiving a detailed consumer purchase record supports receiving the associated unique consumer identification from a network interface. Computer program. 16. A method for determining applicability of an advertisement to a consumer in a networked environment having a plurality of computers interconnected for the purpose of transmitting and receiving data instantaneously. (A) receiving a computer-readable advertisement characterization vector at a first computer, wherein the computer-readable advertisement characterization vector includes a description of any feature corresponding to a target group of the advertisement; (B) receiving a unique consumer ID at the first computer; and (c) retrieving a consumer characterization vector from a storage device in the first computer based on the unique consumer ID. (D) between the computer readable characterization vector and the consumer characterization vector It said method comprising the steps of calculating a correlation factor, and sending the (e) the correlation factor from the first computer to the second computer. 17. The method of claim 16, wherein the consumer characterization vector comprises a demographic characterization of the consumer, and wherein the computer readable advertisement characterization vector comprises a demographic characterization of a target market for the advertisement. The described method. 18. The method of claim 16, wherein the consumer characterization vector includes the consumer's product preference characterization, and wherein the computer-readable advertisement characterization vector includes a product preference target market for the advertisement. . 19. The method of claim 16, wherein the correlation factor is calculated as a scalar product of the consumer characterization vector and the advertisement characterization vector. 20. The method of claim 16, further comprising: (f) determining a price for presenting the advertisement, wherein the price is a function of the correlation factor. 21. The method of claim 20, wherein said price is a monotonic function of increasing said correlation factor. 22. The method of claim 20, wherein said price is a decreasing monotonic function of said correlation factor. 23. A method of targeting a consumer of a particular product to an advertisement in a networked environment having a plurality of computers interconnected to send and receive data instantaneously, comprising: Receiving a computer-readable advertisement characterization vector at a first computer, wherein the advertisement characterization vector includes a description of a target market, and the description of the target market includes at least one indicator corresponding to a target product. (B) receiving a unique consumer ID at the first computer; and (c) retrieving a consumer characterization vector from a storage device in the first computer based on the unique consumer ID. Wherein the consumer characterization vector includes a plurality of purchase indicators, and wherein the purchase indicators are (D) calculating a correlation factor between the advertisement characterization vector and the consumer characterization vector; and (e) transmitting the correlation factor to a second computer. The above method, comprising: 24. The method of claim 23, further comprising: (f) determining a price for presenting the advertisement, wherein the price is a function of the correlation factor. 25. The method of claim 24, wherein the price is a monotonic function of increasing the correlation factor. 26. The method of claim 24, wherein the price is a decreasing monotonic function of the correlation factor. 27. A data processing system for determining applicability of an advertisement to a consumer and for determining a price for displaying the advertisement, comprising: (a) a computer for processing data; Processing means; (b) storage means for storing data on a storage medium; (c) first means for receiving the advertisement characterization vector corresponding to the advertisement, wherein the characterization vector corresponds to the advertisement; (D) second means for receiving a unique consumer ID; (e) third means for searching for a consumer characterization vector based on the unique consumer ID; A fourth factor for determining a correlation factor, wherein the correlation factor is determined from a correlation of the advertising vector with the consumer characterization vector.
Means, (g) fifth means for searching for a pricing function, and (h) sixth means for determining a previous term price for displaying the advertisement to the consumer, wherein the price is The data processing system, wherein the data processing system is determined from the correlation factor and the pricing function. 28. The data processing system of claim 27, wherein the consumer characterization vector includes a demographic characterization of the consumer. 29. The data processing system of claim 27, wherein the consumer characterization vector includes the consumer's product preference characterization. 30. The data processing system of claim 27, wherein said price is a monotonic function of increasing said correlation factor. 31. The method of claim 27, wherein the price is a decreasing monotonic function of the correlation factor. 32. A computer program embodied on a computer readable medium for determining applicability of an advertisement to a consumer, comprising: (a) an advertisement source code segment for receiving an advertisement characterization vector;
(B) a consumer characterization code segment for receiving a unique consumer ID and searching for a consumer characterization vector corresponding to the unique consumer ID; and (c) the advertisement characterization vector and the consumption. Correlating source code segments for calculating a correlation factor between character characterization vectors. 33. The computer program according to claim 32, further comprising: (d) a transmission source code segment for transmitting the correlation factor. 34. A method for constructing a subscriber characterization vector and for enabling access to a subscriber characterization vector without revealing special viewing habits, comprising: (a) at home Creating a record of the viewing habits of at least one subscriber; and (b) determining a subscriber characterization vector based on the recording of the viewing habits, wherein the subscriber characterization vector is identified. And c. Enabling access to the subscriber characterization vector to determine the applicability of the advertisement. 35. The method of claim 34, wherein accessing the subscriber characterization vector is performed by allowing the advertisement characterization vector to correlate with the subscriber characterization vector. 36. A method for constructing a consumer characterization vector and for enabling access to a consumer characterization vector without revealing a particular purchase, comprising: (a) purchasing (B) determining a consumer characterization vector based on the purchase;
Where the consumer characterization vector does not contain specific purchase information and (c) enabling access to the consumer characterization vector to determine the applicability of the advertisement; How to prepare. 37. The method of claim 36, wherein accessing the subscriber characterization vector is performed by allowing the advertisement characterization vector to correlate with the consumer characterization vector.