WO2002061619A1

WO2002061619A1 - Advertisement distribution system

Info

Publication number: WO2002061619A1
Application number: PCT/JP2001/005078
Authority: WO
Inventors: Tomohiko Yukawa; Akio Iijima; Makoto Abe
Original assignee: Dentsu Inc.
Priority date: 2001-02-01
Filing date: 2001-06-14
Publication date: 2002-08-08
Also published as: JP4942281B2; JPWO2002061619A1

Abstract

An advertisement distribution system for distributing advertisements predetermined respective times within periods predetermined for respective advertisers in consideration of the viewing intentions of the viewers and the advertisement distribution intentions of the advertisers. The advertisement distribution system for distributing advertisements to be distributed after weighting the advertisements according to the viewers comprises means for weighting each of the advertisements according to at least either the utility of the advertisers and the utility of the viewers.

Description

Advertisement distribution system

1. Technical Field

The present invention relates to the distribution of advertisements, and in particular, can distribute each advertisement a predetermined number of times while balancing, within a distribution period predetermined for each advertisement. The present invention relates to an advertisement distribution system and method for distributing an advertisement in consideration of a distribution intention, a program for realizing advertisement distribution, and a computer-readable recording medium recording the program.

2. Background technology

In television and radio broadcasts, advertisements are generally distributed integrally with the program being broadcast.However, recently, an advertisement distribution system that separates and distributes advertisements from programs has been proposed. It has become to.

The advertisement distribution system described above determines the broadcast priority by assigning different weights to the viewers for the advertisements stored on the transmission side or the reception side, and the determined broadcast priority is determined. It is configured to perform advertisement distribution based on the advertisement.

However, in the above-described conventional advertisement distribution system, the ad distribution order for each viewer is determined based on the advertiser's intention for the evening gate (that is, the advertiser wants to show the advertisement). Therefore, while meeting the advertiser's willingness to get evening ads, advertising placement is biased toward a particular evening get.

As a result, there has been a problem that each advertisement cannot be delivered a predetermined number of times within a delivery period predetermined for each advertisement. Another problem is that it is not possible to deliver advertisements taking into account the viewers' intentions (including interest, interest, etc.). 3. Disclosure of the Invention

The present invention has been made in view of the above-described problems in the conventional technology, and can distribute each advertisement a predetermined number of times while maintaining a balance within a distribution period predetermined for each advertisement. Another object of the present invention is to provide an advertisement distribution system capable of distributing an advertisement in consideration of an advertisement distribution intention of an advertiser.

In addition, the present invention, in view of the above-described problems in the related art, can distribute each advertisement a predetermined number of times while taking a notice within a distribution period predetermined for each advertisement. It is an object of the present invention to provide an advertisement distribution method capable of distributing an advertisement in consideration of the viewing intention and the advertiser's advertisement distribution intention.

Further, in view of the above-mentioned problems in the conventional technology, the present invention can distribute each advertisement a predetermined number of times while maintaining a balance within a distribution period predetermined for each advertisement. An object of the present invention is to provide a program for realizing advertisement distribution, which can distribute advertisements in consideration of the intention and the advertiser's advertisement distribution intention.

An object of the present invention is to provide a computer-readable recording medium on which a program for realizing the above-described advertisement distribution is recorded.

An object of the present invention is to provide an advertisement distribution system for distributing a plurality of advertisements to be distributed by weighting each of the plurality of advertisements according to each viewer, the utility of the advertiser and each viewer. This is achieved by an advertisement distribution system including means for weighting each of a plurality of advertisements based on at least one of the utilities. An object of the present invention is to provide an advertisement distribution system that distributes a plurality of advertisements by weighting each of the plurality of advertisements to be distributed according to each viewer. This is achieved by an advertisement distribution system that includes means for weighting each of a plurality of advertisements based on a planned distribution period.

The above-mentioned problem of the present invention is that each of a plurality of advertisements to be delivered is adapted to each viewer. A method of delivering at least one of the utility of the advertiser and the utility of each viewer, and the utility of the selected advertiser and the utility of each viewer. Weighting each of the plurality of advertisements based on at least one of the following.

An object of the present invention is to provide an advertisement distribution method for distributing a plurality of advertisements by weighting each of a plurality of advertisements to be distributed according to each viewer, and This is achieved by an advertisement distribution method including a step of setting a planned distribution period, and a step of weighting each of the plurality of advertisements based on the set planned number of distributions and the planned distribution period.

The object of the present invention is to provide a computer with a procedure for selecting at least one of the utility of the advertiser and the utility of each viewer, and the computer based on at least one of the utility of the selected advertiser and the utility of each viewer. Weighting each of a plurality of advertisements according to each viewer, thereby realizing distribution of the plurality of advertisements by weighting each of the plurality of advertisements to be distributed. Achieved by

The object of the present invention is to provide a computer with a procedure for setting a planned number of distributions and a planned distribution period of each advertisement, and a procedure for weighting each of a plurality of advertisements based on the set planned number of distributions and the planned distribution period. This is achieved by a program for realizing distribution of a plurality of advertisements by executing and weighting each of a plurality of advertisements to be distributed according to each viewer.

The object of the present invention is to provide a computer with a procedure for selecting at least one of the utility of the advertiser and the utility of each viewer, and the computer based on at least one of the utility of the selected advertiser and the utility of each viewer. Weighting each of a plurality of advertisements according to each viewer, thereby realizing distribution of the plurality of advertisements by weighting each of the plurality of advertisements to be distributed. This is achieved by a computer-readable recording medium on which is recorded. An object of the present invention is to execute a procedure of setting a planned number of distributions and a planned distribution period of each advertisement in a computer, and a step of weighting each of the plurality of advertisements based on the set planned number of distributions and the planned distribution period. This is achieved by a computer-readable recording medium storing a program for realizing distribution of a plurality of advertisements by weighting each of a plurality of advertisements to be distributed according to each viewer. You.

In the advertisement distribution system of the present invention, when the weighting means weights each of the plurality of advertisements based on the utility of the advertiser, the degree of matching between the attribute of the viewer specified by the advertiser and the attribute of each viewer Depending on the condition, a plurality of advertisements may be weighted.

In the advertisement distribution system of the present invention, the attributes of the viewer are: a) gender of the viewer, b) age of the viewer, c) date of birth of the viewer, d) address or location of the viewer, e) viewer At least one of: f) viewer's household attribute, g) viewer's occupation, h) viewer's monthly or annual income, i) viewer's hobby, j) viewer's preference. May be configured.

In the advertisement distribution system according to the present invention, when the weighting means weights each of the plurality of advertisements based on the utility of the advertiser, P _ik indicates a category weight for one of the i-th category of the attribute k, and y _ik but the i-th category of viewers attribute k. indicates whether or not they belong to Lee, the formula shows the attribute weights w _k is set to attribute ^{k ( ") = ππ ^ ik} (V

The weighting may be performed by k≡≡.

In the advertisement distribution system of the present invention, when weighting each of the plurality of advertisements based on the utility of the viewer, the weighting means includes a field of a hobby or a preference designated by the viewer and A plurality of advertisements may be weighted according to the degree of matching with the field to which the advertisement belongs.

In the advertisement distribution system according to the present invention, the weighting means, when weighting each of the plurality of advertisements based on the utility of the viewer, weights the plurality of advertisements so that continuous viewing of the same advertisement can be adjusted a plurality of times. You may comprise.

In the advertisement distribution system of the present invention, when weighting each of a plurality of advertisements based on the utility of the viewer, the weighting means weights the plurality of advertisements so that the number of views of the same advertisement within a predetermined period can be adjusted. It may be configured as follows.

In the advertisement distribution system of the present invention, the weighting means recognizes a word, a voice, or an image expressed in a program watched by the viewer, and weights a plurality of advertisements according to the recognized word, the voice, or the image. May be configured.

In the advertisement distribution system of the present invention, when the weighting means weights each of the plurality of advertisements based on the utility of the advertiser, P i (n) indicates a category one weight in the i-th interest category of the advertisement n. , Z i indicates 1 if the viewer belongs to the i-th interest category, 0 if it does not, apart (n) indicates the interval since the last viewing of advertisement n, f (apart (n)) is a function of apart (n)

Uv (n) =

May be configured to perform weighting.

In the advertisement distribution system of the present invention, the weighting means determines the remaining amount at any point in time based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period that is the number of days until the distribution end date. A configuration may be adopted in which a distribution quota, which is the number of distributions per day in a distribution period, is calculated for each advertisement, and a plurality of advertisements are weighted according to the degree of the calculated distribution quota. In the advertisement distribution system according to the present invention, the weighting means calculates the planned distribution period based on the planned distribution frequency, the planned roaring period, the distribution frequency, and the remaining distribution period that is the number of days until the distribution end date. Calculated for each advertisement is the planned distribution quota, which is the number of distributions per day, and the distribution quota, which is the number of distributions per day for the remaining distribution period at any point in time. It may be configured to calculate the degree of deviation from planned based on the distribution quota at an arbitrary point in time, and weight a plurality of advertisements according to the calculated degree of degree of deviation from planned.

In the advertisement distribution system of the present invention, the weighting means determines the remaining amount at any point in time based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period that is the number of days until the distribution end date. Calculate the distribution quota, which is the number of distributions per day for the distribution period, for each advertisement, weight multiple ads according to the calculated distribution quota, and distribute the daily distribution for the planned distribution period The planned distribution quota, which is the number of times, and the distribution quota, which is the number of distributions per day, for the remaining distribution period at any time, are calculated for each advertisement, and the planned distribution quota and distribution at any time are performed at any time. The degree of divergence from the plan may be calculated based on the quota, and a plurality of advertisements may be weighted according to the calculated degree of divergence from the plan.

In the advertisement distribution method of the present invention, the step of weighting includes, when weighting each of a plurality of advertisements based on the utility of the advertiser, determining the degree of coincidence between the attribute of the viewer specified by the advertiser and the attribute of each viewer. The method may include a step of determining and a step of weighting a plurality of advertisements according to the result of the determination.

In the advertisement distribution method of the present invention, the attributes of the viewer are: a) gender of the viewer, b) age of the viewer, c) date of birth of the viewer, d) address or location of the viewer, e) viewer. D) the type of residence, f) the household attributes of the viewer,) the occupation of the viewer, h) the monthly or annual income of the viewer, i) the hobby of the viewer, and j) the preference of the viewer. No. In the advertisement distribution method of the present invention, the step of weighting includes, when weighting each of a plurality of advertisements based on the utility of the viewer, the degree of coincidence between the field of the hobby or preference specified by the viewer and the field to which each advertisement belongs. And a step of weighting a plurality of advertisements according to the result of the determination.

In the advertisement distribution method according to the present invention, the weighting step includes, when weighting each of the plurality of advertisements based on the utility of the viewer, weighting the plurality of advertisements so that continuous viewing of the same advertisement can be adjusted a plurality of times. It may be.

In the advertisement distribution method of the present invention, the weighting step includes, when weighting each of a plurality of advertisements based on the utility of the viewer, weighting the plurality of advertisements so that the number of times of viewing the same advertisement within a certain period can be adjusted. You may make it.

In the advertisement distribution method of the present invention, the step of weighting includes recognizing words, sounds, or images expressed in a program being viewed by a viewer, and performing a plurality of operations in accordance with the recognized words, sounds, or images. The advertisement may be weighted.

In the advertisement distribution method of the present invention, the weighting step is performed based on the planned number of distributions, the planned roster S communication period, the number of distributions already performed, and the remaining distribution period that is the number of days until the distribution end date. The method may include a step of calculating a distribution quota, which is the number of distributions per day, of the remaining distribution period at the time, for each advertisement, and a step of weighting a plurality of advertisements according to the calculated distribution quota. It may be.

In the advertisement distribution method of the present invention, the weighting step is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period that is the number of days until the distribution end date. Calculating, for each advertisement, a planned distribution quota, which is the number of distributions per day, and a distribution quota, which is the number of distributions per day, for the remaining distribution period at any time; The method may include a step of calculating a degree of divergence from the plan based on the distribution quota at an arbitrary point in time, and a step of weighting a plurality of advertisements according to the calculated degree of divergence from the plan. In the advertisement distribution method of the present invention, the weighting step is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period, which is the number of days until the distribution end date, for each advertisement. Calculating a delivery quota, which is the number of deliveries per day for the remaining delivery period, for each advertisement; weighting a plurality of advertisements according to the calculated delivery quota; Calculating the distribution quota, which is the number of distributions per day, and the distribution quota, which is the number of distributions per day for the remaining distribution period at any point in time, for each advertisement; and And calculating a degree of divergence from the plan based on the distribution quota at the point of, and weighting a plurality of advertisements according to the degree of the degree of divergence from the plan. There.

In the program of the present invention, the weighting step is to determine the degree of coincidence between the attributes of the viewer specified by the advertiser and the attributes of each viewer when each of the plurality of advertisements is weighted based on the utility of the advertiser. It may be configured to include a procedure and a procedure for weighting a plurality of advertisements according to the result of the determination.

In the program of the present invention, the attributes of the viewer are: a) gender of the viewer, b) age of the viewer, c) date of birth of the viewer, d) address or residence of the viewer, e) residence of the viewer. Configuration, f) viewer's household attribute, g) viewer's occupation, h) viewer's monthly or annual income, i) viewer's hobby, j) viewer's preference. May be.

In the program of the present invention, the weighting step is to determine the degree of coincidence between the field of the hobby or preference designated by the viewer and the field to which each advertisement belongs when weighting each of a plurality of advertisements based on the utility of the viewer. It may be configured to include a procedure for determining and a procedure for weighting a plurality of advertisements according to the result of the determination.

According to the program of the present invention, the weighting procedure can adjust the continuous viewing of the same advertisement a plurality of times when each of a plurality of advertisements is weighted based on the utility of the viewer. A plurality of advertisements may be weighted as described above.

According to the program of the present invention, the weighting step is such that, when weighting each of a plurality of advertisements based on the utility of the viewer, the plurality of advertisements are weighted so that the number of times of viewing the same advertisement within a certain period can be adjusted. It may be configured as follows.

In the program of the present invention, the weighting step includes recognizing words, sounds, or images expressed in the program being viewed by the viewer, and a plurality of advertisements according to the recognized words, sounds, or images. May be configured to be weighted.

In the program of the present invention, the weighting procedure is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period, which is the number of days until the distribution end date, for each advertisement. And a procedure for calculating a distribution quota, which is the number of distributions per day, for each advertisement, and a procedure for weighting a plurality of advertisements in accordance with the degree of the calculated distribution quota. Good.

In the program of the present invention, the weighting procedure is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period, which is the number of days until the distribution end date, for each advertisement. The distribution quota, which is the number of distributions for each advertisement, and the distribution quota, which is the number of distributions per day for the remaining distribution period at any time, for each advertisement, and the planned distribution quota for each time And a procedure for calculating the degree of divergence from the plan based on the distribution quota at the arbitrary point in time, and a procedure for weighting a plurality of advertisements according to the calculated degree of divergence from the plan. Is also good.

In the program of the present invention, the weighting procedure is performed at any time based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period that is the number of days until the distribution end date. A procedure for calculating a delivery quota, which is the number of deliveries per day for the remaining delivery period, for each advertisement; a procedure for weighting a plurality of advertisements according to the calculated delivery quota; Planned distribution, which is the number of distributions per day. A procedure for calculating a luma for each advertisement, a procedure for calculating a planned distribution quota at each time and a distribution quota at the time, and a calculated degree of deviation from the plan. And weighting a plurality of advertisements in accordance with

4. Brief description of drawings

FIG. 1 is a schematic block diagram showing a configuration of an advertisement distribution system according to an embodiment of the present invention;

FIG. 2 is a flowchart for explaining the operation of the advertisement distribution system of FIG. 1; FIG. 3 is a flowchart for explaining the operation of the advertisement distribution system of FIG.

5 BEST MODE FOR CARRYING OUT THE INVENTION

Here, before describing an embodiment of the advertisement distribution system of the present invention, the basic principle of the advertisement distribution system of the present invention will be described.

(1) First, the advertisement distribution system of the present invention is configured to distribute a different advertisement for each viewer without necessarily depending on the program being viewed.

Here, “distribution” means not only transmitting an advertisement from a transmitting station such as a TV station, but also storing the information in a storage device such as a hard disk connected to or built into a television receiver on the viewer side. Means that the advertisement stored in the storage device connected to the relay device on the transmitting side is transmitted from the storage device to the television receiver. .

(2) The advertisement distribution system of the present invention weights each advertisement for each viewer in order to distribute different advertisements for each viewer, and distributes the advertisement according to the weight assigned to each advertisement. It is configured to create a ranking list and distribute each advertisement to viewers based on the created distribution ranking list.

(3) For each viewer, for each advertisement that may be delivered to that viewer, weight each of the advertisements. That is, the total weight W (n) of any advertisement n is Weight each of a plurality of advertisements based on at least one of the utility of the advertiser and the utility of each viewer, and write the weight based on the weight 記 Wl (n) and the planned number of distributions, the planned distribution period, and the distribution quota of each advertisement Weighting each of multiple advertisements, including weight ② W2 (n), represented by the following equation:

W (n) = F (Wl (n), W2 (n)) (A)

Here, it should be noted that the weight ① Wl (n) and the weight ② W2 (n) do not affect each other, they are independent weights, and these two different weights are combined. In this way, the system is designed so that the distribution of advertisements can be controlled in consideration of the utility of the advertiser and the audience, and the schedule of each advertisement.

As described in detail below, the weight ① Wl (n) is expressed as a function using one or both of the advertiser's utility U a (n) and the viewer's utility Uv (n) . Chi

Wl (n) = F (Ua (n), Uv (n)) (B)

Becomes '

On the other hand, the weight ② W2 (n) has two different weighting methods:

(a) The distribution quota of each advertisement (= remaining number of distributions / remaining distribution period) is calculated as a reference value, and each advertisement is weighted in descending order of the reference value.

(b) Based on the degree of discrepancy between the planned delivery quota (= planned delivery count / planned delivery period) of each advertisement and the delivery quota at any time (= remaining delivery count / remaining delivery period) (vs. planned divergence) This is calculated as a value, and each advertisement is weighted in descending order of the reference value. The method (b) enables more autonomous control than the control of only the distribution quota of each advertisement shown in (a).

That is, in the above method (a), it is expected that the distribution quota may rub against the planned distribution quota in the relative relationship of each advertisement. Even if they do not depend on their relative relationship with the notification, they will autonomously rub into the planned distribution quota.

Here, the difference between the method (a) and the method (b) will be specifically described. If there are two types of advertisements, Ad A with an extremely strict planned delivery quota and Ad B with an extremely strict planned delivery quota, according to the above method (a), the overwhelming difference in the delivery quota In this case, only the advertisement A is delivered, and after the delivery period of the advertisement A ends, the delivery of the advertisement B starts. On the other hand, according to the method (b) above, despite the overwhelming difference in the distribution quota between the advertisement A and the advertisement B, the weight is determined based on the degree of divergence between the advertisements A and B with respect to the plan. Therefore, even if the distribution quota of Ad B is loose, Ad B will be delivered if the degree of divergence from Ad A is more than planned. As a result, the bias is reduced as compared with the method (a).

The delivery quota (-number of remaining deliveries / remaining delivery period) may be replaced with the actual delivery results (= number of deliveries / already delivered period).

Further, as the weight ② W 2 (n), a method in which the method according to the above (a) is modified by the method according to the above (b) can be considered.

Hereinafter, preferred embodiments of an advertisement distribution system and method according to the present invention will be described in detail with reference to the drawings.

FIG. 1 is a schematic block diagram showing a configuration of a preferred embodiment of an advertisement distribution system (hereinafter, abbreviated as a system) according to the present invention.

The system 100 in FIG. 1 includes a viewer attribute information input unit 1 for inputting attribute information such as a viewer profile, a viewer attribute database for storing viewer attribute information, and a viewer attribute database 2. A viewing history database 3 for accumulating the viewing history of advertisements, an advertisement distribution condition input section 4 for inputting advertisement distribution conditions 4, an advertisement information database 5 for registering information related to advertisements, and an advertisement information database For registered ads Ad distribution learning judgment unit 6, which has the function of learning the distribution results and the function of judging whether or not to distribute the advertisement, the advertiser utility calculation unit 7, which calculates the advertiser utility of the advertisement, the viewer utility Audience utility calculation unit 8 for calculating the weight, weight ① calculation unit for calculating the weight ①, planned distribution quota calculation unit 10 for calculating the planned distribution quota 10, distribution quota for calculating the distribution quota Calculating unit 11, Pair calculation divergence calculating unit 12 for calculating planned divergence, weight for calculating weight ② calculating unit 13, weight ① weight で calculated by calculating unit 9 and weight (2) The total weight calculation unit (14) that calculates the total weight based on the weight (1) calculated by the calculation unit (13), and the correction value of the total weight is adjusted based on the distribution results of each advertisement learned by the advertisement distribution learning judgment unit (6). , To enter the adjusted value of the adjusted total weight Modification unit 15; Ad distribution order list creation unit 16 to create an ad distribution order list; Ad distribution instruction unit 17 to issue an ad distribution instruction; Advertisement to accumulate advertisement content Server 18, program server 19 for storing program content, program for viewing programs and advertisements, advertisement viewing section 20, words and images expressing the content of the delivered program in the program A program content recognition unit 21 that recognizes by recognizing a program, and an advertisement viewing confirmation unit 22 that confirms the viewer's viewing of an advertisement 2 2 s A content production system 2 for producing program and advertisement contents 2 Has three.

Furthermore, in the present embodiment, the content production system 23 is configured to produce content for programs and advertisements.

Next, the operation of the system 100 will be described in detail based on the operation of the above components.

The viewer inputs viewer attribute information such as a profile of the viewer himself / herself via the television screen to be viewed by the viewer attribute input unit 1 (step S 1). When the attribute information is input, the viewer attribute information is stored in the viewer attribute database 2 (step S2). The advertiser inputs distribution conditions in accordance with the characteristics of the product to be advertised via a terminal or the like through the advertisement distribution condition input section 4 (step S3), and the advertisement to be distributed inputted from the advertisement distribution condition input section 4 Information (advertisement distribution information) is registered and stored in the advertisement information database 5 (step S4).

Tables 1 and 2 show examples of information registered in the advertisement information database 5.

table 1

Table 2

Following the above step S4, the matching between the advertisement distribution conditions input by the advertiser from the advertisement distribution condition input section 4 and the viewer attribute information stored in the viewer attribute database 2 is performed. Then, the advertiser utility U a (n) is calculated by the advertiser utility calculation unit 7 according to the following equation (1) for all the advertisements n that may be delivered to each of the viewers (step S 5) o

Advertiser utility U a (n)

Four H) = Π)

k ≡Π ^ ("(

k ¹ ). Here, P _ik is set by the advertiser and indicates that the category weight is for the i-th category one of the attribute k, and y _ik is the category weight for the i-th category one of the attribute k. Indicates whether the user belongs to the power category.If the user belongs to the power category, y _ik = 1; if not, y _ik = 0 and w _k is set to the attribute k. Indicates attribute weight. In addition, I _k indicates a set of all segment ranges in the attribute k.

Here, the category indicates a range for dividing the attribute.

From the above formula, the advertiser's utility U a (n) for an advertisement n for an arbitrary viewer is represented by the attribute weight w _{k to the} power of the category 'weight P _ik (n) of the category to which the viewer belongs. I can understand that

In the above equation (1), P _ik (n) represents the weight of the i-th power category of the attribute k regarding the advertisement n,

(2)

Satisfy the relationship. That is, the weight P _ik (n) of each category in an arbitrary attribute is allocated and set so that the total number of all the categories included in the attribute is one.

Table 3

Attribute = Age, Attribute Weight = 3, Category 1 to 7, and Category 10 are set as shown in Table 3 above, for example, if you are in your 20s, y ₂ Age = 1 and 10s to 70s excluding those in their twenties, hoof = y ₃ years _ft = y ₄ age = y ₅ age = y ₆ age = y? Therefore, the value of p _ik is as shown in Table 4. Table 4

Following step S5, the viewing history (viewing status) of the advertisement for each viewer is stored in the viewing history data 3 (step S6), and stored in the viewing history data pace 3. Information (that is, information about the advertisement viewing history for each viewer = information about when and which advertisement each viewer watched in the past) In addition to calculating the elapsed time, the content of the program being viewed by the viewer recognized by the program content recognition unit (step S7) and the input from the advertisement distribution condition input unit 4 by the advertiser are input. Based on the matching of the advertisement distribution conditions and the viewer attribute information stored in the viewer history database 3, the viewer utility U v for all advertisements n that may be delivered to each viewer (n) is calculated by the viewer utility calculation unit 8 according to the following equation (3) (step S8).

The viewer utility U v (n) is

Uv (ii) = fi (") ^Zi xf (apart (n)) (3) In other words, the utility U v (n) of each viewer for an arbitrary advertisement n belongs to the advertisement n Calculated based on the category one weight set for the category of interest, whether each viewer is interested in which category of advertisement, and the elapsed time since the viewer last viewed advertisement n Is done.

That is, P i (n) is a category one in the i-th interest category one of the advertisement n. Further, if the viewer is interested in the i-th interest category 1, l is set, and if the viewer is interested, Z i = 0. Therefore, the category one weight P i (n) is derived only for advertisements in categories in which viewers are interested. Then, the category one way (n) of the advertisement that the viewer is not interested in becomes all ones by being multiplied.

In other words, for any advertisement n, the weight sum of the categories in all the interest categories is set to be 1. Here, Z i is 1 when the viewer belongs to the i-th interest category, 0 when it does not belong, and apart (n) represents the interval since the last viewing of advertisement n. , F (apart (n)) represent coefficients that are a function of apart (n). In addition, I _k indicates a set of all division ranges in the attribute k. In the above equation (3), (n) represents ゥ eight of the i-th interest category of advertisement n,

Satisfy the relationship.

Subsequent to step S8 above, based on the advertiser utility calculated by the advertiser utility calculation unit 7 and the viewer utility calculated by the viewer utility calculation unit 8, the weight / calculation unit 9 assigns a weight to each viewer. The weight ① (= advertiser utility X viewer utility) is calculated (step S 9), and the delivery quota is calculated by the delivery quota calculator 11 from the remaining delivery period and the remaining delivery count based on the delivery status of each advertisement. (Step S 10) Further, based on the planned delivery quota calculated by the planned delivery quota calculation unit 10 and the delivery quota calculated by the delivery quota calculation unit 11, the pair is calculated by the planned divergence calculation unit 12. The planned divergence is calculated (step S 11), and the weight is calculated based on the distribution quota calculated by the distribution quota calculator 11 and the planned divergence calculated by the planned divergence calculator 12. (2) The weight ② is calculated by the calculation unit 13 (step S 1 2).

Table 5 shows an example of a basic table for advertisements A to D, and Table 6 shows an example of weight ① and weight ② tables for advertisements A to D, respectively.

Table 5 Advertisement A Advertisement B Advertisement C Advertisement D Viewer①2: 23: 21: 21: 1: 5 Viewer②3: 22: 35: 21: 2 only Viewer3: 1: 1: 4: 2 1: 2: 3

① Audience ② 2: 3 1: 25: 1 2: 1

Audience ⑤ 2: 5 2 ： 3 1 ： 1 1 ： 4 Remaining days 2 3 5 1 Remaining times 8 1 8 2 0 5 Weight ② 4 6 4 5 * Advertiser utility: Viewer utility Table 6

* Weight Advertiser Utility X Viewer Utility

* Weight Delivery quota = number of remaining times Z remaining days

Here, the weight is set to advertiser utility X viewer utility, but this is only an example, and other functions based on advertiser utility and viewer utility, weight (D = advertiser utility + viewer utility, etc. As the weight ②, the weight ②d the number of already-distributed times Z already-distributed period may be used.

The total weight calculation unit 14 calculates the total weight W (n) by the following equation (5) based on the weight ①the weight で calculated by the calculation unit 9 and the weight ②the weight calculated by the calculation unit 13②. (Step S13) The weight correction unit 15 corrects the total weight W (n) based on the correction value, and the total weight calculation unit 14 corrects the corrected total weight W, (n) is calculated (step S14).

The total weight W (n) is

W (n) = Wl (n) +2 (n) (5)

Is defined as That is, the total weight W (n) of an advertisement n for a given viewer is the weight derived from the advertiser utility (for that viewer) and the viewer utility (for advertisement n). It is expressed as the product of ① Wl (n) and the weight より W2 (n) derived from the distribution quota <

Table 7 shows an example of a total weight table for advertisements A to D. Table 7

Also, in this embodiment, the total weight = weight① + weight②, but the total weight = weight

① X weight X can also be set.

Subsequent to the above step S14, referring to the corrected total weight of each advertisement for each viewer calculated by the total weight calculation unit 14, a list of advertisements to be delivered to each viewer in the advertisement distribution order It is created by the list creation unit 16 (step S15).

Table 8 shows an example of the distribution order for advertisements A to D. Table 8

Distribution ranking table

In the example shown in Table 8, for viewer ①, ad B → ad D → ad A → ad C,

For ②, Ad B → Ad Ad C Ad D, For viewer ③, Ad B → Ad]) Advertisement C → Advertisement A, Advertisement A is delivered to viewer ④, Advertisement A → Advertisement C → Advertisement B → Advertisement D, and viewer 視聴 is advertised Ad B → Ad D ~> The advertisement is delivered in the order of the advertisement C.

Subsequently, the advertisement distribution instructing unit 17 issues an advertisement distribution instruction to the advertisement server 18 in accordance with the list created by the advertisement distribution ranking list creating unit 16 (step S16).

The advertisement content produced by the content production system 23 is stored in the advertisement server 18 (step S 17), and the order of the list created by the advertisement distribution order list creation unit 16 by the program's advertisement viewing unit 20 is set. In addition to allowing the viewer to view the advertisement distributed according to the above, information indicating which advertisement is distributed while the television is turned on is returned to the advertisement viewing confirmation unit 22 (step S18).

The advertisement viewing check unit 22 checks the advertisements actually delivered to the viewer (confirmation of the distribution status), and also stores information on the check in the advertisement information database 5 and the viewing history database. Send it to 3 (step S 19).

The advertisement delivery learning determination unit 6 calculates the number of remaining deliveries and the remaining delivery period for the advertisement n (step S20), and when the number of remaining deliveries calculated in step S20 above is equal to or less than 0 (zero). Then, the delivery of the advertisement n is finished (attainment of the goal) (step S21). If the number of remaining deliveries calculated in step S19 is 1 or more and the remaining delivery period is 0 (zero), an instruction is issued to the advertisement delivery ranking list creation unit 16 to place an advertisement at the top of the list. Insert n. (Step S22). Further, when the number of remaining deliveries calculated in the above step S20 is 1 or more and the remaining delivery period is 1 or more, the above steps S1 to S22 are repeated (step S23).

The advertisement distribution learning determination unit 6 has a learning function. This learning function is based on, for example, a “genetic algorithm”, and is used by the weight correction unit 15 to calculate a weight correction value. Used.

The above-described method of the present invention as described above provides a program for operating the above-described system. Can be configured as a system.

Further, such a program may be configured as a recording medium on which the program is recorded and which can be read by a computer.

As described above, the advertisement distribution method of the present invention aims to achieve the objectives of the advertiser, the viewer, and the advertising company by performing two types of independent weighting operations of weight ① and weight ②. Things. The purpose of weight ① is to register the viewer attributes, hobbies, preferences, etc. of each viewer in advance, and to determine the distribution order of advertisements. , And the advertisement can be sequentially distributed based on this. In addition, the purpose of the weight is to avoid distributing the advertisement to a specific audience (evening gate) or at a specific time, and to distribute the advertisement a predetermined number of times within a predetermined period. It is possible to achieve the purpose of the company. The present invention may be embodied in various other forms without departing from its spirit or essential characteristics. Therefore, the above-described embodiment is merely an example in every respect, and should not be construed as limiting. The scope of the present invention is defined by the appended claims, and is not restricted by the specification. Furthermore, all modifications and changes belonging to the equivalent scope of the claims are within the scope of the present invention.

Claims

The scope of the claims

1. An advertisement distribution system that distributes a plurality of advertisements by weighting each of the plurality of advertisements to be distributed according to each viewer,

An advertisement distribution system comprising: means for weighting each of the plurality of advertisements based on at least one of a utility of an advertiser and a utility of each viewer.

2. The weighting means, when weighting each of the plurality of advertisements based on the utility of the advertiser, according to the degree of matching between the attribute of the viewer specified by the advertiser and the attribute of each viewer. The advertisement distribution system according to claim 1, wherein the plurality of advertisements are weighted.

3. The attributes of the viewer are

a) the gender of the viewer, b) the age of the viewer, c) the date of birth of the viewer, d) the address or residence of the viewer, e) the type of residence of the viewer, f) the household attributes of the viewer, g) The method according to claim 1 or 2, wherein the job includes at least one of: occupation of the viewer, h) monthly or annual income of the viewer, i) viewer's hobby, j) viewer's preference. Advertising distribution system.

4. The weighting means, when weighting each of the plurality of advertisements based on the utility of the advertiser, P _ik indicates a category one weight for the i-th category of the attribute k, and y _ik indicates the attribute k of the viewer. 2. The advertisement delivery system according to claim 1, wherein whether or not the item belongs to the i-th category is determined, and w _k is weighted by an expression ik V indicating an attribute weight set to the attribute k.

5. The weighting means, when weighting each of the plurality of advertisements based on the utility of the viewer, determines a field of a hobby or a preference designated by the viewer and a field to which each advertisement belongs. The advertisement distribution system according to claim 1, wherein the plurality of advertisements are weighted according to a degree of coincidence between the advertisements.

6. The weighting means, when weighting each of the plurality of advertisements based on the utility of the viewer, weights the plurality of advertisements so that continuous viewing of the same advertisement can be adjusted a plurality of times. The advertisement delivery system described in Claim 1 or Claim 5

7. The weighting means, when weighting each of the plurality of advertisements based on the utility of the viewer, weights the plurality of advertisements so that the number of views of the same advertisement within a certain period can be adjusted. 7. The advertisement distribution system according to claim 1, wherein the advertisement distribution system according to any one of claims 1, 5, and 6 is characterized by:

8. The weighting means recognizes words, sounds, or images expressed in a program viewed by a viewer, and weights a plurality of advertisements according to the recognized words, sounds, or images. The advertisement distribution system according to any one of claims 1, 5, and 6.

9. When the weighting means weights each of the plurality of advertisements based on the utility of the advertiser, P i (n) indicates a category weight in the i-th interest category of advertisement n, and Z i indicates 1 indicates that the viewer belongs to the i-th interest category, 0 indicates that the viewer does not belong, and apart (n) indicates the interval from the last time the ad n was viewed.f (apart (n 2. The advertisement distribution system according to claim 1, wherein the weighting is performed by an expression (V) that represents a coefficient in which)) is a function of apart (n).

10. An advertisement distribution system for distributing a plurality of advertisements by weighting each of a plurality of advertisements to be distributed according to each viewer,

Each of the plurality of advertisements is calculated based on the planned number of distributions of each advertisement and the planned distribution period. An advertisement distribution system comprising a weighting unit.

1 1. The weighting means calculates one of the remaining distribution periods at any time based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period which is the number of days until the distribution end date. The advertisement distribution according to claim 10, wherein a distribution quota, which is the number of distributions per day, is calculated for each advertisement, and the plurality of advertisements are weighted according to a degree of the calculated distribution quota. system.

1 2. The weighting means, based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period that is the number of days until the distribution end date, for each advertisement, the number of distributions per day in the planned distribution period For each advertisement, the planned distribution quota and the distribution quota, which is the number of distributions per day for the remaining distribution period at an arbitrary point in time, are calculated for each advertisement. 10. The advertisement distribution system according to claim 10, wherein a degree of planned deviation is calculated based on the distribution quota, and the plurality of advertisements are weighted according to the calculated degree of deviation from plan. .

13. The weighting means calculates one of the remaining distribution periods at any time based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period which is the number of days until the distribution end date. The distribution quota, which is the number of distributions per day, is calculated for each advertisement, and according to the calculated distribution quota, the plurality of advertisements are weighted, and the number of distributions per day in the planned distribution period is calculated. For each advertisement, a planned distribution quota and the distribution quota, which is the number of distributions per day for the remaining distribution period at any time, are calculated for each advertisement, and the planned distribution quota and the distribution at any given time 10. The advertisement distribution system according to claim 10, wherein a degree of divergence from the plan is calculated based on the quota, and the plurality of advertisements are weighted according to the calculated degree of divergence from the plan. Beam.

14. An advertisement delivery method for delivering a plurality of advertisements by weighting each of a plurality of advertisements to be delivered according to each viewer, Selecting at least one of the utility of the advertiser and the utility of each viewer; and weighting each of the plurality of advertisements based on at least one of the utility of the selected advertiser and the utility of each viewer.

An advertisement distribution method, comprising:

15. The step of weighting includes, when weighting each of the plurality of advertisements based on the utility of the advertiser, determining the degree of agreement between the attribute of the viewer specified by the advertiser and the attribute of each viewer. Judgment stage,

15. The advertisement distribution method according to claim 14, further comprising a step of weighting the plurality of advertisements according to a result of the determination.

1 6. The attributes of the viewer are

a) the gender of the viewer, b) the age of the viewer, c) the date of birth of the viewer, d) the address or residence of the viewer, e) the type of residence of the viewer, f) the household attributes of the viewer, g) Claim 14 or 15 characterized by including at least one of: occupation of the viewer, h) viewer's monthly or annual income, i) viewer's hobby, j) viewer's preference. ad delivery method _c 1 7 according. step of said weighting if weighting it it of the plurality of advertisements based on the utility of the viewer, and the field of advertising each and fields of hobbies and preferences that specify the viewer belongs Determining the degree of coincidence of

15. The advertisement distribution method according to claim 14, further comprising: weighting the plurality of advertisements according to a result of the determination.

18. The step of weighting includes, when weighting each of the plurality of advertisements based on the utility of a viewer, weighting the plurality of advertisements so that continuous viewing of the same advertisement can be adjusted a plurality of times. The advertisement distribution method according to claim 14 or claim 17, characterized in that:

1 9. In the weighting step, when each of the plurality of advertisements is weighted based on the utility of a viewer, the number of views of the same advertisement within a certain period can be adjusted. The advertisement delivery method according to any one of claims 14 to 17, wherein the plurality of advertisements are weighted.

20. The step of weighting includes recognizing words, sounds, or images expressed in a program viewed by a viewer, and weighting a plurality of advertisements according to the recognized words, sounds, or images. The advertisement delivery method according to any one of claims 14 to 17, characterized in that:

2 1. An advertisement distribution method for distributing a plurality of advertisements by weighting each of a plurality of advertisements to be distributed according to each viewer,

Setting a planned delivery frequency and a planned delivery period for each ad;

An advertisement distribution method, comprising the step of weighting each of the plurality of advertisements based on the set planned number of distributions and the planned distribution period.

2 2. The weighting step is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period, which is the number of days until the distribution end date, for each advertisement. Calculating a delivery quota, the number of deliveries per day, for each ad;

22. The advertisement distribution method according to claim 21, comprising: weighting the plurality of advertisements according to the calculated distribution quota.

23. The weighting step is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period, which is the number of days until the distribution end date, for each advertisement, and distributes each of the advertisements per day. Calculating, for each advertisement, a planned delivery quota that is the number of times and a delivery quota that is the number of times of delivery per day during the remaining delivery period at any time;

Calculating a degree of deviation from the plan based on the planned delivery quota and the delivery quota at the given time at any given time;

A step of weighting the plurality of advertisements according to the calculated degree of deviation from plan 22. The advertisement distribution method according to claim 21, comprising a floor.

24. The weighting step is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period, which is the number of days until the distribution end date, for each advertisement. Calculating a delivery quota, which is the number of deliveries per day, for each ad;

Weighting the plurality of advertisements according to the calculated distribution quota;

Calculating, for each advertisement, a planned distribution quota that is the number of distributions per day of the planned distribution period and a distribution quota that is the number of distributions per day of the remaining distribution period at an arbitrary time;

22. The advertisement distribution method according to claim 21, further comprising a step of weighting the plurality of advertisements according to the calculated degree of deviation from the planned advertisement.

2 5.

A step of selecting at least one of the utility of the advertiser and the utility of each viewer; and a procedure of weighting each of the plurality of advertisements based on at least one of the utility of the selected advertiser and the utility of each viewer. A program that is executed to realize distribution of a plurality of advertisements by weighting each of a plurality of advertisements to be distributed according to each viewer.

26. In the weighting step, when each of the plurality of advertisements is weighted based on the utility of the advertiser, the degree of matching between the attribute of the viewer specified by the advertiser and the attribute of each viewer is determined. The steps to determine

26. The program according to claim 25, further comprising: a step of weighting each of the plurality of advertisements according to a result of the determination.

2 7. The attributes of the viewer are:

a) the gender of the viewer, b) the age of the viewer, c) the date of birth of the viewer, d) the address or residence of the viewer, Θ) the type of residence of the viewer, f) the household attributes of the viewer, g) Claim 25 or 26, characterized by including at least one of: occupation of the viewer, h) viewer's monthly or annual income, i) viewer's hobby, j) viewer's preference. The described program.

28. In the weighting step, when each of the plurality of advertisements is weighted based on the utility of the viewer, the degree of coincidence between the field of the hobby or preference specified by the viewer and the field to which each advertisement belongs is determined. Instructions and

26. The program according to claim 25, further comprising: weighting the plurality of advertisements according to a result of the determination.

29. In the weighting step, when each of the plurality of advertisements is weighted based on the utility of the viewer, the plurality of advertisements are weighted so that continuous viewing of the same advertisement can be adjusted a plurality of times. The program according to claim 25 or claim 28, wherein the program is characterized in that:

30. In the weighting step, when weighting each of the plurality of advertisements based on the utility of a viewer, the plurality of advertisements are weighted so that the number of times of viewing the same advertisement within a certain period can be adjusted. The program according to any one of claims 25, 28, or 29, characterized by:

3 1. The weighting procedure involves recognizing words, sounds or images expressed in a program being viewed by a viewer, and weighting a plurality of advertisements according to the recognized words, sounds or images. The program according to any one of claims 25, 28, or 29, characterized in that:

3 2.

Instructions for setting a planned number of deliveries and a planned delivery period for each ad;

That of the plurality of advertisements based on the set planned delivery times and planned delivery period A program for executing a procedure for weighting the advertisements and weighting each of a plurality of advertisements to be distributed according to each viewer to realize distribution of the plurality of advertisements.

33. The weighting procedure is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period, which is the number of days until the distribution end date, for each advertisement. A request to calculate the delivery quota, which is the number of deliveries per day, for each ad;

33. The program according to claim 32, further comprising: weighting the plurality of advertisements according to the calculated distribution quota.

3 4. The weighting procedure is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period, which is the number of days until the distribution end date, for each advertisement. Calculating for each advertisement the planned delivery quota, which is the number of times, and the delivery quota, which is the number of deliveries per day during the remaining delivery period at any time;

Calculating a planned deviation degree based on the planned delivery quota and the delivery quota at the given time at any given time;

33. The program according to claim 32, further comprising: a step of weighting the plurality of advertisements according to the calculated degree of deviation from the planned degree.

35. The weighting procedure is based on the planned number of distributions, the planned distribution period, the number of distributions already performed, and the remaining distribution period that is the number of days until the distribution end date. Calculating the delivery quota, which is the number of deliveries per day, for each ad;

The planned distribution quota, which is the number of distributions per day in the planned distribution period, and at any time Calculating a delivery quota for each ad, which is the number of deliveries per day in the remaining delivery period at the point,

3 6.

A step of selecting at least one of the utility of the advertiser and the utility of each viewer; and a procedure of weighting each of the plurality of advertisements based on at least one of the utility of the selected advertiser and the utility of each viewer. A plurality of advertisements to be distributed, each of which is weighted according to each viewer, and a program for realizing the distribution of the plurality of advertisements. . 3 7.

By executing a procedure of weighting each of the plurality of advertisements based on the set planned number of times of distribution and the planned distribution period, by weighting each of the plurality of advertisements to be distributed according to each viewer A computer-readable recording medium on which a program for realizing distribution of the plurality of advertisements is recorded.