JP7354331B1 - Information processing device, information processing method, and information processing program - Google Patents

Abstract

[Problem] To appropriately control the distribution pace of each advertisement. [Solution] The information processing device according to the present application includes a setting unit that sets dummy advertisements that do not actually deliver advertisements to some of the advertising spaces that will be generated in the future, and records the quantity of dummy advertisements set by the setting unit. a probability calculation unit that calculates the distribution probability of each trafficked advertisement to be distributed to the advertisement space based on the amount subtracted from the generation quantity of the advertisement space that will occur in the future and the distribution target quantity of each trafficked advertisement; Be prepared. [Selection diagram] Figure 1

Description

The present invention relates to an information processing device, an information processing method, and an information processing program.

In recent years, with the rapid spread of the Internet, advertisement distribution via the Internet has become popular. In advertising distribution, there is a listing advertisement that distributes an advertisement linked to the display screen of a terminal device.

As a technique related to listing advertisements, for example, there is a technique of controlling the pace of budget consumption of advertisements by distributing advertisements according to a preset distribution plan (see, for example, Patent Document 1).

JP2017-016320A

However, in the conventional technology, for example, no consideration is given to ensuring the number of times each advertisement is distributed, and it is not possible to appropriately control the distribution pace of each advertisement.

The present invention has been made in view of the above, and an object of the present invention is to provide an information processing device, an information processing method, and an information processing program that can appropriately control the distribution pace of each advertisement.

In order to solve the above-mentioned problems and achieve the purpose, an information processing device according to the present invention includes a setting unit that sets a dummy advertisement that does not actually deliver an advertisement to a part of advertisement spaces that will occur in the future; Before distributing to the ad space based on the amount obtained by subtracting the quantity of the dummy advertisement set by the setting unit from the quantity of the ad space that will occur in the future, and the distribution target quantity of each trafficked advertisement. It includes a calculation unit that calculates the distribution probability of each written advertisement, and a distribution unit that distributes the written advertisement according to the distribution probability calculated by the calculation unit.

According to the present invention, it is possible to appropriately control the distribution pace of each advertisement.

FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. FIG. 2 is a schematic diagram of a dummy case according to the embodiment. FIG. 3 is a block diagram illustrating a configuration example of an information processing apparatus according to an embodiment. FIG. 4 is a diagram illustrating an example of information stored in the advertisement information storage unit according to the embodiment. FIG. 5 is a diagram illustrating an example of information stored in the inventory prediction information storage unit according to the embodiment. FIG. 6 is a diagram illustrating an example of information stored in the relative priority storage unit according to the embodiment. FIG. 7 is a diagram illustrating an example of processing by the generation unit according to the embodiment. FIG. 8 is a diagram showing the relationship between relative priority and delivery probability according to the embodiment. FIG. 9 is a schematic diagram showing an example of relative priorities according to the embodiment. FIG. 10 is a flowchart illustrating an example of relative priority calculation processing according to the embodiment. FIG. 11 is a flowchart illustrating an example of distribution processing according to the embodiment. FIG. 12 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device according to the embodiment.

DESCRIPTION OF THE PREFERRED EMBODIMENTS An information processing apparatus, an information processing method, and an embodiment of an information processing program (hereinafter referred to as "embodiments") according to the present application will be described in detail below with reference to the drawings. Note that the information processing apparatus, information processing method, and information processing program according to the present application are not limited to this embodiment.

[Embodiment]
[1-1. Information processing〕
First, an example of information processing according to the embodiment will be described using FIG. 1. FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. Note that this information processing method is executed by, for example, the information processing apparatus 10 shown in FIG.

Further, although a case will be described below in which the information processing according to the embodiment is processing related to advertisement distribution, the information processing according to the embodiment can be applied to various inventory management.

As shown in FIG. 1, the information processing apparatus 10 according to the embodiment receives, for example, an advertisement distribution request from a user terminal 50, selects an advertisement corresponding to the distribution conditions of the received distribution request, and then selects an advertisement that corresponds to the distribution conditions of the received distribution request. 50. In addition, below, we will explain the case where the distribution condition specifies the user attributes of the user to whom the advertisement is distributed, but it is not limited to this, and for example, a search query used on a search system. may also be specified.

For example, the information processing device 10 distributes an advertisement to be displayed in an advertisement space 200 on a web page displayed on the user terminal 50. Note that the information processing device 10 is realized by, for example, a server device, a cloud system, or the like.

The user terminal 50 is a terminal device owned by the user, and is a terminal that displays advertisements distributed from the information processing device 10 in an advertisement space 200 set on a web page through data communication with the information processing device 10. . Examples of the user terminal 50 include a smartphone, a tablet terminal, a personal computer, and a wearable terminal.

Here, for example, the business operator of the information processing device 10 (hereinafter simply referred to as a business operator) considers advertising spaces that will occur in the future as inventory (hereinafter simply referred to as inventory), and buys and sells advertising spaces with advertisers. Make a contract. For example, an advertiser specifies a distribution period, distribution conditions, and number of distributions, and requests a business operator to purchase a future advertising space.

On the other hand, if the advertisement space presented in the purchase application is in stock, the business operator will conclude a contract in response to the purchase application, and will not conclude a contract if there is insufficient stock. Note that since a contract regarding an advertising space is usually made in units of one campaign, below, an advertisement regarding one contract will also be referred to as a campaign. Furthermore, even if the advertisement is the same, if different contracts are concluded, it will be treated as a separate campaign.

Under such a contract, it is necessary to guarantee the number of times advertisements will be distributed, and in the information processing according to the embodiment, distribution of each advertisement is optimized using a three-layer mathematical programming method.

Here, an overview of information processing using mathematical programming will be explained. The information processing device 10 according to the embodiment calculates the relative priority, which is the relative priority of each campaign, in advance by mathematical programming, and when an advertisement request is received, the information processing device 10 calculates the relative priority, which is the relative priority of each campaign, in advance, and then, when receiving an advertisement request, calculates the relative priority for the distribution request based on the relative priority. Calculate the distribution probability and distribute advertisements.

For example, as shown in FIG. 1, the information processing device 10 according to the embodiment first obtains inventory prediction information from the web server 100 that provides a web page that serves as a display medium for advertising spaces (step S01).

For example, the inventory prediction information is information about the advertising spaces 200 that will occur in the future on a web page provided by the web server 100. For example, it is information about the advertising spaces 200 that will occur for each unit time, and for each time period. This is predictive information such as how many advertising spaces 200 with what attributes will occur.

Subsequently, the information processing device 10 calculates the relative priority of each advertisement (step S02). The information processing device 10 calculates the relative priority of each campaign by, for example, mathematical programming using the first to third layers.

Here, the relative priority is a relative numerical value of the distribution ratio of each campaign, such as which campaign should be distributed at what ratio among the entire campaign. Note that the mathematical programming method here includes various methods such as linear programming, nonlinear programming, dynamic programming, and discrete programming.

For example, the nodes (S1 to S3...) shown in the first layer correspond to the distribution conditions of each inventory. For example, one node is generated for each stock distribution condition, and such a node is, for example, a matrix corresponding to the number of stocks for each date and time.

Further, nodes (D1 to D3...) shown in the third layer correspond to respective campaigns. For example, a new node is generated for each addition of a campaign, and this node is a matrix corresponding to the target number of deliveries for the entire campaign period, ie, the number of inventory purchased in the campaign.

The nodes (P1 to P3, . . . ) shown in the second layer are a matrix in which each of the nodes in the third layer is subdivided into the target number of distributions per unit time. That is, the nodes of the second layer are, for example, a matrix corresponding to the value obtained by dividing the target number of times of distribution by the distribution period of each node of the third layer.

In the information processing apparatus 10 according to the embodiment, the nodes in the second layer that can be distributed under the distribution conditions of each node in the first layer are associated with each other for each unit time, and then mathematical programming is used. Calculate relative priority.

For example, the calculated relative priority is stored in the relative priority storage section of the information processing device 10. Note that the relative priority may be anything that indicates a relative index for each campaign, and the method for calculating the relative priority is not limited to the example of step S02.

After that, the information processing device 10 receives, for example, an advertisement request to be displayed in the advertisement space 200 from the user terminal 50 (step S03). For example, the advertisement request includes information regarding advertisement distribution conditions, and the information processing device 10 distributes the advertisement after specifying the advertisement corresponding to the information regarding the distribution conditions.

At this time, if there are multiple advertisements corresponding to the distribution condition, the information processing device 10 calculates a relative priority, which is a priority for the advertisement request, from the relative priority stored in the relative priority storage unit (step S04 ).

For example, the information processing device 10 calculates the distribution probability by using a normalization method called L1-Ball Projection. For example, the information processing device 10 sets the value obtained by subtracting the same number from each relative priority as the relative priority of each advertisement so that the sum of the relative priorities of each of a plurality of advertisements competing for an advertisement request is 1. calculate. For example, if the relative priorities of advertisement A and advertisement B are "0.8" and "0.6", the total sum is "1.4". At this time, the distribution probability of advertisement A is calculated as "0.8-0.2=0.6", and the distribution probability of advertisement B is calculated as "0.6-0.2=0.4".

Then, the information processing device 10 distributes an advertisement in response to the distribution request according to the calculated relative priority (step S05). That is, in the above example, advertisement A and advertisement B are distributed at a ratio of 60% and 40% to distribution requests with the same attribute. Note that, for example, if the sum of the relative priorities of advertisement A and advertisement B is less than 1, each relative priority directly becomes the distribution probability.

In this way, the information processing device 10 according to the embodiment can appropriately distribute each advertisement by calculating the distribution probability of advertisements that compete with the distribution request using the relative priority calculated in advance. It is.

[1-2. Calculation process]
Next, a specific procedure for calculating the relative priority will be explained. First, a case will be described in which relative priorities are calculated from two layers, the first layer and the third layer. Let each node shown in the first layer be stock s_i, and each node shown in the third layer be d_j, and then let x_tj be the relative priority and p_ij be the relative priority. Note that i is an inventory index, and j is an issue (campaign) index.

Let the set of items j be A, the set of inventory i belonging to item j be Γ(j), the set of items j belonging to inventory i be Γ(i), the total inventory of item j be S_j, and the full-year scheduled delivery probability (Σs_i Optimization regarding relative priority p_ij that matches p_ij/S_j) and target distribution probability (d_j/S_j) can be calculated using an evaluation function shown in equation (1) below. Note that the full-year scheduled delivery probability is, for example, a relative priority seen from the inventory side, and the target delivery probability is a relative priority seen from the project side.

Note that in the above equation (1), the evaluation function is multiplied by 1/2 in consideration of later differentiation. The relative priority for each campaign is calculated using the above formula (1), and in reality, the relative priority p_ij is calculated by normalizing the relative priority using L1-Ball Projection.

Specifically, when the time period t belonging to inventory i is defined as B(i), the adjustment term δ_i for deriving the relative priority p_ij from the relative priority x_tj of each item is calculated using the following formula (2). I will do it.

On the other hand, when the second layer is further added to the first and third layers, the relative priority p_ij is determined by the relative priority x_tj of the matter j in the time period t and the adjustment term δ_i shown in equation (2) above. It can be expressed by the following equation (3) using , and when this is applied to the optimization equation, equation (4) is obtained.

Regarding equation (4), in order to eliminate the max function, x_tj>δ_i is added to the condition of element i, and when it is set to L, it is finally expressed by equation (5) and optimization regarding relative priority x_tj is performed.

Next, by alternately repeating the derivation of the adjustment term δ_i using the above equation (2) and the update of the relative priority x_tj of the case j in the time period t using the above equation (5), the calculation of each case in each time period is performed. Optimize relative priority x_tj.

[1-3. Adjustment process]
By the way, when selling inventory in advance to such an advertiser, for example, instead of selling all the inventory, it is possible to keep some of the inventory. Therefore, in the information processing according to the embodiment, it was decided to set advertising spaces that are not sold to businesses for some inventory.

For example, as shown in FIG. 2, the information processing device 10. Set the inventory to be sold to advertisers and the unsold inventory that will not be sold to advertisers. Then, the information processing device 10 sets dummy advertisements for the unsold inventory, and calculates the relative priority and delivery probability of each campaign based on the amount obtained by subtracting the quantity of the dummy advertisement from the entire inventory. .

Here, the dummy advertisement is used to secure unsold inventory and reduce the inventory sold to advertisers, and is not actually targeted for distribution. Therefore, the information processing device 10 selects an advertisement to be distributed in response to the distribution request from among advertisements other than dummy advertisements.

For example, if the set of dummy advertisements is SL and the proportion of inventory other than dummy advertisements shown in inventory i is Ki, then the proportion of inventory Ki can be expressed by the following equation (6).

Since the dummy advertisement does not participate in inventory competition, when the adjustment term δ_i of the dummy advertisement is set to 0, the evaluation function can be expressed by the following equation (7).

The information processing device 10 sets the evaluation function so that the sum of the relative priorities of the non-dummy projects and the sum of the relative priorities of the dummy projects is 1 or less, as shown in equation (7). By optimizing the relative priority X_tj, the relative priority x_tj of each case in each time period is optimized.

That is, the information processing device 10 can appropriately calculate the relative priority of each campaign after securing unsold inventory using the logic that optimizes these relative priorities x_tj.

Therefore, according to the information processing device 10 according to the embodiment, the distribution pace of each advertisement can be appropriately controlled. In particular, the information processing device 10 according to the embodiment can be an effective means for securing unsold inventory.

[2. Configuration example of information processing device]
Next, a configuration example of the information processing device 10 will be described using FIG. 3. FIG. 3 is a block diagram showing a configuration example of the information processing device 10 according to the embodiment. As shown in FIG. 3, the information processing device 10 includes a communication section 110, a control section 120, and a storage section 130.

The communication unit 110 is realized by, for example, a NIC (Network Interface Card). The communication unit 110 connects networks such as various wireless communication networks such as 4G (Generation), 5G, LTE (Long Term Evolution), WiFi (registered trademark), or wireless LAN (Local Area Network), or various wired communication networks. Information is sent and received to and from external devices via the external device.

The storage unit 130 is realized by, for example, a semiconductor memory device such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 130 also includes an advertisement information storage unit 131, an inventory prediction information storage unit 132, and a relative priority storage unit 133.

The advertisement information storage unit 131 stores information regarding advertisements of campaigns for which advertisement space has been purchased. FIG. 4 is a diagram illustrating an example of information stored in the advertisement information storage unit 131 according to the embodiment. As shown in FIG. 4, the advertising information storage unit 131 associates information such as "campaign ID," "advertiser ID," "distribution conditions," "distribution period," and "number of purchases" with each other. Remember.

The "campaign ID" is an identifier of the campaign that purchased the advertising space, or in other words, an identifier of the advertisement to be distributed. "Advertiser ID" is an identifier of the advertiser who purchased the advertising space. "Distribution condition" indicates the advertisement distribution condition. Note that distribution conditions may be read as targeting.

"Distribution period" is the distribution period of the corresponding advertisement, and indicates the advertising campaign period. The “number of purchases” indicates the number of advertising spaces purchased in the campaign. Note that, as shown in FIG. 4, for example, the campaign identified by campaign ID "Ca3" is a dummy advertisement, and is intended to limit the number of sales of inventory under distribution condition T1.

Returning to the explanation of FIG. 3, the inventory prediction information storage unit 132 will be explained. The inventory prediction information storage unit 132 stores information regarding the inventory of the advertising space 200. Information predicting the number of inventory of advertising space 200 is stored. For example, the inventory prediction information is information predicted based on past access performance values for the display medium of the advertising space.

FIG. 5 is a diagram illustrating an example of information stored in the inventory prediction information storage unit 132 according to the embodiment. As shown in FIG. 5, the inventory prediction information storage unit 132 stores information on items such as "user attributes," "date and time," and "inventory quantity" in association with each other.

"User attribute" indicates, for example, the attribute of the user who makes the advertisement request. “Date and time” and “inventory quantity” indicate that the inventory quantity corresponding to the corresponding date and time will occur. That is, the example of FIG. 4 shows that "3000" items are in stock for the user group with the user attribute "T1" at 13:00. Note that the number of items in stock here is a predicted value.

Returning to the explanation of FIG. 3, the relative priority storage unit 133 will be explained. The relative priority storage unit 133 stores information regarding relative priorities. FIG. 6 is a diagram illustrating an example of information stored in the relative priority storage unit 133 according to the embodiment.

As shown in FIG. 6, the relative priority storage unit 133 stores information on items such as "campaign ID," "date and time," and "relative priority" in association with each other. "Campaign ID" is an identifier of the campaign that purchased the advertising space.

"Date and time" and "relative priority" each indicate relative priority at the corresponding date and time. Note that the relative priority is, for example, a relative numerical value of the distribution ratio of each campaign based on the relationship between inventory and the distribution target number of the entire campaign. For example, the larger the distribution target number of a campaign, the greater the value of the relative priority, and the greater the inventory (number of distribution requests) corresponding to the campaign, the greater the value of the relative priority.

Returning to the explanation of FIG. 3, the control unit 120 will be explained. The control unit 120 is, for example, a controller, and various programs stored in a storage device inside the information processing device 10 use a RAM as a work area using a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. This is achieved by executing as .

As shown in FIG. 3, the control unit 120 includes an acquisition unit 121, a setting unit 122, a pre-calculation unit 123, a reception unit 124, a specification unit 125, a probability calculation unit 126, and a distribution unit 127.

The acquisition unit 121 acquires inventory prediction information and advertisement information. The acquisition unit 121 acquires, for example, inventory prediction information from the web server 100 and advertisement information from a business operator.

The inventory prediction information acquired by the acquisition unit 121 is stored in the inventory prediction information storage unit 132, and the advertisement information is stored in the advertisement information storage unit 131. Note that instead of acquiring the inventory prediction information, the information processing device 10 may independently calculate information corresponding to the inventory prediction information from the advertisement distribution history.

The setting unit 122 sets dummy advertisements that do not actually deliver advertisements to some of the advertisement spaces that will occur in the future. For example, the setting unit 122 sets a dummy advertisement for inventory that has not yet been sold among the inventory stored in the inventory prediction information storage unit 132.

For example, the setting unit 122 sets a dummy advertisement for inventory based on an operation by a business operator. More specifically, for example, the business operator can set the user attributes and period regarding the setting of the dummy advertisement, and the setting unit 122 sets the dummy advertisement based on the information set by the business operator.

More specifically, the business operator can specify conditions such as limiting the number of inventory sales for advertising space targeted at men in their 30s to 50% from when to when. In other words, in this case, the setting unit 122 stores the dummy advertisement (as shown in FIG. (see Ca3).

The pre-calculation unit 123 calculates a relative priority, which is a relative priority of each advertisement. For example, prior to calculating the relative priority, the preliminary calculation unit 123 generates evaluation functions (first layer, second layer, and third layer) for calculating the relative priority. FIG. 7 is a diagram showing an example of an evaluation function using the first to third layers according to the embodiment. For example, each node in the first layer shown in FIG. 7 corresponds to the distribution condition of each inventory, and is a matrix regarding the number of inventory generated per unit time.

Further, each node in the third layer shown in FIG. 7 corresponds to each campaign, and is a matrix regarding the distribution target number for the entire campaign period. For example, the advance calculation unit 123 refers to the inventory prediction information storage unit 132 and generates the first layer based on the inventory prediction information, and refers to the advertising information storage unit 131 and generates the third layer based on the advertising information. do.

Furthermore, each node in the second layer corresponds to the campaign indicated by each node in the third layer, and converts the matrix regarding the number of distribution targets for the entire campaign period indicated by each node in the third layer into the target number of distributions per unit time. This is the matrix.

After generating each node of the second layer, the pre-calculation unit 123 identifies a combination of each node of the second layer corresponding to each node of the first layer. For example, if an advertisement for campaign A can be distributed for distribution condition A, specify a first layer node corresponding to distribution condition A and a second layer node corresponding to campaign A as one combination. and connect the corresponding nodes.

In addition, if advertisements for campaign A or campaign B can be distributed for distribution condition A or B, nodes in the first layer of distribution condition A and nodes in the second layer corresponding to campaign A and campaign B, respectively. Identify one combination and connect the corresponding nodes. By sequentially performing these processes, it is possible to associate the inventory generated per unit time with the combination of campaigns that can be distributed for each inventory.

Thereafter, the pre-calculation unit 123 calculates the relative priority by linear programming using the predicted number of distribution requests obtained per unit time for each distribution condition and the target number of advertisements submitted per unit time.

At this time, if the total target number of submissions exceeds the number of advertisements in stock, the pre-calculation unit 123 adjusts the relative priority of each advertisement according to the number of advertisements in stock. That is, in this case, the pre-calculation unit 123 sets a distribution target number that is less than the target number of submissions for advertisements with a low relative priority, and then establishes a distribution plan T1 (see FIG. 2A). Become.

The reception unit 124 receives a request for distribution of an advertisement for the advertisement space 200. For example, the reception unit 124 receives from the user terminal 50 a distribution request associated with information regarding a user to whom advertisements are to be distributed. The information regarding the user here may include a user ID that identifies the user, and various personal data such as age, gender, annual income, place of residence, occupation, hobbies and preferences, etc.

Information regarding the distribution request accepted by the reception unit 124 is passed to the identification unit 125. Note that the reception unit 124 accesses a user information storage unit that stores personal data for each user, and acquires information regarding the user attributes of the corresponding user based on an identifier that identifies the user (for example, a user ID). You can also do this.

The specifying unit 125 specifies a combination corresponding to the distribution request received by the receiving unit 124 from among the combinations of advertisement distribution conditions and advertisements to be distributed under each distribution condition.

That is, the specifying unit 125 specifies a campaign that can be distributed in response to a distribution request. For example, the identifying unit 125 refers to the distribution conditions of each campaign and identifies a campaign that matches the user attributes of the distribution request.

The specifying unit 125 then passes information regarding the specified campaign to the probability calculating unit 126. Note that, for example, if there is only one campaign that can be distributed in response to a distribution request, the advertisement corresponding to that campaign will be distributed by the distribution unit 127.

The probability calculation unit 126 calculates the distribution probability of each advertisement included in the combination specified by the identification unit 125 based on the relative priority calculated in advance by the advance calculation unit 123.

When the probability calculation unit 126 receives information regarding the campaign specified as a distribution target from the identification unit 125, the probability calculation unit 126 acquires information regarding the relative priority of each campaign from the relative priority storage unit 133.

Next, the probability calculation unit 126 calculates the distribution probability using the obtained relative priority of each campaign. The probability calculation unit 126 passes information regarding the calculated distribution probability to the distribution unit 127.

Here, the relationship between relative priority and delivery probability will be explained using FIGS. 8 and 9. FIG. 7 is a diagram showing the relationship between relative priority and delivery probability according to the embodiment. FIG. 8 is a schematic diagram showing an example of distribution probability according to the embodiment.

For example, as shown on the left side of FIG. 8, campaigns A to D... have relative priorities of 0.2, 0.8, 0.6, and 0.3, respectively. For example, assume that the specifying unit 125 narrows down the campaigns that can be distributed in response to the distribution request (step S11), and as a result, the campaigns are narrowed down to three campaigns A, B, and C in total.

Next, the probability calculation unit 126 calculates the distribution probability by subtracting the same number from the respective relative priorities of campaigns A, B, and C so that the total sum becomes 1 (step S12). For example, in this case, the sum of the relative priorities of campaigns A, B, and C is "0.2+0.8+0.6=1.6."

Therefore, the probability calculation unit 126 calculates a value obtained by subtracting 0.2 from the relative priority of each of campaigns A, B, and C as the relative priority of each of campaigns A, B, and C. That is, in this case, the relative priorities of campaigns A, B, and C are 0, 0.6, and 0.4, respectively.

In other words, in this case, as shown in FIG. 8, campaign A is not distributed in response to distribution requests with the same attribute, and campaigns B and C are distributed at a probability of 0.6 to 0.4, so that each campaign is The pace of delivery can be appropriately controlled.

In addition, in the above example, campaign A will not be distributed at all in response to a distribution request with the same attributes as the above distribution request, but campaign A will not be distributed at all in response to a distribution request with the same attributes as the above distribution request, but campaign A will not be distributed at all in response to a distribution request with the same attributes as the above distribution request. Therefore, the distribution pace of campaign A can also be appropriately controlled.

That is, for example, when the distribution conditions of two or more campaigns have an inclusive relationship and a distribution request corresponding to a subset part comes, the distribution probability is set so that the distribution conditions of the subset are distributed preferentially. Calculated. In this case, if a distribution request corresponding to a subset other than the subset is received, it is sufficient to distribute the other campaign, so it is possible to appropriately control the distribution pace of each campaign.

To give a more specific example, assume that the distribution condition for campaign A is male, the distribution condition for campaign B is male, in his 30s, and the distribution condition for campaign C is male, living in Tokyo.

In this case, if campaign A is distributed in a distribution request that does not conflict with the distribution conditions of campaign B and campaign C, it is possible to maintain an appropriate distribution pace with respect to inventory. In other words, Campaign A may be distributed to men other than men in their 30s living in Tokyo, and even if the distribution probability for such distribution requests is set to 0%, it is distributed to other distribution requests. Therefore, the distribution pace of each campaign can be appropriately controlled.

Returning to the explanation of FIG. 3, the distribution unit 127 will be explained. The distribution unit 127 selects advertisements as distribution candidates based on the distribution probability calculated by the probability calculation unit 126 and distributes them to the user terminal 50.

Thereby, the information processing apparatus 10 according to the embodiment can appropriately control the delivery pace of each advertisement delivered to the user terminal 50.

[3. Processing flow]
Next, a processing procedure executed by the information processing apparatus 10 according to the embodiment will be described using FIGS. 10 and 11. FIG. 10 is a flowchart illustrating an example of relative priority calculation processing according to the embodiment. FIG. 11 is a flowchart illustrating an example of distribution processing according to the embodiment.

As shown in FIG. 9, the information processing device 10 first obtains inventory prediction information (step S101) and sets a dummy advertisement (step S102). Subsequently, the information processing device 10 sets a combination of inventory distribution conditions and advertisements (campaigns) that can be distributed for each distribution condition (step S103).

Subsequently, the information processing device 10 identifies the relationship between the number of items in stock per unit time and the target number of distributions for each combination (step S104). Note that through the processing up to step S104, nodes for each of the first to third layers are generated.

Subsequently, the information processing device 10 calculates the relative priority for each campaign (step S105), and ends the relative priority calculation process.

As shown in FIG. 11, in the distribution process, the information processing device 10 first receives a distribution request (step S201), and then identifies a campaign that can be distributed in response to the distribution request (step S202).

Subsequently, the information processing device 10 calculates a relative priority from the relative priorities of each of the distributable campaigns (step S203), distributes an advertisement according to the relative priority (step S204), and performs distribution processing. end.

[4. Modified example]
Incidentally, in the above-described embodiment, a case has been described in which the advertisement distribution pace is controlled based on the advertisement demand forecast and the distribution target, but the present invention is not limited to this. That is, resources are not limited to advertisements, but may be resources related to product delivery, for example, and the product itself may be regarded as a resource.

That is, the present invention may be applied to various types of inventory management. In this case, the present invention can be applied by replacing the above-mentioned inventory forecasting information with targeted demand forecasting information and replacing advertising information with targeted supply target information.

Further, in the above-described embodiment, a case has been described in which the time units of the first layer and the third layer are the same in the second layer, but the present invention is not limited to this. That is, in the second layer, for example, any parameter other than the time unit may be arranged. As an example of this, for example, parameters related to the region of advertisement distribution may be arranged.

[5. effect〕
The information processing device 10 according to the embodiment described above includes a setting unit 122 that sets dummy advertisements that do not actually deliver advertisements to some of the advertising spaces that will occur in the future, and a quantity of dummy advertisements set by the setting unit 122. A probability calculation unit 126 that calculates the distribution probability of each trafficked advertisement to be distributed to the advertisement space based on the subtracted quantity obtained by subtracting the amount of the trafficked advertisement from the generation quantity of the advertisement space that will be generated in the future, and the distribution target quantity of each trafficked advertisement. Equipped with. Therefore, according to the information processing device 10 according to the embodiment, the distribution pace of each advertisement can be appropriately controlled.

In addition, the information processing device 10 according to the embodiment uses a mathematical programming method that uses the subtracted quantity of advertising spaces per unit time and the distribution target number of dummy advertisements and trafficked advertisements per unit time. The probability calculation unit 126 includes a pre-calculation unit 123 that calculates in advance the relative priority of each submitted advertisement, and a probability calculation unit 126 distributes the advertisement to the ad space based on the relative priority calculated by the pre-calculation unit 123. Calculate the delivery probability of each possible trafficked advertisement. Therefore, according to the information processing apparatus 10 according to the embodiment, the distribution probability can be calculated without delay in response to the distribution request.

In the information processing device 10 according to the embodiment, the pre-calculation unit 123 calculates the dummy advertisements so that the sum of the relative priorities of each submitted advertisement and the sum of the relative priorities of each dummy advertisement becomes less than 1. Calculate the relative priority of each advertisement and trafficked advertisement. Therefore, the information processing apparatus 10 according to the embodiment can calculate an appropriate relative priority for each submitted item in consideration of the dummy advertisement.

Furthermore, in the information processing apparatus 10 according to the embodiment, the probability calculation unit 126 calculates the distribution probability by normalizing the relative priority of each submitted advertisement that can be distributed to the advertisement space. Therefore, according to the information processing device 10 according to the embodiment, it is possible to calculate an appropriate distribution probability in consideration of dummy advertisements.

Further, the information processing method according to the embodiment is an information processing method executed by a computer, and includes a setting step of setting a dummy advertisement that does not actually deliver an advertisement to a part of an advertisement space that will occur in the future; Delivery of each trafficked ad to the ad space based on the amount of dummy ads set by subtracted from the amount of ad space that will occur in the future and the distribution target quantity of each trafficked ad. and a probability calculation step of calculating the probability. Therefore, according to the information processing method according to the embodiment, it is possible to appropriately control the distribution pace of each advertisement.

Further, the information processing program according to the embodiment includes a setting procedure for setting dummy advertisements that do not actually deliver advertisements to some of the advertising spaces to be generated in the future, and a quantity of the dummy advertisements set by the setting procedure. Probability calculation that calculates the distribution probability of each of the drafted advertisements to be distributed to the advertisement space based on the amount subtracted from the number of advertisement spaces that will occur in the future and the distribution target quantity of each of the trafficked advertisements. Make the computer perform the steps. Therefore, according to the information processing program according to the embodiment, it is possible to appropriately control the distribution pace of each advertisement.

[6. Hardware configuration]
Further, the information processing apparatus 10 according to the embodiments described above is realized by, for example, a computer 1000 having a configuration as shown in FIG. 12. FIG. 12 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device 10 according to the embodiment. Computer 1000 has CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I/F) 1500, input/output interface (I/F) 1600, and media interface (I/F) 1700.

CPU 1100 operates based on a program stored in ROM 1300 or HDD 1400, and controls each section. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started, programs depending on the hardware of the computer 1000, and the like.

The HDD 1400 stores programs executed by the CPU 1100, data used by the programs, and the like. Communication interface 1500 receives data from other devices via network (communication network) N and sends it to CPU 1100, and sends data generated by CPU 1100 to other devices via network N.

The CPU 1100 connects output devices such as displays and printers, and input devices such as keyboards and mice (in FIG. 11, the output devices and input devices are collectively referred to as "input/output devices") via the input/output interface 1600. ). CPU 1100 obtains data from an input device via input/output interface 1600. Further, CPU 1100 outputs the generated data to an output device via input/output interface 1600.

Media interface 1700 reads programs or data stored in recording medium 1800 and provides them to CPU 1100 via RAM 1200. CPU 1100 loads this program from recording medium 1800 onto RAM 1200 via media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. etc.

For example, when the computer 1000 functions as the information processing device 10 according to the embodiment, the CPU 1100 of the computer 1000 realizes the functions of the control unit 120 by executing a program loaded onto the RAM 1200. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired from another device via the network N.

Some of the embodiments of the present application have been described above in detail based on the drawings, but these are merely examples, and various modifications and variations may be made based on the knowledge of those skilled in the art, including the embodiments described in the disclosure section of the invention. It is possible to carry out the invention in other forms with modifications.

[7. others〕
Further, among the processes described in the above embodiments and modified examples, all or part of the processes described as being performed automatically can be performed manually, or are described as being performed manually. All or part of this processing can also be performed automatically using known methods. In addition, information including the processing procedures, specific names, and various data and parameters shown in the above documents and drawings may be changed arbitrarily, unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

Furthermore, each component of each device shown in the drawings is functionally conceptual, and does not necessarily need to be physically configured as shown in the drawings. In other words, the specific form of distributing and integrating each device is not limited to what is shown in the diagram, and all or part of the devices can be functionally or physically distributed or integrated in arbitrary units depending on various loads and usage conditions. Can be integrated and configured.

Furthermore, the above-described embodiments and modifications can be combined as appropriate within a range that does not conflict with the processing contents.

Further, the above-mentioned "section, module, unit" can be read as "means", "circuit", etc. For example, the acquisition unit 121 can be replaced with an acquisition means or a user information acquisition circuit.

10 Information processing device 50 User terminal 100 Web server 110 Communication unit 120 Control unit 121 Acquisition unit 122 Setting unit 123 Pre-calculation unit 124 Reception unit 125 Specification unit 126 Probability calculation unit 127 Distribution unit 130 Storage unit 131 Advertisement information storage unit 132 Inventory prediction Information storage unit 133 Relative priority storage unit 200 Advertising space

Claims (6)

a setting section that sets dummy advertisements that do not actually deliver advertisements to some of the advertisement spaces that will occur in the future;
Deliver to the ad space based on a subtraction amount obtained by subtracting the quantity of the dummy advertisement set by the setting unit from the quantity of the ad space to be generated in the future, and a delivery target quantity of each trafficked advertisement. a probability calculation unit that calculates the delivery probability of each of the posted advertisements;
An information processing device comprising: The relative quantity of each of the dummy advertisement and the posted advertisement is determined by mathematical programming using the subtracted quantity per unit time of the advertising space and the target number of distributions per unit time of the dummy advertisement and the posted advertisement. It is equipped with a pre-calculation unit that calculates the relative priority in advance,
The probability calculation unit is
The information processing according to claim 1, characterized in that the distribution probability of each of the posted advertisements that can be distributed to the advertising space is calculated based on the relative priority calculated by the pre-calculation unit. Device. The pre-calculation unit is
The relative priorities of each of the dummy advertisement and the posted advertisement are set such that the sum of the relative priority of each of the posted advertisements and the sum of the relative priority of each of the dummy advertisements is less than 1. The information processing device according to claim 2, further comprising: calculating. The probability calculation unit is
The information processing device according to claim 2 or 3, wherein the distribution probability is calculated by normalizing the relative priority of each of the posted advertisements that can be distributed to the advertisement space. An information processing method performed by a computer, the method comprising:
A setting process of setting dummy advertisements that do not actually deliver advertisements to some of the advertisement spaces that will occur in the future;
Deliver to the advertising space based on the amount obtained by subtracting the quantity of the dummy advertisement set in the setting step from the quantity of the advertising space that will occur in the future, and the distribution target quantity of each trafficked advertisement. a probability calculation step of calculating the distribution probability of each of the posted advertisements;
An information processing method comprising: Setting procedure for setting dummy ads that will not actually deliver ads to some of the ad spaces that will occur in the future,
Deliver to the advertising space based on the amount obtained by subtracting the quantity of the dummy advertisement set by the setting procedure from the quantity of the advertising space that will occur in the future, and the distribution target quantity of each trafficked advertisement. a probability calculation procedure for calculating the distribution probability of each of the posted advertisements;
An information processing program that causes a computer to execute.

Images