Information processing device, information processing method, and information processing program

Abstract

To appropriately estimate conversion of a distributed advertisement.SOLUTION: An information processing apparatus relating to the present application comprises: an acquisition unit which acquires both statistical information about an object behavior performed by users who have selected any of a plurality of advertisements being distribution objects and permitted behavior information indicating whether permitting users giving permission out of the users who have selected any of the advertisements have performed an object behavior or not; and a generation unit which generates probability information indicating a probability with which the individual users who have selected any of the advertisements have performed the object behavior, on the basis of the statistical information and the permitted behavior information acquired by the acquisition unit.SELECTED DRAWING: Figure 2

Description

【Technical Field】 【0001】 The present invention relates to an information processing apparatus, an information processing method, and an information processing program. 【Background Art】 【0002】 Conventionally, various metrics have been used in advertising distribution via the Internet. For example, techniques for measuring the effect of advertising distribution using various metrics such as CVR (Conversion Rate; also referred to as "conversion rate") are known (see, for example, Patent Document 1). 【Prior Art Documents】 【Patent Documents】 【0003】 【Patent Document 1】 Japanese Patent Application Laid-Open No. 2018-116345 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0004】 However, in recent years, anonymization of user information has been required from the perspective of protecting personal information. Under regulations, there may be cases where information necessary for conversion measurement cannot be obtained, and it is desirable to estimate finer-grained information from coarsely grained information in which user information is anonymized. 【0005】 The present invention has been made in view of the above, and an object thereof is to provide an information processing apparatus, an information processing method, and an information processing program capable of generating finer-grained information from coarsely grained information in which user information is anonymized. 【Means for Solving the Problems】 【0006】 To solve the above-mentioned problems and achieve the objective, the information processing device according to the present invention comprises: an acquisition unit that acquires statistical information relating to target actions taken by users who have selected one of a plurality of advertisements to be delivered, and authorized action information indicating whether or not a target action was taken by authorized users who have given permission among the users who have selected one of the advertisements; and a generation unit that generates accuracy information indicating the probability that each user who has selected one of the advertisements has taken a target action, based on the statistical information and authorized action information acquired by the acquisition unit. [Effects of the Invention] 【0007】 According to the present invention, it is possible to generate finer-grained information from coarse-grained information in which user information has been anonymized. [Brief explanation of the drawing] 【0008】 [Figure 1] Figure 1 shows an example of an information processing system according to an embodiment. [Figure 2] Figure 2 shows an example of information processing according to the present invention. [Figure 3] Figure 3 is a block diagram showing an example of an information processing device according to the present invention. [Figure 4] Figure 4 shows an example of a conversion database according to the present invention. [Figure 5] Figure 5 shows an example of accuracy information using the predicted values ​​according to the present invention. [Figure 6] Figure 6 is a flowchart showing an example of a processing procedure performed by the information processing device according to the embodiment. [Figure 7] Figure 7 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device according to the embodiment. [Modes for carrying out the invention] 【0009】 The following describes in detail, with reference to the drawings, the embodiments for implementing the information processing device, information processing method, and information processing program relating to this application (hereinafter referred to as "embodiments"). However, these embodiments do not limit the information processing device, information processing method, and information processing program relating to this application. 【0010】 [Embodiment] [1.1. Information Processing Systems] First, an overview of the information processing system and information processing according to the embodiment will be described using Figures 1 and 2. Figure 1 is a diagram showing an example of the information processing system according to the embodiment. Figure 2 is a diagram showing an example of information processing according to the embodiment. 【0011】 As shown in Figure 1, the information processing system 1 according to this embodiment includes a user terminal 50 connected by a network N, an information processing device 100, and an advertiser server 200. Although Figure 1 shows the case where there is one user terminal 50, one information processing device 100, and one advertiser server 200 included in the information processing system 1, there may be multiple user terminals 50, one information processing device 100, and one advertiser server 200. 【0012】 The information processing device 100 shown in Figure 1 is an information processing device that provides content related to various web services to the user terminal 50. For example, the information processing device 100 delivers advertisements submitted by advertisers to the user terminal 50 along with content related to web services. 【0013】 Furthermore, the information processing device 100 measures the conversions (hereinafter also referred to as CV) of the delivered advertisements provided to the user terminal 50. The information processing device 100 then optimizes the CVR (Conversion Rate) of each delivered advertisement according to the measured conversions. For example, the information processing device 100 classifies each request into its respective attribute and optimizes conversions by performing machine learning using the conversions for each attribute as training data. 【0014】 For example, the attributes of a request can be arbitrarily classified according to the user who is the source of the request (e.g., psychographic attributes or demographic attributes), the type of distribution frame or distribution base where the advertisement is distributed, the URL, and the domain, past performance, etc. In the example of FIG. 1, it shows that there were requests identified by request IDs "r1~r3" for the distributed advertisement identified by the campaign ID "Camp.1". 【0015】 Also, in the example of FIG. 1, it shows that each of the requests r1~r3 is classified into attributes f1~f3. Note that "f1~f3" indicates the indexes of the feature amounts of the corresponding attributes, and the above machine learning is performed using the feature amounts corresponding to the indexes. 【0016】 The user terminal 50 is a terminal owned by the user. For example, the user terminal 50 displays the content distributed from the information processing device 100 together with the distributed advertisement. The user terminal 50 is various client terminals such as, for example, a smartphone, a tablet terminal, a personal computer, and a wearable terminal. 【0017】 Also, when the distributed advertisement displayed together with the content is clicked by the user, the user terminal 50 moves to the advertiser site operated by the advertiser from the content. Then, when a predetermined conversion is performed at the advertiser site, the user terminal 50 transmits the conversion information to the information processing device 100. Thereby, the information processing device 100 can measure the conversion of the distributed advertisement. 【0018】 The advertiser server 200 is a server device that operates the advertiser site of the advertiser. For example, the advertiser server 200 provides the advertiser site to the user terminal 50 in response to an access by the user terminal 50. 【0019】 〔1.2. An Example of Information Processing〕 By the way, in recent years, the use of third-party cookies and the like has been restricted by the so-called ITP (Intelligent Tracking Prevention) installed by platforms. 【0020】 Under such constraints, anonymization of personal information is also required in advertising distribution. Therefore, the information processing apparatus 100 is becoming unable to track the actions of each user on the advertiser site, and is becoming unable to measure conversions in units of requests, which has been possible until now. 【0021】 Under such circumstances, some information regarding conversions may be missing. For example, conventionally, as shown in the graph of FIG. 1, the information processing apparatus 100 can measure by associating a campaign ID, a request ID (including request attributes), and a CV (conversion) respectively by tracking the actions of each user. 【0022】 On the other hand, due to the installation of ITP, for example, even for a request in which a conversion has actually occurred, the conversion may be missing. In the example shown in FIG. 1, although a conversion has occurred in request r3, the presence or absence of the conversion is unknown, and a case where the conversion, which should be "1", has become "0" is shown. Note that for a permitted user who has permitted tracking, conversions can be measured in units of requests as before. In FIG. 1, it is assumed that request r1 is a request made by a permitted user, and "1 time" conversion has been measured for request r1. 【0023】 In addition, the information processing apparatus 100 can obtain the number of conversions in units of campaigns, which have a coarser granularity than requests, and generate probability information indicating the probability that a conversion has occurred in each request using these information. 【0024】 For example, as shown in Figure 2, suppose the number of conversions per campaign is "2", and the sum of the number of conversions measured for each request r1 to r3 is "1". In this case, excluding request r1, which does not have any missing conversions, a total of "1" conversions occurred in either request r2 or request r3. 【0025】 In other words, a gap in the conversion occurred in either request r2 or request r3. Therefore, the information processing device 100 generates probability information indicating the likelihood that a conversion occurred at the missing location. Here, "missing" means that it is unclear whether or not a conversion occurred. 【0026】 Specifically, as shown in Figure 2, for example, the information processing device 100 assigns a numerical value to the conversion (CV) of request r2 or request r3, which may have missing conversions. The example shown in Figure 2 illustrates the case where "1 / 2" is assigned, which is the total number of conversions divided equally between the number of missing conversions (request r2 and request r3). 【0027】 In other words, the information processing device 100 estimates that "1 / 2" conversions occurred for both request r2 and request r3, and interpolates the missing parts. The information processing device 100 can also generate accuracy information using a predictive model. The predictive model is a model that has learned the relationship between requests and conversions for each attribute using, for example, the conversion history before the implementation of ITP and data from authorized users. 【0028】 The information processing device 100 then performs machine learning processing using the information obtained by interpolating the missing parts. The model generated by this machine learning processing is a delivery model that has learned the relationship between conversions for each attribute and the advertisements to be delivered, so that the CVR (Conversion Rate) is optimized. 【0029】 For example, as will be described later, the information processing device 100 can optimize the conversion rate (CVR) of each delivered advertisement by determining which advertisement to deliver to each user terminal 50 using a delivery model. 【0030】 Thus, the information processing device 100 according to the embodiment generates information indicating the probability that each user will perform a predetermined action regarding the missing parts of the conversion. Therefore, according to the information processing device 100 according to the embodiment, it is possible to generate finer-grained information from coarse-grained information in which user information has been anonymized. 【0031】 [2. Example of Information Processing Device Configuration] Next, an example of the configuration of the information processing device 100 will be described using Figure 3. Figure 3 is a block diagram showing an example of the configuration of the information processing device 100 according to this embodiment. As shown in Figure 3, the information processing device 100 includes a communication unit 110, a control unit 120, and a storage unit 130. 【0032】 The communication unit 110 is implemented, for example, by a NIC (Network Interface Card). The communication unit 110 transmits and receives information with external devices via a network N, such as various wireless communication networks or wired communication networks, including 4G (Generation), 5G, LTE (Long Term Evolution), WiFi (registered trademark), or wireless LAN (Local Area Network). 【0033】 The storage unit 130 is implemented by, for example, semiconductor memory elements such as RAM and flash memory, or storage devices such as hard disks and optical discs. The storage unit 130 also includes an advertising information database 131, a conversion database 132, an interpolation information database 133, and a distribution model database 134. 【0034】 The advertising information database 131 is a database that stores various advertising information related to advertisements requested by advertisers to be delivered. Advertising information includes, for example, information on the content to be displayed as an advertisement (images, videos, audio, etc.), delivery conditions (delivery target), delivery target number, target cost per conversion, and landing page URL. 【0035】 The conversion database 132 is a database that stores information about conversions. This information includes, for example, the association between the index of each attribute (e.g., f1 to f3) and the number of requests and conversions. 【0036】 Figure 4 shows an example of a conversion database 132 according to the embodiment. As shown in Figure 4, the conversion database 132 stores information on items such as "Campaign ID," "Request ID," "Attribute ID," and "Conversion (CV)" in a manner that associates them with each other. For the sake of explanation, information on a single "Campaign ID" is shown here. 【0037】 The "Campaign ID" is an identifier used to identify the campaign of the advertisement being delivered, and the "Request ID" is an identifier used to identify each request. The "Attribute ID" is an identifier (index) used to identify each attribute. Note that the example in Figure 4 shows the case where there are three attributes, "f1 to f3," but it is not limited to this. "Conversion" indicates whether or not a conversion occurred for the corresponding request. For example, it will be "1" if a conversion occurred for the corresponding request, and "0" if no conversion occurred. Note that if, for example, multiple conversions occur for a single request, "Conversion" will be the number of corresponding conversions. 【0038】 For example, Figure 4 shows that request r11 is included in the attribute of attribute ID "f1", and request r12 is included in the attribute of attribute ID "f2". Also, Figure 4 shows that one conversion occurred with request r11, meaning that no conversion occurred or the conversion is missing for request r12 or request r13. 【0039】 Returning to the explanation of Figure 3, let's describe the interpolation information database 133. The interpolation information database 133 is a database that stores interpolation information. Interpolation information is information obtained by interpolating missing conversion information in the conversion database 132. 【0040】 This section describes the delivery model database 134. The delivery model database 134 is a database that stores delivery models. A delivery model is a model that has learned the relationship between delivered advertisements and conversions for each attribute in order to optimize the CVR (Conversion Rate). 【0041】 For example, as will be described later, the information processing device 100 can optimize the conversion rate (CVR) of each delivered advertisement by using a delivery model to select the advertisement to deliver in response to each request from each user terminal 50. 【0042】 The control unit 120 is, for example, a controller, and is realized by the execution of various programs stored in the memory device inside the information processing device 100 using RAM as the working area by a CPU (Central Processing Unit) or MPU (Micro Processing Unit), etc. Alternatively, the control unit 120 is a controller and can be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). 【0043】 As shown in Figure 3, the control unit 120 includes an acquisition unit 121, a generation unit 122, a learning unit 123, a distribution unit 124, and a provision unit 125. 【0044】 The acquisition unit 121 acquires various information from the user terminal 50. The information acquired by the acquisition unit 121 includes, for example, requests for delivered advertisements and information related to access to delivered advertisements. 【0045】 Furthermore, the acquisition unit 121 acquires statistical information regarding predetermined actions taken by users who have selected one of several advertisements to be delivered. In addition, the acquisition unit 121 acquires authorized action information indicating whether or not a predetermined action was taken by an authorized user who has given permission among the users who have selected one of the advertisements. 【0046】 The statistical information here refers to, for example, the number of conversions at the campaign level, while the consent behavior information refers to conversions at the request level by consenting users. 【0047】 Based on the statistical information and consent behavior information acquired by the acquisition unit 121, the generation unit 122 generates information indicating the probability that each user who has selected one of the advertisements will perform a predetermined action. In other words, the generation unit 122 generates information indicating the probability that each user will achieve conversion. 【0048】 First, the generation unit 122 calculates the difference between the total number of conversions for the target campaign and the number of conversions measured for each request. The total number of conversions for the campaign can be obtained from statistical information, and the number of conversions measured for each request can be obtained from consent behavior information. 【0049】 For example, if the total number of conversions for a campaign ID is "5" and the total number of conversions measured for each request is "2", then the difference would be "5-2=3". 【0050】 Next, the generation unit 122 generates information indicating the probability for conversions in the conversion database 132 where the conversion is missing (CV = "0"). 【0051】 For example, as already explained, the generation unit 122 generates a value that indicates the probability of reaching a conversion by dividing the calculated difference equally by the number of missing values ​​corresponding to a conversion of "0". 【0052】 Furthermore, the generation unit 122 may generate predicted values ​​from the prediction model as information indicating the probability of conversion. For example, the prediction model is a model that has learned the relationship between requests and conversions using data from before the implementation of ITP or data from authorized users. 【0053】 For example, the generation unit 122 inputs information about each request where the conversion is "0" in the conversion database 132 into the prediction model, thereby generating information that indicates the probability of conversion for each request. 【0054】 Figure 5 shows an example of accuracy information using predicted values ​​according to the embodiment. For example, as shown in Figure 5, suppose the total number of conversions for a campaign is "2" and the total number of conversions for each request is "1". 【0055】 In this case, the difference between the total number of conversions for the campaign and the total number of conversions measured for each request is "2-1=1". The generation unit 122 inputs information about each request where the conversion that may be missing is "0" (for example, information about the attributes of each request) into the prediction model to calculate the predicted value of the conversion for each request. 【0056】 In other words, in this case, the more likely a request is to result in a conversion according to the prediction model, the higher the predicted value will be. The generation unit 122 may, for example, generate conversion probability information by taking into account the data of authorized users. 【0057】 For example, if request r22, which has a conversion of "0", is data from a user who has given permission, then the likelihood of a conversion occurring in request r22 is low, and the predicted value should be set to "0", or to a value as low as possible compared to other requests. In this way, data from users who have given permission for tracking may be weighted against data from users who have not given permission for tracking. 【0058】 Returning to the explanation of Figure 3, let's describe the learning unit 123. Based on the information generated by the generation unit 122, the learning unit 123 trains a model that estimates the probability that a user who selects each advertisement will perform a predetermined action. 【0059】 In other words, the learning unit 123 learns the distribution model stored in the distribution model database 134 based on the interpolation information stored in the interpolation information database 133. In other words, the learning unit 123 can continuously optimize the CVR by sequentially updating the distribution model. 【0060】 The distribution unit 124 delivers advertisements in response to requests from the user terminal 50. For example, the distribution unit 124 determines which advertisements to deliver to the user terminal 50 by inputting information about the attributes corresponding to the request into the distribution model. Once the distribution unit 124 has determined which advertisements (campaigns) to deliver in response to the request, it delivers the advertisements to the user terminal 50 that made the request. 【0061】 The provisioning unit 125 provides advertisers with advertiser reports. For example, the provisioning unit 125 uses interpolation information stored in the interpolation information database 133 to generate reports that include information on conversions (CV) and conversion rates (CVR). The advertiser reports generated by the provisioning unit 125 are provided to each advertiser via the communication unit 110. 【0062】 [3. Processing Flow] Next, the processing procedure executed by the information processing device 100 according to the embodiment will be described using Figure 6. Figure 6 is a flowchart showing the processing procedure executed by the information processing device 100 according to the embodiment. 【0063】 As shown in Figure 6, first, the information processing device 100 acquires statistical information and permission behavior information (step S101). Next, based on the statistical information and permission behavior information acquired in step S101, the information processing device 100 generates information indicating the probability for each missing location (step S102). 【0064】 Next, the information processing device 100 learns the delivery model based on the interpolated information, including the accuracy information, generated in step S102 (step S103). The information processing device 100 also provides a report to the advertiser based on the interpolated information, including the accuracy information, generated in step S102 (step S104). Finally, the information processing device 100 completes the series of processes. 【0065】 [4. Variations] By the way, the above-described method of operation explains the case where the target action is a conversion, but it is not limited to this. The target action may be any action, such as requesting information. 【0066】 [5. Effects] The information processing device 100 according to the above embodiment includes an acquisition unit 121 that acquires statistical information regarding target actions taken by users who have selected one of a plurality of advertisements to be delivered, and authorized action information indicating whether or not a target action was taken by authorized users who have given permission among the users who selected one of the advertisements, and a generation unit 122 that generates accuracy information indicating the probability that each user who selected one of the advertisements performed a target action, based on the statistical information and authorized action information acquired by the acquisition unit 121. 【0067】 Furthermore, the information processing device 100 includes a learning unit 123 that learns a model for estimating the probability that a user who selects each advertisement will perform the target action, based on the probability information generated by the generation unit 122 and the consent action information. 【0068】 Furthermore, the generation unit 122 generates accuracy information such that the difference between the number of target actions indicated by the statistical information and the number of target actions indicated by the consent action information equals the sum of the accuracy values. In addition, the generation unit 122 generates accuracy information for the unknown parts where the presence or absence of a target action is unknown, in the correspondence information which associates the presence or absence of a target action by the consenting user with the selection history of each user for the advertisement. 【0069】 Furthermore, the generation unit 122 generates the sum of the accuracies divided equally by the number of unknown locations as the accuracy information for each unknown location. In addition, the generation unit 122 generates predicted values ​​as accuracy information using a prediction model that has learned past measured values ​​regarding the presence or absence of the target behavior. 【0070】 Furthermore, the generation unit 122 generates accuracy information on a user attribute basis. The generation unit 122 generates accuracy information on a group basis classified using at least one of the following: the distribution platform where the advertisement is delivered, the URL, and the domain. 【0071】 By any or a combination of the above-described processes, the information processing device according to the present invention can generate finer-grained information from coarse-grained information in which user information has been anonymized. 【0072】 [6. Hardware Configuration] Furthermore, the information processing device 100 according to the above embodiment is realized by a computer 1000 having a configuration such as that shown in Figure 7. Figure 7 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device 100 according to the embodiment. The computer 1000 has a 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. 【0073】 The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, controlling various components. The ROM 1300 stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware. 【0074】 The HDD1400 stores programs executed by the CPU1100, as well as data used by such programs. The communication interface1500 receives data from other devices via the network (communication network) N and sends it to the CPU1100, and transmits data generated by the CPU1100 to other devices via the network N. 【0075】 The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice (in Figure 7, output devices and input devices are collectively referred to as "input / output devices") via the input / output interface 1600. The CPU 1100 acquires data from input devices via the input / output interface 1600. The CPU 1100 also outputs the generated data to output devices via the input / output interface 1600. 【0076】 The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the 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 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. 【0077】 For example, when computer 1000 functions as an information processing device 100 according to the embodiment, the CPU 1100 of computer 1000 realizes the functions of the control unit 120 by executing programs loaded on RAM 1200. The CPU 1100 of computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be obtained from other devices via a network N. 【0078】 Although some embodiments of the present invention have been described in detail above with reference to the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention. 【0079】 [7. Other] Furthermore, among the processes described in the above embodiments and modifications, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown. 【0080】 Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions. 【0081】 Furthermore, the embodiments and modifications described above can be combined as appropriate, provided that the processing content is not inconsistent. 【0082】 Furthermore, the terms "section, module, unit" mentioned above can be replaced with "means" or "circuit," etc. For example, the acquisition unit can be replaced with acquisition means or acquisition circuit. [Explanation of symbols] 【0083】 1. Information Processing System 50 User terminals 100 Information Processing Devices 110 Communications Department 120 Control Unit 121 Acquisition Department 122 Generation part 123 Learning Department 124 Distribution Department 125 Provision Department 130 Storage section 131 Advertising Information Database 132 Conversion Database 133 Interpolation Information Database 134 Distribution Model Database 200 advertiser servers

Claims

[Claim 1] An acquisition unit that acquires statistical information regarding target actions taken by users who selected one of several advertisements to be delivered, and authorized action information indicating whether or not the target action was taken by authorized users who selected one of the advertisements and gave their consent. Based on the statistical information and consent behavior information acquired by the acquisition unit, a generation unit generates accuracy information indicating the probability that each user who selected one of the advertisements performed the target action. Equipped with, The generating unit is In correspondence information that associates the presence or absence of the target action by the authorized user with each user's selection history for advertisements, the probability information for the unknown parts where the presence or absence of the target action is unknown is generated, and the sum of the probabilities is divided equally by the number of unknown parts to generate the probability information for each of the unknown parts. An information processing device characterized by the following. [Claim 2] The generating unit is The accuracy information is generated such that the difference between the number of target actions indicated by the statistical information and the number of target actions indicated by the permission action information equals the sum of the accuracy values. The information processing apparatus according to claim 1, characterized by the following: [Claim 3] The generating unit is The accuracy information is generated on a per-user attribute basis. The information processing apparatus according to claim 1, characterized by the following: [Claim 4] The generating unit is The accuracy information is generated for attribute units classified using at least one of the distribution platform, URL, and domain on which the advertisement is delivered. The information processing apparatus according to claim 1, characterized by the following: [Claim 5] A method of information processing performed by a computer, The process includes obtaining statistical information regarding target actions taken by users who selected one of several advertisements to be delivered, and authorized action information indicating whether or not the target action was taken by authorized users who selected one of the advertisements and gave their consent. A generation step that generates accuracy information indicating the probability that each user who selected one of the advertisements performed the target action, based on the statistical information and consent behavior information obtained by the acquisition step, Includes, The aforementioned generation step is In correspondence information that associates the presence or absence of the target action by the authorized user with each user's selection history for advertisements, the probability information for the unknown parts where the presence or absence of the target action is unknown is generated, and the sum of the probabilities is divided equally by the number of unknown parts to generate the probability information for each of the unknown parts. An information processing method characterized by the following. [Claim 6] A procedure for obtaining statistical information regarding target actions taken by users who selected one of several advertisements to be delivered, and authorized action information indicating whether or not the target action was taken by authorized users who selected one of the advertisements and gave their consent, A generation procedure that generates probability information indicating the probability that each user who selected one of the advertisements performed the target action, based on the statistical information and consent behavior information obtained by the acquisition procedure, Have the computer run it, The aforementioned generation procedure is: In correspondence information that associates the presence or absence of the target action by the authorized user with each user's selection history for advertisements, the probability information for the unknown parts where the presence or absence of the target action is unknown is generated, and the sum of the probabilities is divided equally by the number of unknown parts to generate the probability information for each of the unknown parts. An information processing program characterized by the following. [Claim 7] An acquisition unit that acquires statistical information regarding target actions taken by users who have selected one of a plurality of advertisements to be delivered, and authorized action information indicating whether or not a target action was taken by an authorized user who has given permission among the users who have selected one of the advertisements, Based on the statistical information and consent behavior information acquired by the acquisition unit, a generation unit generates accuracy information indicating the probability that each user who selected one of the advertisements performed the target action. Equipped with, The generating unit is In correspondence information that associates the presence or absence of a target action by the authorized user with each user's selection history for advertisements, the accuracy information is generated for the unknown parts where the presence or absence of a target action is unknown, and a predictive model that has learned past measured values ​​regarding the presence or absence of a target action is used, with user attribute information and the presence or absence of the target action as training data, and the predictive value generated by the predictive model that predicts the presence or absence of a target action from the input attribute information is used as the accuracy information. An information processing device characterized by the following. [Claim 8] A computer-based information processing method, The process includes obtaining statistical information regarding target actions taken by users who selected one of several advertisements to be delivered, and authorized action information indicating whether or not the target action was taken by authorized users who selected one of the advertisements and gave their consent. A generation step that generates accuracy information indicating the probability that each user who selected one of the advertisements performed the target action, based on the statistical information and consent behavior information obtained by the acquisition step, Includes, The aforementioned generation step is In correspondence information that associates the presence or absence of a target action by the authorized user with each user's selection history for advertisements, the accuracy information is generated for the unknown parts where the presence or absence of a target action is unknown, and a predictive model that has learned past measured values ​​regarding the presence or absence of a target action is used, with user attribute information and the presence or absence of the target action as training data, and the predictive value generated by the predictive model that predicts the presence or absence of a target action from the input attribute information is used as the accuracy information. An information processing method characterized by the following. [Claim 9] A procedure for obtaining statistical information regarding target actions taken by users who selected one of a plurality of advertisements to be delivered, and authorized action information indicating whether or not a target action was taken by an authorized user who selected one of the advertisements and gave permission; A generation procedure that generates probability information indicating the probability that each user who selected one of the advertisements performed the target action, based on the statistical information and consent behavior information obtained by the acquisition procedure, Have the computer run it, The aforementioned generation procedure is: In correspondence information that associates the presence or absence of a target action by the authorized user with each user's selection history for advertisements, the accuracy information is generated for the unknown parts where the presence or absence of a target action is unknown, and a predictive model that has learned past measured values ​​regarding the presence or absence of a target action is used, with user attribute information and the presence or absence of the target action as training data, and the predictive value generated by the predictive model that predicts the presence or absence of a target action from the input attribute information is used as the accuracy information. An information processing program characterized by the following.

Images