Information processing system and information processing method

Abstract

One aspect of the present disclosure relates to obtaining a dataset of a plurality of consumers, the dataset having, for each consumer, an identification code of the respective consumer and characteristic data. Based on a degree of usage of each digital platform determined from the respective characteristic data of the plurality of consumers, one or more of the plurality of consumers are assigned to the respective digital platform for each digital platform.

Description

[0001] Cross-references to related applications

[0002] This international application claims priority to Japanese Patent Application No. 2023-044019, filed on March 20, 2023, with the Japan Patent Office, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to information processing systems and information processing methods. Background Technology

[0004] In recent years, numerous digital platform operators have provided digital platforms as venues for offering digital services to users on the internet. For example, they offer digital services such as search services, content delivery services, online shopping malls, and social networking services to users through multiple digital platforms (see, for example, Patent Document 1).

[0005] Users can access the digital platform through information terminals and use the services provided by the digital platform. The digital platform provider's system then accumulates information about users based on their usage of the digital platform.

[0006] Digital platform operators that provide advertising delivery services or information services utilizing accumulated information are also known. For example, in advertising delivery services, advertisements are targeted to users estimated to have high advertising effectiveness based on accumulated information. In information service provision, information accumulated on the digital platform is provided to requesters as, for example, statistical information.

[0007] Existing technical documents

[0008] Patent documents

[0009] Patent Document 1: Japanese Patent Application Publication No. 2022-153536 Summary of the Invention

[0010] The problem the invention aims to solve

[0011] Additionally, it can be assumed that when advertising to consumers is delivered through advertising services on multiple digital platforms, businesses using these services must request advertising on one or more of the multiple digital platforms to ensure that the advertising is effectively delivered to actual and potential customers.

[0012] Alternatively, it can be considered that when a service collects customer information through multiple digital platforms, the enterprise using the service needs to provide customer information to one or more of the requests from these digital platforms. In such cases, it may be necessary to select a digital platform.

[0013] Therefore, according to one aspect of this disclosure, in the context of using digital services such as advertising delivery services and information service provision, it is desirable to provide a new technology that is useful for selecting digital platforms, particularly for allocating multiple consumers to multiple digital platforms.

[0014] Technical solutions to the problem

[0015] According to one aspect of this disclosure, an information processing system is provided. The information processing system includes an acquisition unit, an allocation unit, and a list generation unit. The acquisition unit is configured to acquire a dataset concerning multiple consumers.

[0016] The dataset is tailored to each consumer, containing their unique identifier and characteristic data. This characteristic data includes information about multiple digital platforms and the extent to which the consumer uses each platform.

[0017] The allocation department is configured to allocate one or more consumers to the corresponding digital platform based on the degree of usage of each digital platform determined by the characteristic data of multiple consumers.

[0018] The list generation unit is configured to generate a consumer list for each digital platform in order to send data to the operator system of the corresponding digital platform, wherein the consumer list records the identification code of one or more consumers assigned to the corresponding digital platform.

[0019] According to one aspect of this disclosure, the information processing system may further include a sending unit to replace the list generation unit, or the information processing system may include a sending unit in addition to the list generation unit. The sending unit may be configured to send identification codes assigned to one or more consumers of each digital platform to the operator system of that digital platform.

[0020] The more a consumer uses a particular digital platform, the more information about that consumer accumulates. Therefore, by allocating consumers using the methods described above, it is possible to distribute consumers among multiple digital platforms that possess richer information about them. This allows access to high-quality advertising services, such as those utilizing accumulated information, from digital platform operators. Thus, according to one aspect of this disclosure, a new technology can be provided that facilitates the allocation of multiple consumers to multiple digital platforms.

[0021] According to one aspect of this disclosure, the allocation unit may be configured to allocate more than one consumer to each of the multiple digital platforms in such a way that multiple consumers are preferentially allocated to the platforms where the corresponding consumers have a high usage level.

[0022] According to one aspect of this disclosure, the allocation unit can be configured to calculate a score for each consumer across multiple digital platforms based on the consumer's usage of those platforms. The allocation unit can then, based on this score, allocate one or more consumers to each digital platform.

[0023] According to one aspect of this disclosure, the allocation unit may be configured to calculate scores in a weighted manner based on the demographic attributes of the corresponding consumers. According to another aspect of this disclosure, the allocation unit may be configured to use weights to calculate the scores, wherein the weights correspond to a specified consumer distribution with respect to the demographic attributes.

[0024] For example, when considering advertising through digital platforms, by weighting scores according to the distribution of desired demographic attributes among the target audience, consumers can be allocated to various digital platforms based on the desired distribution. This enables effective advertising.

[0025] According to one scheme of this disclosure, the allocation department can calculate a score for each consumer on each of a plurality of digital platforms based on the degree to which the consumer uses the corresponding digital platform.

[0026] The allocation department can be configured to allocate one or more consumers to each digital platform in a manner that assigns the corresponding consumer to a digital platform among multiple digital platforms where the score reaches or exceeds a benchmark.

[0027] The allocation department can be configured such that, when there are two or more digital platforms with scores above the benchmark for a given consumer, it allocates one or more consumers to each of the two or more digital platforms with scores above the benchmark.

[0028] The allocation unit can be configured to weight scores based on the demographic attributes of the corresponding consumers when calculating scores, using weights that correspond to the specified consumer distribution with respect to the demographic attributes.

[0029] According to one aspect of this disclosure, the information processing system may include a dataset generation unit that generates datasets about multiple consumers. An acquisition unit may be configured to acquire the datasets about multiple consumers generated by the dataset generation unit.

[0030] According to one aspect of this disclosure, the dataset generation unit can be configured to generate a dataset about multiple consumers based on a first dataset about multiple first consumers and a second dataset about multiple second consumers.

[0031] The first dataset may be a dataset that does not include information about the usage levels of at least some of the multiple digital platforms. The second dataset may be a dataset that includes information about the usage levels of at least some of the multiple digital platforms.

[0032] The dataset generation unit can be configured to estimate the usage of at least a portion of the digital platforms among multiple digital platforms by multiple first consumers, and add information about the estimated usage to the first dataset, thereby generating a dataset about the multiple consumers.

[0033] By generating and acquiring datasets in the manner described above, it is possible to appropriately allocate consumer sets, including those lacking information about usage levels, to multiple digital platforms.

[0034] According to one aspect of this disclosure, the list generation unit can be configured to, with reference to a database of consumer groups, expand one or more consumers assigned to a corresponding digital platform into two or more consumers with similar or identical characteristics for each digital platform, wherein the database contains an identification code and characteristic data for each consumer. The list generation unit can be configured to generate a consumer list describing the identification codes of the expanded two or more consumers.

[0035] According to one aspect of this disclosure, the sending unit may be configured to send submission data to the operator system of the corresponding digital platform for each digital platform, wherein the submission data records the identification code of one or more consumers assigned to the corresponding digital platform as the identification code of the consumer targeted by the advertisement.

[0036] According to one aspect of this disclosure, each of a plurality of digital platforms may be configured to have a database that associates and records the characteristics of multiple users with the corresponding user identification codes, and based on the database, expands the target audience for advertising from one or more consumers whose identification codes are recorded in the submitted data to a group of users whose characteristics are related to the consumers who are the target audience for advertising, thereby targeting advertising to that user group.

[0037] According to one aspect of this disclosure, the sending unit can be configured to specify an upper limit on the expanded number of users targeted for advertising when sending the submitted data. By specifying this upper limit, the advertising requester can appropriately control the expanded targeting.

[0038] According to one aspect of this disclosure, the sending unit may be configured to specify an upper limit on the number of users who are the targets of advertising in each division of the demographic attributes based on the distribution of demographic attributes of multiple consumers, wherein the distribution of demographic attributes is determined based on the dataset of multiple consumers obtained by the acquisition unit.

[0039] By specifying an expansion limit based on the distribution of demographic attributes, the distribution of demographic attributes in the expanded ad targeting audience can be controlled to achieve a desired distribution. For example, it is possible to control the distribution of demographic attributes in the ad targeting audience to make it approximate the distribution of the population.

[0040] According to one aspect of this disclosure, the sending unit may be configured to send one or more consumer identification codes assigned to the corresponding digital platform, along with one or more consumer characteristic data, to the operator system of the corresponding digital platform for each digital platform.

[0041] Digital platforms can leverage their existing user characteristic data to expand the received characteristic data and use it for advertising services, information services, and so on.

[0042] According to one aspect of this disclosure, a computer program can be provided for enabling a computer to implement at least a portion of the acquisition unit, allocation unit, and list generation unit in the aforementioned information processing system. According to one aspect of this disclosure, a computer program can be provided for enabling a computer to implement at least a portion of the acquisition unit, allocation unit, and transmission unit in the aforementioned information processing system. The computer program can be recorded on a computer-readable non-transitory recording medium.

[0043] According to one aspect of this disclosure, an information processing method corresponding to the aforementioned information processing system can be provided. The information processing method can be an information processing method executed by a computer.

[0044] According to one aspect of this disclosure, the information processing method includes acquiring a dataset of multiple consumers, the dataset including, for each consumer, an identification code and characteristic data of the respective consumer, and the characteristic data of the respective consumer including information about multiple digital platforms and the degree to which the respective consumer uses each digital platform.

[0045] Information processing methods may include assigning one or more consumers to a corresponding digital platform based on the degree of usage of each digital platform determined by the characteristic data of multiple consumers.

[0046] The information processing method may include, for each digital platform, generating a consumer list for sending data to the operator system of the corresponding digital platform, wherein the consumer list records identification codes assigned to one or more consumers of the corresponding digital platform.

[0047] Information processing methods may include, instead of generating as described above or in addition to generating as described above, sending identification codes assigned to one or more consumers of the respective digital platform to the operator system of the respective digital platform for each digital platform.

[0048] According to one scheme of this disclosure, allocation may include, for each consumer, calculating a score for each of a plurality of digital platforms based on the extent to which the consumer uses the respective digital platform.

[0049] The allocation may include allocating a corresponding consumer to a digital platform with a score above a benchmark among multiple digital platforms, and, when there are two or more digital platforms with a score above a benchmark for a corresponding consumer, allocating the corresponding consumer to two or more digital platforms with a score above a benchmark for each digital platform.

[0050] The allocation may include weighting the scores based on the demographic attributes of the corresponding consumers when calculating the scores, where the weights correspond to the consumer distribution specified with respect to the demographic attributes.

[0051] According to one aspect of this disclosure, a computer program can be provided, comprising instructions for causing a computer to at least partially perform the information processing method described above when executed by the computer. According to another aspect of this disclosure, a computer-readable recording medium for storing the computer program can be provided. The recording medium may be a non-transitory and tangible computer-readable recording medium.

[0052] According to one aspect of this disclosure, a computer program can be provided that, after being loaded and executed by a computer, can perform the aforementioned information processing method. A computer program product with the computer program built-in can be provided. A computer-readable recording medium storing the computer program can be provided. According to one aspect of this disclosure, a computer-readable storage medium can be provided, storing a computer program / instructions that, when executed by a processor, implement the steps of the aforementioned information processing method. According to one aspect of this disclosure, a computer program product can be provided, including a computer program / instructions that, when executed by a processor, implement the steps of the aforementioned information processing method. Attached Figure Description

[0053] Figure 1This is a block diagram showing the structure of an information processing system.

[0054] Figure 2 It is a flowchart of data generation and processing executed by the processor.

[0055] Figure 3 It is a diagram illustrating the structure of the dataset.

[0056] Figure 4 This is a flowchart (1) illustrating the selection and transmission process performed by the processor.

[0057] Figure 5 This is a flowchart (2) showing the selection send process performed by the processor.

[0058] Figure 6 This is a flowchart illustrating the deployment-related processes performed by the operator's system.

[0059] Figure 7 This is a flowchart illustrating the data extension processing performed by the operator's system.

[0060] Figure 8 This is a flowchart illustrating a portion of the selection and sending process in a variant example.

[0061] Figure 9 This is a diagram illustrating an example of expanding consumers based on a consumer database.

[0062] Explanation of reference numerals in the attached figures

[0063] 1… Information processing system; 5… Carrier system; 11… Processor; 13… Memory; 15… Memory; 15A, 15B, 15C… Datasets; 16… Consumer database; 17… User interface; 19… Communication interface; Pr… Computer program. Detailed Implementation

[0064] Exemplary embodiments of this disclosure are described below with reference to the accompanying drawings.

[0065] The information processing system 1 of this embodiment is constructed by installing a dedicated computer program Pr into a general-purpose computer. For example... Figure 1 As shown, the information processing system 1 includes a processor 11, a memory 13, a storage device 15, a user interface 17, and a communication interface 19.

[0066] Processor 11 executes processing according to computer program Pr stored in memory 15. Memory 13 is a primary storage device with RAM (random access memory) and is used as a work area when processor 11 executes processing.

[0067] The memory 15 is a secondary storage device, such as a hard disk drive or a solid-state drive, which stores not only the computer program Pr, but also various data provided when the computer program Pr is processed.

[0068] The user interface 17 includes: an input device for inputting operation signals from an operator of the operation information processing system 1 to the processor 11; and a display for displaying various information to the operator. Examples of input devices include a keyboard and a pointing device.

[0069] The communication interface 19 includes a LAN (Local Area Network) interface and a USB (Universal Serial Bus) interface. The information processing system 1 is configured to communicate with multiple digital platform operator systems 5 via the communication interface 19.

[0070] On the Internet, which serves as an example of a wide area network, multiple digital platforms are provided by multiple digital platform operators as venues for providing digital services to users.

[0071] For example, providing users with digital services such as search services, content delivery services, online shopping malls, and social networking services through multiple digital platforms.

[0072] Users can access the digital platform through information terminals and use the services provided by the digital platform. The digital platform accumulates information about users based on their usage.

[0073] In the following text, the digital platform will be referred to as "Platform" for short, and the digital platform operator as "PF Operator." PF stands for "platform." PF Operator refers to an operator that operates a digital platform.

[0074] The processor 11 generates an expanded dataset 15C by extending the first dataset 15A with the second dataset 15B through processing according to the computer program Pr.

[0075] For each platform, processor 11 selects one or more consumers based on extended dataset 15C and sends descriptive data describing the selected consumers to the corresponding platform's operator system 5. Operator system 5, as used herein, refers to the computer system of the operator operating the corresponding platform. Examples of operator system 5 may include the corresponding platform's data clean room.

[0076] Specifically, in order to generate the extended dataset 15C, the processor 11 can perform operations based on instructions input by the operator through the user interface 17. Figure 2 The data generation and processing are shown.

[0077] When the data generation process begins, the processor 11 acquires the first dataset 15A (S110). The processor 11 can read the first dataset 15A pre-stored in the memory 15. Alternatively, the processor 11 can acquire the first dataset 15A from a first external system via the communication interface 19.

[0078] The first dataset 15A is, for example, panel data collected through a panel survey. The first dataset 15A contains first feature data for each consumer of a first set of consumers corresponding to a panel, and this first feature data is associated with the identification data of the corresponding consumer.

[0079] The first dataset 15A contains consumer identification data that records multiple IDs as the corresponding consumer's identification code. These IDs include the hash value of the email address used by the consumer, the advertising ID of the information terminal used by the consumer, and the cookie ID of the information terminal used by the consumer. The first dataset 15A may also contain demographic attributes describing the corresponding consumer, such as gender, age group, and occupation, as the primary feature data for each consumer.

[0080] In the subsequent S120, the processor 11 acquires the second dataset 15B. The processor 11 is capable of reading the second dataset 15B pre-stored in the memory 15. Alternatively, the processor 11 is capable of acquiring the second dataset 15B from a second external system via the communication interface 19.

[0081] The second dataset 15B contains second feature data for each consumer in the second set of consumers, and this second feature data is associated with the corresponding consumer's identification data. Similar to the first dataset 15A, the identification data of each consumer in the second dataset 15B can also use the aforementioned multiple IDs of the corresponding consumer as the consumer's identification code.

[0082] The second dataset 15B includes secondary feature data for each consumer, including their PF usage performance data, media consumption data, and purchase data. Regarding multiple platforms, the PF usage performance data describes the consumer's usage performance on each platform.

[0083] Regarding multiple media outlets, media exposure data indicates whether a consumer has been exposed to that media outlet for each individual. Regarding multiple products, purchase data indicates whether a consumer has purchased that product for each individual, or the quantity purchased.

[0084] In the subsequent S130, the processor 11 uses data fusion technology to combine the acquired first dataset 15A and second dataset 15B to generate an extended dataset 15C.

[0085] For example, processor 11 can determine first feature data and second feature data that are associated with the same identification code, based on the consumer identification codes included in the first dataset 15A and the second dataset 15B. Processor 11 can combine the first feature data with the second feature data that is associated with the same identification code as the first feature data, thereby combining the first dataset 15A and the second dataset 15B.

[0086] The extended dataset 15C contains feature data for each consumer in the first consumer set corresponding to the panel. This feature data combines the first feature data as basic data, PF usage performance data, media contact data, and purchase data as extended data.

[0087] Within the first consumer set, there exist consumers who belong to the second consumer set and consumers who do not belong to the second consumer set. That is, in the extended dataset 15C, there are consumers with missing performance data regarding PF usage.

[0088] Therefore, in S140, processor 11 performs the aforementioned missing data completion process. For example, processor 11 can estimate the parameters of the PF usage performance data through statistical processing. According to other examples, processor 11 can use feature data about a set of consumers without missing PF usage performance data as training data to train an estimation model using a machine learning algorithm. This estimation model is used to infer the parameters of the PF usage performance data based on the parameters of the basic data.

[0089] Processor 11 inputs the values ​​of various parameters of the basic data of consumers with missing PF usage performance data into the trained estimation model, and obtains estimated values ​​of various parameters related to the PF usage performance data of the corresponding consumers as the output of the estimation model. Processor 11 can complete the missing PF usage performance data in extended dataset 15C by adding PF usage performance data describing these estimated values ​​to extended dataset 15C.

[0090] As described above, the processor 11 stores the missing-completed extended dataset 15C in the memory 15. Hereinafter, the missing-completed extended dataset 15C will be simply referred to as dataset 15C.

[0091] Figure 3The structure of dataset 15C is shown. As illustrated, dataset 15C, for each consumer, contains basic consumer data, PF usage performance data, media contact data, and purchase data associated with the consumer's identification data. Similar to the first dataset 15A, the consumer identification data records multiple IDs as the consumer's identification code.

[0092] As described above, the basic data in dataset 15C corresponds to the feature data of the first dataset 15A and represents the demographic attributes of the corresponding consumers. PF uses performance data, media exposure data, and purchase data as the second feature data of the corresponding consumers in the second dataset 15B, or data with missing completion.

[0093] PF performance data includes PF performance data for each platform. Each PF performance data point represents the usage performance of the corresponding consumer (i.e., the consumer identified by the aforementioned identification code) on the corresponding platform, specifically representing the degree of usage. The degree of usage is a numerical value related to the level of usage performance.

[0094] Figure 3 The 1PF performance data shown represents the data on the usage performance of the corresponding consumer on the 1st platform, the 2ndPF performance data represents the data on the usage performance of the corresponding consumer on the 2nd platform, and the 3rdPF performance data represents the data on the usage performance of the corresponding consumer on the 3rd platform.

[0095] PF performance data records the values ​​of PF usage rate F1, PF main user usage rate F2, PF login usage rate F3, and PF usage frequency F4, which are used as numerical values ​​to represent the degree of use of the corresponding consumer on the corresponding platform.

[0096] PF usage rate F1 is a parameter that represents, in probabilistic terms, whether a consumer uses a particular platform. If a consumer has already used the platform, PF usage rate F1 can be 1; otherwise, it can be 0.

[0097] PF primary usage rate F2 is a parameter that represents, in probabilistic terms, whether a consumer primarily uses a particular platform. When a consumer primarily uses the platform, PF primary usage rate F2 can be 1, while in other cases it can be 0. Whether a consumer primarily uses a particular platform can be determined by whether they use its main services, whether they have used the platform within the most recent specified period, or by their relative frequency of use compared to other platforms.

[0098] PF Login Usage Rate F3 represents the probability of whether a consumer has logged in and used the platform. If a consumer has logged in and used the platform, PF Login Usage Rate F3 can be 1; otherwise, it can be 0. For example, if a consumer has used the platform without logging in, the corresponding PF Login Usage Rate F3 is 0.

[0099] PF usage frequency F4 is a parameter representing the frequency of a consumer's use of a particular platform according to a predefined scale, and the higher the usage frequency, the larger its positive value. PF usage frequency F4 can, for example, take values ​​corresponding to the usage frequency within a range of 0 to 1. Usage frequency can represent, for example, the value corresponding to the platform's usage time or number of visits within a certain period.

[0100] When processor 11 receives an instruction from the operator to perform the selection and transmission process via user interface 17, processor 11 executes... Figure 4 and Figure 5 The selective transmission process is shown. In this selective transmission process, processor 11, for each platform, extracts the set of consumers whose information should be sent to the operator system 5 of the corresponding platform from the first consumer set whose feature data is registered in dataset 15C.

[0101] The processor 11 sends description data describing the identification code and characteristics of each consumer to the operator system 5 of the corresponding platform as description data describing the set of consumers of the extracted sending objects.

[0102] when Figure 4 and Figure 5 When the selection and sending process begins, processor 11 refers to dataset 15C and extracts multiple consumers belonging to the group designated as the sending object from the first consumer set (S210).

[0103] The target group designated for delivery may be a consumer group categorized based on demographic attributes, psychometric attributes, behavioral characteristics, or a combination thereof. Examples of demographic attributes include gender, age group, and occupation. Examples of psychometric attributes include attributes related to interests, concerns, and preferences. Examples of behavioral characteristics include behavioral characteristics related to purchasing and media consumption.

[0104] When men are designated as the target group, processor 11 extracts a set of male consumers from the first consumer set as the target group. When a group interested in SUVs (Sport Utility Vehicles) is designated as the target group, processor 11 extracts a set of consumers interested in SUVs from the first consumer set. When the operator specifies the number of consumers to be extracted as the size of the target group, processor 11 extracts the specified number of consumers from the first consumer set A, prioritizing consumers with a high degree of conformity to the target group.

[0105] In the subsequent S220, the processor 11 selects one consumer from the multiple consumers extracted in S210 as the consumer to be processed. Hereinafter, the selected consumer as the object of processing will also be referred to as the object consumer.

[0106] In the subsequent S230, the processor 11 refers to the characteristic data of the object in the dataset 15C, and in particular to the PF usage performance data, to calculate a score Scr for the usage performance of each platform related to the object.

[0107] For example, processor 11 calculates the score Scr for each platform in such a way that the higher the usage frequency of the target user, the larger the value. For example, for each platform, processor 11 can use the target user's PF usage rate F1, PF main usage rate F2, PF login usage rate F3, PF usage frequency F4, and predefined coefficients k1, k3, k4 to calculate the score Scr according to the following mathematical formula.

[0108] Scr=k1·F1·F2·(1+k3·F3)·(1+k4·F4).

[0109] The coefficients k1, k3, and k4 in the above mathematical formula are set according to each platform. Coefficients k1, k3, and k4 can be defined based on the number of users and logged-in users on the respective platform. For example, coefficients k1, k3, and k4 can be set to suppress statistical differences in the score Scr between platforms. However, coefficients k1, k3, and k4 can also be defined as values ​​shared by multiple platforms.

[0110] In the above mathematical formula, the values ​​identified by the object's PF performance data on the corresponding platform according to dataset 15C are used as PF usage rate F1, PF main usage rate F2, PF login usage rate F3, and PF usage frequency F4.

[0111] According to the above mathematical formula, when either the PF utilization rate F1 or the PF main utilization rate F2 is 0, the score Scr is calculated as 0. If both the PF utilization rate F1 and the PF main utilization rate F2 are 1, then the score Scr is calculated as a value corresponding to the PF login utilization rate F3 and the PF usage frequency F4, and is greater than zero.

[0112] Among platforms with a PF login usage rate (F3) of 1, the platform with the higher usage frequency is assigned a higher score (Scr). Among platforms with the same PF usage frequency (F4), the platform with login usage is assigned a higher score (Scr) compared to the platform that has never logged in.

[0113] In the subsequent S240, the processor 11 assigns the object to one or more platforms that meet the specified conditions based on the score Scr of each platform. According to the first example, the processor 11 assigns the object to the platform with the highest score Scr among the multiple platforms.

[0114] According to the second example, processor 11 assigns objects to one or more platforms among multiple platforms whose score Scr is above the benchmark. As mentioned above, processor 11 prioritizes assigning objects to platforms with higher usage.

[0115] Based on the second example, there may be cases where an object is not assigned to any platform. A variation of the second example could be the following: if no platform has a score (Scr) above the baseline, then the object is assigned to the platform with the highest score (Scr).

[0116] Subsequently, processor 11 determines whether all consumers extracted in S210 have been selected as objects (S250). If it is determined that there are still consumers not selected as objects (which is not the case in S250), processor 11 selects the consumers not yet selected as objects from the multiple consumers extracted in S210 as new objects (S220). Subsequently, processor 11 performs the processing from S230 onwards for the new object and allocates a platform to the object.

[0117] If the processor 11 determines that the selection of all consumers has been completed (yes in S250), then in the subsequent S260, the processor 11 selects one of the multiple platforms as the platform to which the data is sent.

[0118] In the subsequent S270, the processor 11 generates description data as transmission data to be sent to the selected platform, wherein the description data describes one or more consumers assigned to the selected platform. This description data corresponds to a consumer list.

[0119] The description data pertains to more than one assigned consumer and stores a consumer identification code and characteristic data for each consumer. The consumer identification code stored in the description data may be part or all of multiple IDs recorded as consumer identification codes in dataset 15C.

[0120] For example, the description data can record the advertisement ID as an identifier for each consumer. Some or all of the characteristic data of the corresponding consumers in dataset 15C can be stored in the description data as characteristic data for each consumer.

[0121] For example, as characteristic data for the corresponding consumer, the data may only associate and store the basic data included in dataset 15C with the identification code. The data may also store one or more of the following: PF usage performance data, media contact data, and purchase data. The data may also only store the consumer's identification code (e.g., advertising ID) without storing the corresponding consumer's characteristic data.

[0122] According to one example, the explanatory data identifies more than one consumer as an object of advertising delivered through the advertising delivery service provided by the respective platform operator. In this case, the explanatory data may correspond to the submission data submitted to the respective platform operator. According to one example, the explanatory data identifies more than one consumer as an object of analysis performed in the data cleanroom provided by the respective platform operator.

[0123] In subsequent step S280, processor 11 sends the instruction data generated in S270 to the operator system 5 of the selected platform. For example, processor 11 is capable of sending the instruction data as submission data to the operator system 5 of the platform. According to other examples, processor 11 is capable of sending the instruction data to the data cleanroom of the selected platform.

[0124] In the subsequent S290, the processor 11 determines whether all platforms have been selected as data transmission targets. If it is determined that there are still platforms that have not been selected as data transmission targets (no in S290), the processor 11 moves to S260, reselects the platforms that have not yet been selected as data transmission targets, generates the aforementioned explanatory data (S270), and then sends it to the operator system 5 of the selected platform (S280).

[0125] When the processor 11 determines that all platforms have been selected (yes in S290), the selection and transmission process ends.

[0126] The operator system 5 of each platform can execute based on the aforementioned data from the information processing system 1. Figure 6 The processing shown. Figure 6This is a flowchart of the delivery-related processing performed by the operator system 5 when the aforementioned explanatory data is sent from the information processing system 1 as submission data.

[0127] In the delivery-related processing, the operator system 5 receives the submitted data from the information processing system 1 (S310) and expands the advertising delivery targets mentioned in the submitted data based on the database of the operator system 5 (S320).

[0128] The database of operator system 5 regarding users of the corresponding platform may contain characteristic data for each user associated with a user's identifier (e.g., advertising ID). In the following text, to distinguish it from characteristic data included in submitted data, the characteristic data held by operator system 5 will be referred to as retained characteristic data.

[0129] In S320, the operator system 5 can retrieve the retained feature data of users presumed to be the same person from its database based on the consumer's identification code (e.g., advertising ID) and optionally also based on the consumer's feature data, for each consumer described in the submitted data.

[0130] The operator system 5 expands the advertising target by adding more than one user to the advertising target in its database, wherein the more than one user has retained characteristic data similar to the retained characteristic data of users presumed to be the same person.

[0131] The submitted data can record the upper limit of the number of people after the expansion of the advertising target audience. The upper limit can be recorded through the processing of S270 executed by the processor 11 of the information processing system 1.

[0132] You can specify an upper limit for each segment. This segment can be a consumer segment based on demographic attributes, specifically gender and age. In other words, for each segment, you can specify the upper limit of the expanded target audience for the advertisement in the submitted data.

[0133] The first consumer set in the panel is a sample from the parent body. The demographic composition of the first consumer set may differ from that of the parent body. Without setting an upper limit, the demographic composition of the expanded ad targeting audience may deviate from that of the parent body.

[0134] Therefore, an expanded upper limit on the number of people can be set for each segment to ensure that the demographic composition of the expanded consumer set matches that of the parent population. For example, if each consumer in a segment of the first consumer set represents N people in the parent population, the expanded upper limit on the number of people in each segment can be set to a value proportional to N. Information about the demographic composition can be pre-stored in memory 15.

[0135] In S330, the operator system 5 delivers advertisements to each user belonging to the extended advertising target group. The advertisement content to be delivered can be provided to the operator system 5 in advance from the information processing system 1.

[0136] On the other hand, when the aforementioned explanatory data is used as consumer explanatory data for analysis and provided from information processing system 1 to operator systems 5 of various platforms, operator system 5 executes... Figure 7 The data expansion process is shown.

[0137] For example, when the operator system 5 receives the description data (S410), it expands the characteristic data of each consumer as the analysis object based on the database of the platform (S420).

[0138] The operator system 5 can expand the characteristic data of each consumer included in the data based on the characteristic data of the corresponding users included in its database. The operator system 5 can send the expanded characteristic data of each consumer to the information processing system 1 or output it to the data cleanroom (S430). The expanded characteristic data can be used, for example, for consumer behavior analysis.

[0139] In the example described in this embodiment, where the information processing system 1 sends description data of consumers targeted for advertising to multiple platform operator systems 5 for advertising placement, the following effect is achieved: That is, the information processing system 1 can request advertising placement based on the information of each consumer from the platform operator, which stores rich information about that consumer.

[0140] As described above, information processing system 1 can, for each consumer in a set of consumers targeted for advertising, request advertising from platform operators with high usage records for that consumer. Platforms with high usage records are likely to hold a wealth of information about the corresponding consumer. Platforms with a history of consumer logins and usage are likely to hold even more information about the corresponding consumer.

[0141] Therefore, by using the above method, the information processing system 1 calculates a score Scr for each consumer's performance on each of the multiple platforms, and sends the information of the corresponding consumer to the operator system 5 of the platform with the higher score Scr. In this way, the target audience for advertising can be appropriately expanded from the consumers selected in the dataset 15C, thereby achieving effective advertising.

[0142] [Other Implementation Methods]

[0143] This disclosure is not limited to the above-described embodiments, and can take various forms.

[0144] For example, the selection of sending processing can be as follows: Figure 8 The transformation is shown. That is, the processor 11 can weight and correct the score Scr of the object calculated in S230 on each platform based on the demographic attributes of the object.

[0145] Variation example selection, transmission processing and Figure 4 and Figure 5 The selection and sending process shown is the same, the difference being that... Figure 8 The processes shown in S241 and S245 are used to replace Figure 4 and Figure 5 The process shown is S240 in the selection and transmission process. Therefore, Figure 8 Only some steps of the selective transmission process, including S241 and S245, are shown selectively. Regarding the selective transmission process in the variant examples, descriptions of steps other than those in S241 and S245 are appropriately omitted.

[0146] It can be assumed that advertisers specify the expected distribution of demographic attributes within their target audience. For example, it can be assumed that for gender and age groups defined by a combination of gender and age groups, advertisers specify the expected composition ratio of their target audience according to this division. For example, it can be assumed that for groups interested in SUVs, advertisers want to target ads based on a weighted composition ratio of men in age groups with higher purchase intent (e.g., 20-29, 30-39, 40-49 years old).

[0147] Therefore, the score Scr is weighted as W·Scr by multiplying it by a weight W corresponding to the proportion of the desired demographic attribute. For example, it can be considered that when the desired male-to-female ratio is male:female = 3:1, the score Scr for males is weighted by W = 3 / 4, and the score Scr for females is weighted by W = 1 / 4.

[0148] Alternatively, weights W=Wθ can be assigned to each partition θ of the demographic attribute based on a histogram provided by the advertiser, such that the higher the frequency of the attribute partition θ identified from the histogram, the higher the weight W=Wθ of that attribute partition. Here, the expected distribution of the histogram across the demographic attribute represents the frequency of the attribute partition θ. The operator of information processing system 1 can pre-set the weights Wθ of the attribute partition θ based on the requirements from the advertiser.

[0149] according to Figure 8 In the variant shown, in S241 after S230, the processor 11 weights the score Scr of the object on each platform with the weight W = Wθ corresponding to the attribute partition θ of the object, and calculates the weight value Sc = W·Scr of the score Scr.

[0150] In subsequent S245, processor 11 is able to assign an object to one or more platforms whose weight value Sc on each platform is above a baseline, based on the object's weight value Sc on each platform. Afterward, processor 11 executes the processing in S250 (see...). Figure 4 ).

[0151] Based on the weighting method described above, it is possible to make the set of consumers allocated to each platform as the target of advertising closer to the expected distribution, and to achieve advertising delivery according to the expected distribution corresponding to the advertiser.

[0152] Additionally, in S270, when generating description data, the processor 11 can extend the distribution of more than one consumer assigned to the corresponding platform to two or more similar or consistent consumers.

[0153] For expansion, the memory 15 can be configured as follows: Figure 9 The consumer database 16 shown may contain characteristic data for each consumer in a third consumer set, which is associated with the corresponding consumer's identification data. The third consumer set may be a larger consumer set than the first consumer set. The third consumer set may at least partially include consumers belonging to the first consumer set.

[0154] Similar to dataset 15C, the identification data of each consumer in consumer database 16 can also have the aforementioned multiple IDs of the corresponding consumer as the consumer's identification code.

[0155] Third-person feature data is data that represents the characteristics of the corresponding consumer. Third-person feature data may be data that partially represents the same type of characteristics as dataset 15C. For example, third-person feature data may include information about the demographic attributes of the corresponding consumer. However, third-person feature data may not include PF performance data.

[0156] In S270, processor 11 is able to compare dataset 15C and consumer database 16. Based on this comparison, processor 11 can extract at least one consumer from the third consumer set that has at least similar characteristics to at least one consumer already assigned to the corresponding platform. Processor 11 can also expand the number of consumers assigned to a corresponding platform from one or more extracted consumers to two or more consumers. "At least similar characteristics" includes both feature similarity and feature consistency.

[0157] In S270, processor 11 is capable of generating descriptive data about two or more consumers that have been expanded, describing those two or more consumers. Processor 11 is capable of storing characteristic data of each consumer in the descriptive data, which includes information that can be extracted from dataset 15C and consumer database 16.

[0158] By expanding the number of consumers assigned in the S240 processing to two or more consumers when generating the description data to be sent to the corresponding platform, it is possible to expand the target audience for advertising, such as in an advertising service.

[0159] The functions of one constituent element in the above embodiments can be distributed among multiple constituent elements. Alternatively, the functions of multiple constituent elements can be integrated into one constituent element. A portion of the configuration of the above embodiments can be omitted. At least a portion of the configuration of the above embodiments can be added to the configurations of the other embodiments described above, or at least a portion of the configuration of the above embodiments can be replaced by the configurations of the other embodiments described above. All embodiments of the technical concept defined by the statements in the claims are implementations of this disclosure.

Claims

1. An information processing system, characterized in that, have: The acquisition unit is configured to acquire datasets for multiple consumers, each dataset containing a corresponding consumer's identification code and characteristic data. The characteristic data for each consumer includes information about multiple digital platforms and the extent to which the consumer uses each digital platform. An allocation unit, configured to, based on the degree of usage of each digital platform determined according to the characteristic data of each of the plurality of consumers, allocate one or more consumers to a corresponding digital platform for each digital platform. The list generation unit is configured to generate a consumer list for each digital platform in order to send data to the operator system of the corresponding digital platform, wherein the consumer list records the identification codes of one or more consumers assigned to the corresponding digital platform.

2. An information processing system, characterized in that, have: The acquisition unit is configured to acquire a dataset about multiple consumers. For each consumer, the dataset includes an identification code and characteristic data. The characteristic data for each consumer includes information about multiple digital platforms and the extent to which the consumer uses each digital platform. An allocation unit, configured to, based on the degree of usage of each digital platform determined according to the characteristic data of each of the plurality of consumers, allocate one or more consumers to a corresponding digital platform for each digital platform. The sending unit is configured to send, for each digital platform, the identification code of the one or more consumers assigned to the corresponding digital platform to the operator system of the corresponding digital platform.

3. The information processing system according to claim 1 or claim 2, characterized in that, The allocation unit is configured to allocate one or more consumers to each of the multiple digital platforms in a manner that prioritizes allocating each of the multiple consumers to the platform where the corresponding consumer has a high usage level.

4. The information processing system according to any one of claims 1 to 3, characterized in that, The allocation unit is configured to, for each consumer, calculate a score for each of the plurality of digital platforms based on the degree of usage of the corresponding digital platform by the corresponding consumer, and, based on the score, allocate the more than one consumer to the corresponding digital platform for each digital platform.

5. The information processing system according to claim 4, characterized in that, The allocation department calculates the score in a weighted manner based on the demographic attributes of the corresponding consumer.

6. The information processing system according to claim 5, characterized in that, The allocation unit uses weights to perform weighted calculations on the scores, wherein the weights correspond to a specified consumer distribution with respect to the demographic attribute.

7. The information processing system according to claim 1 or claim 2, characterized in that, The allocation unit, for each consumer, calculates a score for each of the plurality of digital platforms based on the consumer's usage of the corresponding digital platform. The allocation unit allocates the corresponding consumer to a digital platform among the plurality of digital platforms in a manner that the score reaches or exceeds a benchmark. Furthermore, when there are two or more digital platforms with the score reaching or exceeding the benchmark for the corresponding consumer, the unit allocates the corresponding consumer to both of those platforms. This process is repeated for each digital platform to assign one or more consumers to a given digital platform. When calculating the score, the allocation unit uses weights to weight the score based on the demographic attributes of the corresponding consumers, wherein the weights correspond to a specified consumer distribution with respect to the demographic attributes.

8. The information processing system according to any one of claims 1 to 7, characterized in that, The information processing system includes a dataset generation unit. This unit estimates the usage levels of at least a portion of the multiple digital platforms by the multiple first consumers based on a first dataset about multiple first consumers and a second dataset about multiple second consumers. Information regarding the estimated usage levels is added to the first dataset to generate the dataset about the multiple consumers. The first dataset is a dataset that does not include information regarding the usage levels of the at least a portion of the multiple digital platforms, while the second dataset is a dataset that includes information regarding the usage levels of the at least a portion of the multiple digital platforms. The acquisition unit acquires the dataset about the multiple consumers generated by the dataset generation unit.

9. The information processing system according to claim 1, characterized in that, The list generation unit refers to a database about consumer groups, and for each digital platform, expands the one or more consumers assigned to the corresponding digital platform into two or more consumers with similar or identical characteristics, and generates a consumer list describing the identification codes of the two or more consumers as the consumer list, wherein the database has a corresponding consumer identification code and characteristic data for each consumer.

10. The information processing system according to claim 2, characterized in that, The sending unit is configured to send submission data to the operator system of the corresponding digital platform for each digital platform, wherein the submission data records the identification code of the one or more consumers assigned to the corresponding digital platform as the identification code of the consumer to whom the advertisement is targeted.

11. The information processing system according to claim 10, characterized in that, Each of the plurality of digital platforms has a database that associates and records the characteristics of multiple users with their respective user identification codes. Each of the plurality of digital platforms is configured, based on the database, to expand the target audience for advertising from the single or multiple consumers whose identification codes are recorded in the submitted data to a group of users whose characteristics are related to the consumers targeted by the advertising, thereby delivering advertisements to this user group. When sending the submission data, the sending unit specifies the upper limit of the number of users who are the target audience for the extended advertisement.

12. The information processing system according to claim 11, characterized in that, The sending unit specifies an upper limit on the number of users who will be targeted for advertising in each division of the demographic attributes based on the distribution of the demographic attributes of the plurality of consumers, wherein the distribution of the demographic attributes is determined based on the dataset of the plurality of consumers obtained by the acquisition unit.

13. The information processing system according to claim 2, characterized in that, The sending unit sends the one or more consumer identification codes assigned to the corresponding digital platform, along with the one or more consumer characteristic data, to the operator system of the corresponding digital platform for each digital platform.

14. An information processing method, executed by a computer, characterized in that it includes: Acquire a dataset for multiple consumers, each dataset containing a unique identifier and characteristic data. The characteristic data for each consumer includes information about multiple digital platforms, the extent to which the consumer uses each digital platform, and... Based on the degree of usage of each digital platform determined according to the characteristic data of each of the plurality of consumers, for each digital platform, one or more of the plurality of consumers are assigned to the corresponding digital platform. Furthermore, the information processing method further includes any one of the following steps: For each digital platform, a consumer list is generated for sending data to the operator system of the corresponding digital platform, wherein the consumer list records the identification codes of one or more consumers assigned to the corresponding digital platform; and For each digital platform, the identification code assigned to one or more consumers on the corresponding digital platform is sent to the operator system of the corresponding digital platform.

15. The information processing method according to claim 14, characterized in that, The allocation includes: For each consumer, a score for each of the plurality of digital platforms is calculated based on the consumer's usage of the corresponding digital platform. The method involves assigning the corresponding consumer to a digital platform among the plurality of digital platforms whose score reaches or exceeds a benchmark. Furthermore, when there are two or more digital platforms with scores exceeding the benchmark for the corresponding consumer, the method involves assigning the corresponding consumer to one or more of these platforms. When calculating the score, the score is weighted based on the demographic attributes of the corresponding consumer, and the weights correspond to a specified consumer distribution with respect to the demographic attributes.

16. A computer program, characterized in that, The computer program includes instructions, when executed by a computer, for causing the computer to perform the information processing method of claim 14 or claim 15.

17. A computer-readable recording medium, characterized in that, The computer program stores instructions for causing the computer to perform the information processing method of claim 14 or claim 15 when the computer executes the computer program.

Images