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

Abstract

This provides a new technology for inferring the relationship between the user of a content output device and the owner of an information terminal. [Solution] In the information processing device, the acquisition unit is configured to acquire combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information. The attribute determination unit is configured to determine attribute information relating to the attributes of the network identification information. The inference unit is configured to infer the relationship between the user of the content output device and the owner of the account for at least one of the multiple combinations to be inferred. The inference unit is configured to infer the relationship based on the attribute information of the network identification information relating to at least one combination to be inferred.

Description

Technical Field

[0004] , ,

[0005] , , , , , ,

[0001] This disclosure relates to an information processing system.

Background Art

[0002] Techniques for determining whether a user of a content output device such as a television and an owner of an information terminal such as a smartphone or a personal computer are the same person are known. For example, Patent Document 1 discloses an information processing system that identifies an owner of an information terminal who is the same person as the user of a television based on a combination of a viewing log of the television and an advertisement ID of the information terminal.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] There is a further need for improvement in the method for determining whether a user of a content output device and an owner of an information terminal are the same person. One aspect of this disclosure is to provide a new technique for inferring the relationship between a user of a content output device and an owner of an information terminal.

Means for Solving the Problems

[0006] With this configuration, based on the results of the above-mentioned correlation estimation, it is possible to infer the relationship between the user of the content output device and the account owner. Therefore, by identifying the account owner, it is possible to indirectly infer the relationship between the user of the content output device and the owner of the information terminal.

[0007] In one aspect of this disclosure, the combination information may include account identification information that can identify an account for each of a plurality of combinations. The inference unit may be configured to infer associations based on the account identification information relating to at least one combination to be inferred. This configuration can improve the accuracy of inferring correlations.

[0008] In one aspect of this disclosure, the inference unit may be configured to infer associations based on the output of a machine learning model obtained by inputting features relating to attribute information into a pre-trained machine learning model. With this configuration, the accuracy of predicting relationships can be improved by using a machine learning model. It can improve.

[0009] One aspect of this disclosure may further include a correspondence determination unit. The correspondence determination unit is configured to determine correspondence information relating to the relationship between network identification information and at least one of a content output device and an account, based on combination information. The prediction unit may be configured to further input the correspondence information as features into a machine learning model. This configuration can improve the accuracy of inferring correlations.

[0010] In one aspect of this disclosure, the combination information may describe, for each of a plurality of network identifiers, a combination of content output devices and accounts whose communication has been confirmed via the corresponding network identifier, in association with the corresponding network identifier. The correspondence information may include at least one of a first correspondence number, a second correspondence number, a third correspondence number, and a fourth correspondence number in the combination information. The first correspondence number may be the number of network identifiers associated with the corresponding content output device for each content output device. The second correspondence number may be the number of content output devices associated with the corresponding network identifier for each network identifier. The third correspondence number may be the number of accounts associated with the corresponding network identifier for each network identifier. The fourth correspondence number may be the number of network identifiers associated with the corresponding account for each account.

[0011] With this configuration, it is possible to infer relationships from multiple combinations of information. Therefore, the accuracy of inferring relationships can be improved.

[0012] In one aspect of this disclosure, the network identification information may be an IP address. With this configuration, relationships can be inferred using IP addresses.

[0013] In one aspect of this disclosure, attribute information may include at least one of IP address class information, ASN information, and carrier information. This configuration allows us to infer relationships using features with higher contributions. Therefore, we can improve the accuracy of relationship inference.

[0014] One aspect of the present disclosure is an information processing device comprising an acquisition unit and an inference unit. The acquisition unit is configured to acquire combination information describing a plurality of combinations of content output devices and accounts whose communication has been confirmed via the same network identification information. The inference unit is configured to infer the relationship between a user of a content output device and an account owner for at least one of the plurality of combinations to be inferred. The combination information includes account identification information that can identify an account for each of the plurality of combinations. The inference unit is configured to infer the relationship based on the account identification information for at least one of the combinations to be inferred.

[0015] With this configuration, based on the results of the above-mentioned correlation estimation, it is possible to infer the relationship between the user of the content output device and the account owner. Therefore, by identifying the account owner, it is possible to indirectly infer the relationship between the user of the content output device and the owner of the information terminal.

[0016] In one aspect of this disclosure, the inference unit may be configured to infer relevance based on the output of a machine learning model obtained by inputting features relating to account identification information into a pre-trained machine learning model. With this configuration, the accuracy of predicting relationships can be improved by using machine learning models.

[0017] One aspect of this disclosure is an information processing system comprising the information processing device described above and a display device. The display device is configured to display the results predicted by the prediction unit. With this configuration, it is possible to realize a system in which the information processing device displays the results it has inferred on a display device.

[0018] One aspect of the present disclosure is a computer-based information processing method comprising: obtaining combination information describing a plurality of combinations of content output devices and accounts whose communication has been confirmed via the same network identification information; determining attribute information relating to the attributes of the network identification information; and inferring a relationship between a user of a content output device and an account owner with respect to at least one of the plurality of combinations. The inference includes inferring the relationship based on the attribute information of the network identification information relating to at least one combination of the plurality of combinations. This information processing method produces the same effects as the information processing device described above.

[0019] One aspect of the present disclosure is a computer program for causing a computer to perform a process, the process including: obtaining combination information describing a plurality of combinations of content output devices and accounts whose communication has been confirmed via the same network identifier; determining attribute information relating to the attributes of the network identifier; and inferring a relationship between a user of a content output device and an account owner with respect to at least one of the plurality of combinations. The inference includes inferring the relationship based on the attribute information of the network identifier relating to at least one combination of the plurality of combinations. This computer program produces the same effect as the information processing device described above.

[0020] One aspect of this disclosure is a computer-based information processing method comprising: obtaining combination information describing a plurality of combinations of content output devices and accounts whose communication has been confirmed via the same network identification information; and inferring a relationship between a user of a content output device and an account owner with respect to at least one of the plurality of combinations. The combination information includes, for each of the plurality of combinations, account identification information that can identify an account. The inference includes inferring a relationship based on the account identification information with respect to at least one of the inferred combinations. According to this information processing method, the same effects as those of the above-described information processing apparatus are achieved.

[0021] One aspect of the present disclosure is a computer program for causing a computer to execute processing, the processing including: obtaining combination information that describes a plurality of combinations of a content output device and an account for which communication has been confirmed via the same network identification information; and inferring a relationship between a user of the content output device and an owner of the account for at least one combination to be inferred among the plurality of combinations. The combination information includes account identification information that can identify an account for each of the plurality of combinations. Inferring includes inferring the relationship based on the account identification information for at least one combination to be inferred. According to this computer program, the same effects as those of the above-described information processing apparatus are achieved.

Brief Description of the Drawings

[0022] [Figure 1] It is a block diagram showing the configuration of an information processing system. [Figure 2] It is a diagram for explaining that a viewing log server accumulates first specific information received from a content output device. [Figure 3] It is a diagram for explaining that an access log server accumulates second specific information received from an information terminal. [Figure 4] It is a flowchart showing the processing of an information processing apparatus. [Figure 5] It is a diagram for explaining a feature amount input to a machine learning model and a relevance probability output by the machine learning model.

Mode for Carrying Out the Invention

[0023] Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the drawings. [1. First Embodiment] ​​The information processing system 100 shown in Figure 1 comprises an information processing device 1, a content output device 21, an information terminal 22, a router 3, a viewing log server 41, an access log server 42, and a machine learning model 8.

[0024] The information processing device 1 comprises a processor 11, memory 12, storage 13, a user interface 14, and a communication interface 15. The information processing device 1 may be a personal computer or the like.

[0025] The processor 11 is configured to perform various functions of the information processing device 1 by executing one or more instructions contained in the computer program stored in the memory 12.

[0026] Memory 12 is configured to store computer programs. Memory 12 is used as a workspace when the processor 11 executes processing. An example of memory 12 is RAM (Random Access Memory).

[0027] Storage 13 holds computer programs and various data used when executing processes that follow the computer programs. Examples of storage 13 include HDDs (Hard Disk Drives), SSDs (Solid State Drives), and flash memory.

[0028] The user interface 14 is a general term for interfaces that accept various input operations from the user and interfaces that output various information to the user. Examples of user interfaces 14 include keyboards, mice, touch panels, displays, etc.

[0029] The communication interface 15 is an interface that can communicate various types of data with external devices in accordance with a predetermined standard. The communication may be wired or wireless. The information processing device 1 is configured to communicate with the viewing log server 41, the access log server 42, and the machine learning model 8 through the communication interface 15.

[0030] The content output device 21 is a device capable of outputting content. The content output device 21 is, for example, a television. The television (i.e., television receiver) referred to here is a device that has the function of receiving at least one of the following broadcasts: terrestrial broadcasting, broadcasting satellite (BS), communications satellite (CS), etc. The television may also have the function of receiving internet broadcasting (i.e., internet television).

[0031] The content output device 21 may be a mobile device such as a mobile phone, smartphone, or tablet, or it may be a personal computer. The content output device 21 may also be an information device capable of outputting content. In this embodiment, the television (TV) includes multi-function televisions, smart TVs, IP (Internet Protocol) TVs, etc.

[0032] The content output device 21 can store the viewing log 411 related to the content output device 21. The system is configured as follows. The viewing log 411 is information that records, for example, the content viewed on the content output device 21 in chronological order. The viewing log 411 may include information about the programs viewed, the commercials viewed, etc. The viewing log 411 may also include information that can identify at least one of the viewing actions via the content output device 21, such as viewing terrestrial broadcasts, time-shift playback, browser startup, or execution of internet services. It is preferable that the viewing log 411 is accompanied by a timestamp indicating the viewing time, corresponding to the program viewed, viewing action, etc.

[0033] As shown in Figure 2, the content output device 21 is configured to send viewing logs 411 to the viewing log server 41 at a timely interval. In addition to the viewing logs 411, the content output device 21 also sends first identification information 412 to the viewing log server 41. The first identification information 412 is information that can identify the content output device 21. In this embodiment, the device ID of the content output device 21 is used as an example of the first identification information 412.

[0034] The device ID is a unique identifier. The device ID may be a serial number, manufacturing number, product model number, or a combination thereof. The device ID may also be a value calculated based on these values. For example, the device ID may be a hashed value of the serial number. Each content output device 21 is associated with a unique device ID.

[0035] The information terminal 22 is a device for accessing a platform that provides various services. The information terminal 22 may be, for example, a mobile phone, smartphone, tablet device, or personal computer. The information terminal 22 may also be a device capable of outputting content such as television, but it is a different device from the content output device 21. That is, the device ID of the information terminal 22 is different from the device ID of the content output device 21.

[0036] Users of the platform accessed by the information terminal 22 are issued a dedicated account 22a. Each account 22a is associated with a unique account ID. This account ID may be a so-called platform ID, similar to a login ID for using the platform. Users can access the platform using their account 22a. In other words, a user's account 22a can access the platform through the information terminal 22.

[0037] The information terminal 22 is configured to store access logs 421 when accessing a corresponding platform using various accounts 22a. The access log 421 is information that records, for example, the platform accessed and the date and time of access to that platform for each account 22a. The access log 421 may also include the details of the services used by each account 22a on the platform.

[0038] As shown in Figure 3, the information terminal 22 is configured to send the access log 421 to the access log server 42 at a timely interval. In addition to the access log 421, the information terminal 22 also sends the second identification information 422 to the access log server 42. The second identification information 422 is information that can identify account 22a. In this embodiment, the account ID is exemplified as the second identification information 422.

[0039] Router 3 relays communication between devices within the first network NW1 and devices within the second network NW2, which is outside the first network NW1. Router 3 has network identification information 31 that can identify the network. In this embodiment, Router 3 is assigned an IP address by the ISP (Internet Service Provider) as the network identification information 31. However, the network identification information 31 is not limited to an IP address.

[0040] In this embodiment, the first network NW1 is, for example, a LAN (Local Area Network). The second network NW2 is, for example, the Internet. The IP address is a so-called global IP address.

[0041] The first network NW1 includes a content output device 21, an information terminal 22, and a router 3. The second network NW2 includes an information processing device 1, a viewing log server 41, an access log server 42, and a router 3.

[0042] The content output device 21 and the information terminal 22 are connected to the router 3 by wire or wireless connection. The content output device 21 and the information terminal 22, which are included in the same first network NW1, are configured to access the second network NW2, which is outside the first network NW1, via the router 3.

[0043] In the configuration shown in Figure 1, the IP address visible to external communication partners on the first network NW1 during communication between the content output device 21 and the information terminal 22 is the IP address of router 3 (i.e., the same IP address). This is achieved through the IP masquerading (NAPT: Network Address Port Translation) function of router 3.

[0044] An IP address may be an address conforming to IPv4 or IPv6, or it may be an extended or modified version of these. In this embodiment, as an example, the IP address is an address conforming to IPv4.

[0045] The viewing log server 41 is a device that manages viewing logs 411 related to the content output device 21. The viewing log server 41 may be owned by the manufacturer of the content output device 21, or it may be owned by a broadcasting station (e.g., a television station) that broadcasts via the content output device 21.

[0046] As shown in Figure 2, the viewing log server 41 is configured to store the first identification information 410 received from the content output device 21. The first identification information 410 includes viewing logs 411 related to the content output device 21, first identification information 412 of the content output device 21, and network identification information 31 of the content output device 21 (i.e., the IP address of router 3). The first identification information 410 is different for each content output device 21. The viewing log server 41 is configured to transmit the first identification information 410 to the information processing device 1 in response to a request from the information processing device 1.

[0047] As shown in Figure 3, the access log server 42 is a device that manages account logs related to accounts 22a owned by users of the information terminal 22. The access log server 42 is configured to store second identification information 420 received from the information terminal 22. The second identification information 420 includes the access log 421 related to account 22a, the second identification information 422 of account 22a, and the network identification information 31 of the information terminal 22 (i.e., the IP address of router 3). The second identification information 420 is different for each information terminal 22. The access log server 42 is configured to transmit the second identification information 420 to the information processing device 1 in response to a request from the information processing device 1.

[0048] Machine learning model 8 is a pre-trained machine learning model. Machine learning model 8 may be, for example, a deep learning model. The training method for machine learning model 8 will be described later. In the example shown in Figure 1, machine learning model 8 is configured to communicate directly with the information processing device 1, but it may also be configured to communicate with the information processing device 1 via other networks such as the internet.

[0049] [1-2. Processing] The series of processes performed by processor 11 will be explained using the flowchart in Figure 4 and Figure 5. When processor 11 receives instructions from the user through the user interface 14, it starts the process shown in Figure 4.

[0050] First, in S100, the processor 11 obtains the first specific information 410 from the viewing log server 41. Next, in S110, the processor 11 obtains the second specific information 420 from the access log server 42.

[0051] Next, in S120, the processor 11 obtains combination information 50 as shown in Figure 5, based on the first identification information 410 and the second identification information 420. The combination information 50 describes multiple combinations of content output devices 21 and accounts 22a whose communication has been confirmed via the same network identification information 31. That is, each of the multiple combinations is a combination of one content output device 21 and one account 22a.

[0052] In this embodiment, the combination information 50 includes a first identification information 412 and a second identification information 422 for each of the multiple combinations. That is, the combination of the content output device 21 and the account 22a is described by the first identification information 412 and the second identification information 422.

[0053] The combination information 50 may further include network identification information 31. In other words, for each of the multiple network identification information 31s, the combination information 50 may describe the combination of content output device 21 and account 22a whose communication has been confirmed via the corresponding network identification information 31, in relation to the corresponding network identification information 31. In the example shown in Figure 5, multiple combinations of a device ID as the first identification information 412 and an account ID as the second identification information 422 are associated with the same IP address as the network identification information 31.

[0054] Next, in S130, the processor 11 determines correspondence information 60 regarding the relationship between the network identification information 31 and at least one of the content output device 21 and account 22a, based on the combination information 50.

[0055] The correspondence information 60 includes at least one of the first correspondence number 61, the second correspondence number 62, the third correspondence number 63, and the fourth correspondence number 64 in the combination information 50. The first correspondence number 61 is the number of network identification information 31 associated with each corresponding content output device 21. In the example shown in Figure 5, the number of network identification information 31 associated with the content output device 21 whose device ID is "94EU-RM5V-KQUU" is shown to be "11".

[0056] The second correspondence number, 62, is the number of content output devices 21 associated with each corresponding network identification information 31. In the example shown in Figure 5, the number of content output devices 21 associated with the IP address "11.22.33.44" is shown to be "12".

[0057] The third correspondence number, 63, is the number of accounts 22a associated with each corresponding network identification information 31. In the example shown in Figure 5, the number of accounts 22a associated with the IP address "11.22.33.44" is shown to be "34".

[0058] The fourth correspondence number, 64, is the number of network identification information 31 associated with each corresponding account 22a. In the example shown in Figure 5, the number of network identification information 31 associated with account 22a, whose account ID is "FF6DAD21", is shown to be "1".

[0059] Next, in S140, the processor 11 determines the attribute information 70 relating to the attributes of the network identification information 31 related to the combination information 50. Attribute information 70 represents the characteristics and properties of one or more network identification information 31. In this embodiment, network identification information 31 is an IP address. Attribute information 70 includes at least one of the IP address class information 71, ASN information 72, and carrier information 73.

[0060] Class information 71 represents a classification of IP addresses based on their range. For example, if IPv4 IP addresses are used as network identification information 31, IP addresses within the range of "0.0.0.0" to "127.255.255.255" may be classified as Class A, IP addresses within the range of "128.0.0.0" to "191.255.255.255" may be classified as Class B, IP addresses within the range of "192.0.0.0" to "223.255.255.255" may be classified as Class C, IP addresses within the range of "224.0.0.0" to "239.255.255.255" may be classified as Class D, and IP addresses within the range of "240.0.0.0" to "255.255.255.255" may be classified as Class E. Each class is not limited to these ranges and can be classified by any range.

[0061] ASN information 72 represents the ASN (Anonymous System Number) to which an IP address belongs. An ASN is a number used to identify a network on the internet. ASN information 72 can be determined based on the IP address.

[0062] Carrier information 73 is information that represents the telecommunications carrier corresponding to the IP address. Carrier information 73 can be identified based on the IP address. Figure 5 shows that the attribute information 70 for the IP address "11.22.33.44" includes class information 71, which is "Class A", ASN information 72, which is "ASxxxx", and carrier information 73, which is "Company P".

[0063] Next, in S150, the processor 11 infers the relationship between the user of the content output device 21 and the owner of account 22a with respect to at least one of the multiple combinations to be inferred. The processor 11 infers the relationship based on at least one of the attribute information 70 of the network identification information 31, the first identification information 412, and the second identification information 422 with respect to at least one combination to be inferred.

[0064] The method of expressing the results of correlation estimation is not particularly limited. For example, correlation estimation results may be shown using probabilities representing the degree of correlation, classes that evaluate the degree of correlation in stages, or binary data indicating the presence or absence of correlation, such as "correlated" and "not related." Correlation estimation results may also be shown using predetermined thresholds. For example, correlation estimation results may be shown as "correlated" when a numerical value representing the degree of correlation exceeds a predetermined threshold.

[0065] In this embodiment, as shown in Figure 5, the association is estimated based on the output of the machine learning model 8 obtained by inputting a feature quantity 81 including combination information 50, correspondence information 60, and attribute information 70 into the machine learning model 8. As a result of estimating the association, the machine learning model 8 outputs an association probability 82 for each combination, which is a probability representing the degree of association. The processor 11 may determine that the user of the content output device 21 and the owner of account 22a are the same person if the association probability 82 is above a predetermined threshold.

[0066] The training process for machine learning model 8 may be performed using supervised learning of machine learning model 8, where the output of machine learning model 8 in S150 is assigned a correct label indicating whether each combination is "related" or "unrelated" as training data.

[0067] Next, in S160, the processor 11 displays the results of its correlation estimation to the user. In this embodiment, the processor 11 displays the results through the user interface 14. The processor 11 may also display the results through a display device other than the user interface 14.

[0068] After that, processor 11 terminates the process shown in Figure 4. In this way, the processor 11 obtains first identification information 410, which is associated with first identification information 412 and network identification information 31, and second identification information 420, which is associated with second identification information 422 and network identification information 31. Based on the first identification information 410 and the second identification information 420, the processor 11 obtains combination information 50. Based on the combination information 50, the processor 11 determines correspondence information 60 regarding the relationship between network identification information 31 and at least one of content output device 21 and account 22a. The processor 11 determines attribute information 70 regarding the attributes of network identification information 31 related to the combination information 50. Based on the output of the machine learning model 8, which is obtained by inputting a feature quantity 81 including the combination information 50, correspondence information 60, and attribute information 70 into the machine learning model 8, the processor 11 infers the relationship.

[0069] [1-3. Effects] According to the embodiments described above, the following actions and effects can be obtained. (1a) The processor 11 can infer a relationship between the user of the content output device 21 and the owner of account 22a with respect to at least one of the multiple combinations to be inferred.

[0070] With this configuration, based on the results of inferring the above-mentioned relationship, the owner of account 22a, who is the same person as the user of content output device 21, can be identified. Since account 22a accesses the platform through information terminal 22, by identifying the owner of account 22a, the owner of information terminal 22 can be indirectly inferred. Therefore, the relationship between the user of content output device 21 and the owner of information terminal 22 can be inferred.

[0071] In addition, by associating the user of the content output device 21 with the owner of account 22a, even if the owner of account 22a accesses the platform using an information terminal different from the information terminal 22, if account 22a is the same, it can be inferred that the owner of account 22a is also the same. Therefore, the access log 421 of the owner of account 22a can be obtained independently of the information terminal 22. In other words, the relationship between the user of the content output device 21 and the owner of the information terminal 22 can be inferred without using information terminal 22-specific identification information such as advertising IDs or cookie IDs.

[0072] (1b) The processor 11 may infer the above relationship based on the attribute information 70. The attribute information 70 relates to the attributes of the network identification information 31 in the combination to be inferred. With this configuration, the above-mentioned relationships can be inferred by considering the attributes of the network identification information 31. Therefore, the accuracy of inferring relationships can be improved.

[0073] In particular, when using the machine learning model 8, the attributes of the network identification information 31 are more abstract than the network identification information 31 itself, allowing for the input of features with higher contributions to the machine learning model 8. This improves the accuracy of predicting relationships.

[0074] (1c) In this embodiment, the network identification information 31 is an IP address. The attribute information 70 relates to the attributes of the IP address in the combination to be inferred. The attribute information 70 is, for example, the IP address class information 71, ASN information 72, and carrier information 73.

[0075] With this configuration, attribute information 70 can be determined based on the IP address. In other words, attribute information 70 can be determined without using information other than the IP address itself, such as the usage time or usage status of the IP address. Therefore, attribute information 70 can be easily determined.

[0076] (1d) The processor 11 may infer the above relationship based on the first identification information 412 and the network identification information 31. The first identification information 412 is information that can identify the content output device 21. For example, the processor 11 infers the above relationship based on the correspondence information 60 which includes at least one of the first correspondence number 61 and the second correspondence number 62 in the combination information 50. The first correspondence number 61 is the number of network identification information 31 associated with the corresponding content output device 21 for each content output device 21. The second correspondence number 62 is the number of content output devices 21 associated with the corresponding network identification information 31 for each network identification information 31.

[0077] With this configuration, an account 22a whose communication with the content output device 21 corresponding to the first identification information 412 has been confirmed via the same network identification information 31 can be inferred from multiple combinations in the combination information 50. Therefore, the accuracy of inferring relationships can be improved.

[0078] (1e) The processor 11 may infer the above relationship based on the second identification information 422 and the network identification information 31. The second identification information 422 is information that can identify account 22a. For example, the processor 11 infers the above relationship based on the correspondence information 60 which includes at least one of the third correspondence number 63 and the fourth correspondence number 64 in the combination information 50. The third correspondence number 63 is the number of accounts 22a associated with the corresponding network identification information 31 for each network identification information 31. The fourth correspondence number 64 is the number of network identification information 31 associated with the corresponding account 22a for each account 22a.

[0079] With this configuration, a content output device 21 that has communicated with account 22a corresponding to the second identification information 422 via the same network identification information 31 can be inferred from multiple combinations in the combination information 50. Therefore, the accuracy of inferring relationships can be improved.

[0080] [1-4. Correspondence between terms] In the above embodiment, the IP address corresponds to an example of network identification information, the second identification information 422 corresponds to an example of account identification information, and the user interface 14 corresponds to an example of a display device.

[0081] The process in S120 corresponds to an example of a process executed as an acquisition unit, the process in S130 corresponds to an example of a process executed as a correspondence determination unit, the process in S140 corresponds to an example of a process executed as an attribute determination unit, and the process in S150 corresponds to an example of a process executed as an inference unit. ru.

[0082] [2. Other Embodiments] While embodiments of this disclosure have been described above, it goes without saying that this disclosure is not limited to the embodiments described above and can take various forms.

[0083] (2a) In the above embodiment, the processor 11 executes the processes from S100 to S160 in order. However, the order in which the processor 11 executes the processes is not limited thereto. For example, the order in which the processes S100 and S110 are executed is not particularly limited and may be executed in any order. The order in which the processes S130 and S140 are executed is not particularly limited and may be executed in any order. The processor 11 does not have to execute the process S130 or S140.

[0084] (2b) In the above embodiment, the network identification information 31 is an IP address. The attribute information 70 includes at least one of the IP address class information 71, ASN information 72, and carrier information 73. However, the attribute information 70 is not limited to these. For example, the attribute information 70 may be location information, hostname, domain name, etc., identified from the network identification information 31. The attribute information 70 may also be time-series information relating to the network identification information 31. The time-series information may be information relating to the time of day when communication via the network identification information 31 is frequently used, such as during the day or at night.

[0085] (2c) In the above embodiment, the feature quantity 81 includes combination information 50, correspondence information 60, and attribute information 70. However, the feature quantity 81 is not limited to this. For example, the feature quantity 81 does not have to include correspondence information 60 or attribute information 70. The feature quantity 81 may further include features other than correspondence information 60 and attribute information 70.

[0086] For example, feature 81 may further include features related to viewing logs 411, access logs 421, etc. Alternatively, feature 81 may further include features related to identification information specific to the information terminal 22, such as advertising IDs and cookie IDs. The features related to the identification information specific to the information terminal 22 may be the number of information terminal 22-specific identification pieces corresponding to network identification information 31, or the number of network identification pieces 31 corresponding to the same information terminal 22-specific identification piece.

[0087] (2d) In the above embodiment, the processor 11 infers the relationship using the machine learning model 8. However, the processor 11 may infer the relationship without using the machine learning model 8. For example, the relationship may be inferred based on at least one of the correspondence information 60 and the attribute information 70. For example, it can be inferred that combinations with the same attribute information 70 have a higher relationship than combinations with different attribute information 70. Alternatively, the relationship may be inferred based on the magnitude of at least one of the first correspondence number 61, second correspondence number 62, third correspondence number 63, and fourth correspondence number 64 of the correspondence information 60. For example, it can be inferred that the larger the value of the fourth correspondence number 64, the higher the relationship.

[0088] (2e) Multiple functions of one component in the above embodiment may be realized by multiple components, or one function of one component may be realized by multiple components. Multiple functions of multiple components may be realized by one component, or one function realized by multiple components may be realized by one component. Some of the configurations of the above embodiment may be omitted. At least some of the configurations of the above embodiment may be added to or replaced with the configurations of other above embodiments.

[0089] (2f) This disclosure can be implemented in various forms other than the information processing device described above. For example, it can be realized in the form of a system comprising the information processing device, a computer program for causing a computer to function as the information processing device, a non-transitional physical recording medium such as semiconductor memory on which the computer program is recorded, or an information processing method.

[0090] [Technical Concept Disclosed in This Specified Specification] [Item 1] An information processing device, An acquisition unit configured to acquire combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, An attribute determination unit configured to determine attribute information relating to the attributes of the network identification information, An inference unit is configured to infer the relationship between the user of the content output device and the owner of the account with respect to at least one of the plurality of combinations to be inferred, Equipped with, The estimation unit is configured to estimate the relationship based on the attribute information of the network identification information relating to the at least one combination of items to be estimated. Information processing device.

[0091] [Item 2] The information processing device described in item 1, The combination information includes, for each of the plurality of combinations, account identification information that can identify the account, The estimation unit is configured to estimate the relationship based on the account identification information relating to the at least one combination of items to be estimated. Information processing device.

[0092] [Item 3] The information processing device described in item 1, The estimation unit is configured to estimate the relationship based on the output of a machine learning model obtained by inputting the feature quantities relating to the attribute information into a pre-trained machine learning model. Information processing device.

[0093] [Item 4] The information processing device described in item 3, The system further includes a correspondence determination unit configured to determine correspondence information relating to the relationship between the network identification information and at least one of the content output device and the account, based on the aforementioned combination information. The estimation unit is configured to further input the corresponding information as features into the machine learning model. Information processing device.

[0094] [Item 5] The information processing device described in item 4, The aforementioned combination information describes, for each of the multiple network identification pieces, the combination of the content output device and the account whose communication was confirmed via the corresponding network identification piece, in relation to the corresponding network identification piece. The aforementioned correspondence information includes the first correspondence number, the second correspondence number, the third correspondence number, and and include at least one of the fourth corresponding number, The first correspondence number is the number of network identification pieces associated with the corresponding content output device for each content output device. The second correspondence number is the number of content output devices associated with the corresponding network identification information for each network identification information. The third correspondence number is the number of accounts associated with the corresponding network identification information for each network identification information. The fourth correspondence number is the number of network identification pieces associated with the corresponding account for each account. Information processing device.

[0095] [Item 6] An information processing device described in any one of items 1 to 5, The aforementioned network identification information is an IP address. Information processing device.

[0096] [Item 7] The information processing device described in item 6, The attribute information includes at least one of the IP address's class information, ASN information, and carrier information. Information processing device.

[0097] [Item 8] An information processing device, An acquisition unit configured to acquire combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, An inference unit is configured to infer the relationship between the user of the content output device and the owner of the account with respect to at least one of the plurality of combinations to be inferred, Equipped with, The combination information includes, for each of the plurality of combinations, account identification information that can identify the account, The estimation unit is configured to estimate the relationship based on the account identification information relating to the at least one combination of items to be estimated. Information processing device.

[0098] [Item 9] The information processing device described in item 8, The estimation unit is configured to estimate the association based on the output of a machine learning model obtained by inputting the features related to the account identification information into a pre-trained machine learning model. Information processing device.

[0099] [Item 10] An information processing device described in any one of items 1 through 9, A display device configured to display the results predicted by the prediction unit, An information processing system equipped with the following features.

[0100] [Item 11] A method of information processing performed by a computer, To obtain combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, To determine attribute information relating to the attributes of the aforementioned network identification information, With respect to at least one of the aforementioned multiple combinations, the relationship between the user of the content output device and the owner of the account is inferred. Includes, The aforementioned inference includes inferring the association based on the attribute information of the network identification information relating to the at least one combination of inference targets, Information processing methods.

[0101] [Item 12] A computer program that causes a computer to perform a process, The aforementioned process is, To obtain combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, To determine attribute information relating to the attributes of the aforementioned network identification information, With respect to at least one of the aforementioned multiple combinations, the relationship between the user of the content output device and the owner of the account is inferred. Includes, The aforementioned inference includes inferring the association based on the attribute information of the network identification information relating to the at least one combination of inference targets, Computer program.

[0102] [Item 13] A method of information processing performed by a computer, To obtain combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, With respect to at least one of the aforementioned multiple combinations, the relationship between the user of the content output device and the owner of the account is inferred. Includes, The combination information includes, for each of the plurality of combinations, account identification information that can identify the account, The aforementioned inference includes inferring the association based on the account identification information relating to the at least one combination of inference targets, Information processing methods.

[0103] [Item 14] A computer program that causes a computer to perform a process, The aforementioned process is, To obtain combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, With respect to at least one of the aforementioned multiple combinations, the relationship between the user of the content output device and the owner of the account is inferred. Includes, The combination information includes, for each of the plurality of combinations, account identification information that can identify the account, The aforementioned inference includes inferring the association based on the account identification information relating to the at least one combination of inference targets, Computer program. [Explanation of Symbols]

[0104] 1... Information processing device, 11... Processor, 21... Content output device, 22... Information terminal, 22a... Account, 31... Network identification information, 70... Attribute information, 100... Information processing system, 422... Second identification information.

Claims

1. An information processing device, An acquisition unit configured to acquire combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, An attribute determination unit configured to determine attribute information relating to the attributes of the network identification information, An inference unit is configured to infer the relationship between the user of the content output device and the owner of the account with respect to at least one of the plurality of combinations to be inferred, Equipped with, The estimation unit is configured to estimate the relationship based on the attribute information of the network identification information relating to the at least one combination of items to be estimated. Information processing device.

2. An information processing apparatus according to claim 1, The combination information includes, for each of the plurality of combinations, account identification information that can identify the account, The estimation unit is configured to estimate the relationship based on the account identification information relating to the at least one combination of items to be estimated. Information processing device.

3. An information processing apparatus according to claim 1, The estimation unit is configured to estimate the relationship based on the output of a machine learning model obtained by inputting the feature quantities relating to the attribute information into a pre-trained machine learning model. Information processing device.

4. An information processing apparatus according to claim 3, The system further includes a correspondence determination unit configured to determine correspondence information relating to the relationship between the network identification information and at least one of the content output device and the account, based on the aforementioned combination information. The estimation unit is configured to further input the corresponding information as features into the machine learning model. Information processing device.

5. An information processing apparatus according to claim 4, The aforementioned combination information describes, for each of the multiple network identification pieces, the combination of the content output device and the account whose communication was confirmed via the corresponding network identification piece, in relation to the corresponding network identification piece. The correspondence information includes at least one of the first correspondence number, second correspondence number, third correspondence number, and fourth correspondence number in the combination information. The first correspondence number is the number of network identification pieces associated with the corresponding content output device for each content output device. The second correspondence number is the number of content output devices associated with the corresponding network identification information for each network identification information. The third correspondence number is the number of accounts associated with the corresponding network identification information for each network identification information. The fourth correspondence number is the number of network identification pieces associated with the corresponding account for each account. Information processing device.

6. An information processing device according to any one of claims 1 to 5, The aforementioned network identification information is an IP address. Information processing device.

7. An information processing apparatus according to claim 6, The attribute information includes at least one of the IP address's class information, ASN information, and carrier information. Information processing device.

8. An information processing device according to any one of claims 1 to 5, A display device configured to display the results predicted by the prediction unit, An information processing system equipped with the following features.

9. A method of information processing performed by a computer, To obtain combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, To determine attribute information relating to the attributes of the aforementioned network identification information, With respect to at least one of the aforementioned multiple combinations, the relationship between the user of the content output device and the owner of the account is inferred. Includes, The aforementioned inference includes inferring the association based on the attribute information of the network identification information relating to the at least one combination of inference targets, Information processing methods.

10. A computer program that causes a computer to perform a process, The aforementioned process is, To obtain combination information describing multiple combinations of content output devices and accounts whose communication has been confirmed via the same network identification information, To determine attribute information relating to the attributes of the aforementioned network identification information, With respect to at least one of the aforementioned multiple combinations, the relationship between the user of the content output device and the owner of the account is inferred. Includes, The aforementioned inference includes inferring the association based on the attribute information of the network identification information relating to the at least one combination of inference targets, Computer program.

Images