SNS information processing device, SNS information processing method and program

Abstract

This SNS information processing device comprises a first acquisition unit, a second acquisition unit, a first score acquisition unit, and a second score acquisition unit. The first acquisition unit acquires object person information including a plurality of sets of first item information corresponding to a plurality of items, respectively. The second acquisition unit acquires candidate account information which relates to a candidate for an account used by an object person and which includes a plurality of sets of second item information respectively corresponding to a plurality of items extracted from SNS information managed by an SNS system. The first score acquisition unit obtains, for each of the plurality of items, an individual-item score indicating the degree of similarity between the first item information and the second item information. The second score acquisition unit uses the individual-item score obtained for each of the plurality of items to obtain an identity score indicating the possibility that the user of the candidate account is the object person.

Description

SNS information processing device, SNS information processing method, and recording medium

[0001] The present invention relates to an SNS information processing device, an SNS information processing method, and a recording medium.

[0002] Social networking services (SNS) are widely used, and the information provided by these systems is used for various purposes, such as marketing and personnel surveys of job seekers and those hoping to enter higher education.

[0003] For example, Patent Document 1 proposes a technique for estimating the identity of a target person and a user of an account on an SNS.

[0004] The social network information processing device described in Patent Literature 1 acquires candidate social information, which is information on a social network. The candidate social information is information related to the account owners of candidate accounts extracted by narrowing down accounts on the social network based on at least some information of multiple items of target person information.

[0005] The social network information processing device described in Patent Literature 1 generates candidate-corrected social information including values ​​of multiple items that can be compared with person information based on the acquired candidate social information and a person model. The social network information processing device then compares the target person information with the candidate-corrected social information for each candidate account to determine the identity of the account owner and the target person for each candidate account.

[0006] According to Patent Literature 1, a person model is model data that uses information obtained from a social networking site in relation to a person's account as input and obtains values ​​(numerical values ​​indicating content) for each item of the person's account that can be compared with the person's information. When a value cannot be obtained from the social networking site for one item of a candidate's social information, the person model is applied to the candidate. This allows multiple types of values ​​and the occurrence probability of each value to be obtained for the item of the candidate's social information for which a value cannot be obtained in relation to other information obtained from the social networking site.

[0007] JP 2014-206791 A

[0008] As described above, in the technology described in Patent Literature 1, when a value cannot be obtained from the SNS, a person model is used to obtain the value in relation to other information obtained from the SNS. Therefore, information not included in the SNS information obtained from the SNS system is created in the person model based on the SNS information, which may result in an inaccurate determination of the identity of the account owner and the target person.

[0009] In view of the above-mentioned problems, an example of an object of the present invention is to provide an SNS information processing device, an SNS information processing method, a program, a recording medium, etc. that accurately estimate the account used by a subject on an SNS.

[0010] According to one aspect of the present invention, there is provided an SNS information processing device comprising: a first acquisition means for acquiring subject information, which is information about a subject and includes a plurality of first item information corresponding to each of a plurality of items; a second acquisition means for acquiring candidate account information, which is information about a candidate account to be used by the subject and includes a plurality of second item information corresponding to each of the plurality of items extracted from SNS information managed in an SNS system; a first score acquisition means for calculating an item-specific score indicating the degree of similarity between the first item information and the second item information for each of the plurality of items; and a second score acquisition means for calculating an identity score indicating the possibility that the user of the candidate account is the subject, using the item-specific score calculated for each of the plurality of items.

[0011] According to one aspect of the present invention, an SNS information processing method is provided in which one or more computers: acquire subject information, which is information about a subject and includes a plurality of first item information corresponding to each of a plurality of items; acquire candidate account information, which is information about a candidate account to be used by the subject and includes a plurality of second item information corresponding to each of the plurality of items extracted from SNS information managed in an SNS system; calculate an item-specific score for each of the plurality of items indicating the degree of similarity between the first item information and the second item information; and use the item-specific score calculated for each of the plurality of items to calculate an identity score indicating the possibility that the user of the candidate account is the subject.

[0012] According to one aspect of the present invention, there is provided a program for causing one or more computers to execute the following: acquire subject information, which is information about a subject, including a plurality of first item information corresponding to each of a plurality of items; acquire candidate account information, which is information about a candidate account to be used by the subject, including a plurality of second item information corresponding to each of the plurality of items extracted from SNS information managed in an SNS system; calculate an item-specific score indicating the degree of similarity between the first item information and the second item information for each of the plurality of items; and calculate an identity score indicating the likelihood that a user of the candidate account is the subject, using the item-specific score calculated for each of the plurality of items. Also, according to another aspect of the present invention, there is provided a recording medium having this program recorded thereon.

[0013] According to one aspect of the present invention, it is possible to accurately estimate the account that a subject uses on an SNS.

[0014] 1 is a diagram illustrating an overview of an SNS information processing device according to embodiment 1. FIG. 2 is a diagram illustrating an overview of SNS information processing according to embodiment 1. FIG. 3 is a diagram illustrating an example of the configuration of an SNS information processing system according to embodiment 1. FIG. 4 is a diagram illustrating an example of the physical configuration of an SNS information processing device according to embodiment 1. FIG. 5 is a flowchart illustrating an example of SNS information processing according to embodiment 1. FIG. 6 is a flowchart illustrating an example of search processing according to embodiment 1. FIG. 7 is a diagram illustrating an example of the functional configuration of a second score acquisition unit according to embodiment 2. FIG. 8 is a flowchart illustrating an example of SNS information processing according to embodiment 2. FIG. 9 is a diagram illustrating an example of item-specific weights according to embodiment 2. FIG. 10 is a diagram illustrating an example of the functional configuration of a second acquisition unit according to embodiment 2. FIG. 11 is a flowchart illustrating an example of SNS information processing according to embodiment 3. FIG. 12 is a diagram illustrating an example of the functional configuration of an SNS information processing device according to embodiment 4. FIG. 13 is a flowchart illustrating an example of SNS information processing according to embodiment 4.

[0015] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, like components are designated by like reference numerals, and the description thereof will be omitted as appropriate.

[0016] First Embodiment FIG. 1 is a diagram showing an overview of an SNS information processing device 102 according to a first embodiment.

[0017] The SNS information processing device 102 includes a first acquisition unit 122 , a second acquisition unit 123 , a first score acquisition unit 124 , and a second score acquisition unit 125 .

[0018] The first acquisition unit 122 acquires subject information that is information about the subject and includes a plurality of first item information corresponding to a plurality of items, respectively.

[0019] The second acquisition unit 123 acquires candidate account information, which is information regarding candidate accounts to be used by the target person, and includes multiple second item information corresponding to multiple items extracted from SNS information managed in the SNS system.

[0020] The first score acquisition unit 124 obtains an item score indicating the degree of similarity between the first item information and the second item information for each of the plurality of items.

[0021] The second score acquisition unit 125 uses the item-specific scores obtained for each of the multiple items to obtain an identity score indicating the possibility that the user of the candidate account is the target person.

[0022] According to this SNS information processing device 102, it is possible to accurately estimate the account that a subject uses on the SNS.

[0023] FIG. 2 is a diagram illustrating an overview of SNS information processing according to the first embodiment.

[0024] The first acquisition unit 122 acquires subject information that is information about the subject and includes a plurality of first item information corresponding to a plurality of items (Step S102).

[0025] The second acquisition unit 123 acquires candidate account information, which is information regarding candidate accounts to be used by the target person, and includes multiple second item information corresponding to multiple items extracted from SNS information managed in the SNS system (step S104).

[0026] The first score acquisition unit 124 obtains an item score indicating the degree of similarity between the first item information and the second item information for each of the plurality of items (step S105).

[0027] The second score acquisition unit 125 uses the item-specific scores obtained for each of the multiple items to obtain an identity score indicating the possibility that the user of the candidate account is the target person (step S106).

[0028] This SNS information processing makes it possible to accurately estimate the account that the subject uses on the SNS.

[0029] A detailed example of the SNS information processing system according to the first embodiment will be described below.

[0030] 3 is a diagram illustrating an example of the configuration of an SNS (social networking service) information processing system 100 according to embodiment 1. The SNS information processing system 100 is a system for estimating an account used by a subject person on an SNS.

[0031] The SNS information processing system 100 includes, for example, an SNS system 101 and an SNS information processing device 102. The SNS system 101 and the SNS information processing device 102 are connected to each other via a network NT so as to be able to send and receive information to and from each other. The network NT is a communication line that may be wired, wireless, or a combination of these.

[0032] (Configuration Example of SNS System 101 According to First Embodiment) The SNS system 101 is a system that provides an SNS to a plurality of SNS users. The SNS system 101 may be a general system for providing an SNS, and an example thereof will be described here.

[0033] The SNS system 101 includes, for example, an SNS server 111 that manages SNS information, and a plurality of terminals 112 that are used by a plurality of SNS users. The SNS information is, for example, information about each of the SNS users.

[0034] The SNS server 111 transmits and receives SNS information to and from each of the multiple terminals 112, and stores the SNS information, for example.

[0035] The SNS information includes, for example, one or more of a user ID (Identification), profile information, sending information, receiving information, etc. for each account of an SNS user. In the SNS information, for example, other information is associated with the user ID.

[0036] The user ID is information for identifying each account in the SNS system 101. For example, the user ID is automatically assigned to the SNS user when the SNS user registers as a user on the SNS. For example, the user ID does not have to be changeable by the SNS user.

[0037] The profile information includes a profile of the SNS user, which may include, for example, one or more of: name(s), date of birth, address(es), alma mater(s), phone number(s), and comment(s).

[0038] Each of the one or more names is, for example, a name that an SNS user uses to identify himself or herself to other SNS users on the SNS.

[0039] In more detail, for example, the name is a name arbitrarily set by the SNS user. The name may be a name that can identify each SNS user in the SNS system 101 (i.e., a name that is not duplicated among multiple SNS users), or a name that allows duplication among multiple SNS users. Examples of names that allow duplication among multiple SNS users include the name of the SNS user himself / herself, the names of the SNS user's relatives (e.g., parents, siblings, etc.), and names used by the SNS user on the SNS.

[0040] Here, the name may be either the last name only, the first name only, or both the last name and the first name, and the same applies hereinafter.

[0041] The address indicates one or more of the SNS user's place of residence, place of work, permanent domicile, and the like.

[0042] The alma mater is one or more of the elementary school, junior high school, high school, university, etc. that the SNS user attended.

[0043] Each of the one or more phone numbers is a phone number used by the SNS user, and may include, for example, one or more of a phone number used by the SNS user personally, a phone number at the SNS user's workplace, and the like.

[0044] The comment is, for example, a comment that an SNS user arbitrarily writes on the SNS to introduce themselves or the like.

[0045] The transmitted information is information transmitted from an account of an SNS user. In other words, the transmitted information is information received by the SNS server 111 from the terminal 112 used by the SNS user together with the user ID, for example.

[0046] The transmitted information includes, for example, one or more of unspecified transmitted information, specified transmitted information, etc. Unspecified transmitted information is information transmitted from an account of an SNS user to unspecified others, such as posted information. Specific transmitted information is information transmitted from an account of an SNS user to one or more specific accounts, such as comments, replies, chats, etc. in response to posts by others.

[0047] Received information is information received by an SNS user's account from another account, such as a comment, reply, or chat from another person.

[0048] Each of the transmitted information and the received information may include, for example, location information indicating the location from which the information was transmitted. This location information may be obtained using, for example, a GPS (Global Positioning System) function provided in the terminal 112.

[0049] The transmitted information and the received information may each include, for example, a transmission time when the transmitted information was transmitted and a reception time when the received information was received. The reception time may be at least one of the time when the SNS server 111 received the information and the time when the SNS user viewed the information. The transmission time and the reception time may each be expressed, for example, by a date and time.

[0050] It should be noted that the information included in the SNS information is not limited to the examples given here.

[0051] (Example of functional configuration of SNS information processing device 102 according to embodiment 1) The SNS information processing device 102 is a device that processes SNS information managed by the SNS system 101. The SNS information processing device 102 processes the SNS information and uses the SNS information to estimate accounts used by a subject on the SNS. The SNS information processing device 102 functionally includes, for example, a search unit 121, a first acquisition unit 122, a second acquisition unit 123, a first score acquisition unit 124, a second score acquisition unit 125, and a related account acquisition unit 126.

[0052] The search unit 121 acquires SNS information about one or more candidate accounts based on, for example, search conditions that are input as appropriate.

[0053] The candidate account is an account that may be used by the target person among the accounts in the SNS system 101. The search criteria is information including criteria for searching the accounts in the SNS system 101 for accounts that may be used by the target person.

[0054] The first acquisition unit 122 acquires subject information.

[0055] The subject information is information about a subject, and includes a plurality of first item information corresponding to a plurality of predetermined items.

[0056] The second acquiring unit 123 acquires the candidate account information.

[0057] The candidate account information is information about candidates for accounts to be used by the target person. The candidate account information includes a plurality of second item information corresponding to a plurality of items extracted from the SNS information managed by the SNS system 101.

[0058] Here, the multiple items used in each of the first item information and the second item information may be, for example, predetermined common items. The multiple items include, for example, two or more of name, date of birth, places visited, address, alma mater, area of ​​activity, and belongings. As described above, the name may be only the last name, only the first name, or both the last name and first name. The places visited are places visited during travel, business trips, etc. As described above, the address is an address indicating one or more of the residence, workplace, and registered domicile. In the first item information, the activity area is the area in which the subject is active, such as the subject's residence or workplace, where the subject is active with a predetermined frequency or more. In the second item information, the activity area is the area in which the user of the candidate account is active.

[0059] The first score acquisition unit 124 obtains an item score for each of the plurality of items.

[0060] The item score is a value (similarity) indicating the degree of similarity between first item information and second item information relating to a common item.

[0061] The second score obtaining unit 125 obtains an identity score using the item-specific scores obtained for each of the plurality of items.

[0062] The identity score is a value that indicates the likelihood that the user of the candidate account is the target person.

[0063] The related account acquisition unit 126 identifies the target person's account from among the candidate accounts based on the identity score. Furthermore, the related account acquisition unit 126 may identify other accounts based on the identified target person's account.

[0064] (Example of Physical Configuration of SNS Information Processing System 100 According to First Embodiment) The SNS information processing system 100 is physically configured, for example, by an SNS server 111, a plurality of terminals 112, and an SNS information processing device 102, which are connected via a network NT.

[0065] 4 is a diagram showing an example of the physical configuration of the SNS information processing device 102 according to embodiment 1. The SNS information processing device 102 physically includes, for example, a bus 1010, a processor 1020, a memory 1030, a storage device 1040, a network interface 1050, an input interface 1060, and an output interface 1070.

[0066] The bus 1010 is a data transmission path for transmitting and receiving data among the processor 1020, memory 1030, storage device 1040, network interface 1050, input interface 1060, and output interface 1070. However, the method of connecting the processor 1020 and the like to each other is not limited to bus connection.

[0067] The processor 1020 is implemented as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit).

[0068] The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.

[0069] The storage device 1040 is an auxiliary storage device realized by a hard disk drive (HDD), a solid state drive (SSD), a memory card, a read-only memory (ROM), or the like. The storage device 1040 stores program modules for realizing the functions of the device that includes it (in the example of FIG. 4, the SNS information processing device 102). The processor 1020 loads each of these program modules into the memory 1030 and executes them, thereby realizing the function corresponding to that program module.

[0070] The network interface 1050 is an interface for connecting a device having the network interface 1050 (the SNS information processing device 102 in the example of FIG. 4) to the network NT.

[0071] The input interface 1060 is an interface for the user to input information, and is configured from, for example, a touch panel, a keyboard, a mouse, and the like.

[0072] The output interface 1070 is an interface for presenting information to the user, and is configured, for example, by a liquid crystal panel, an organic EL (Electro-Luminescence) panel, or the like.

[0073] It should be noted that the SNS server 111 and each of the multiple terminals 112 may be physically configured in the same manner as, for example, the SNS information processing device 102. The SNS information processing device 102 may be physically configured from multiple devices having the configuration described with reference to FIG.

[0074] (Example of Operation of SNS Information Processing System 100 According to First Embodiment) The SNS information processing system 100 according to the first embodiment executes various information processes, including SNS information processing.

[0075] FIG. 5 is a flowchart illustrating an example of SNS information processing according to the first embodiment.

[0076] The SNS information processing is a process for estimating an account used by a subject on an SNS by using SNS information managed by the SNS system 101 (for example, SNS information stored in the SNS server 111). The SNS information processing is started, for example, in response to an instruction from an operator of the SNS information processing device 102.

[0077] The search unit 121 executes a search process (step S101). The search process is a process of acquiring SNS information about a candidate account from the SNS information managed by the SNS system 101.

[0078] FIG. 6 is a flowchart showing an example of the search process (step S101) according to the first embodiment.

[0079] The search unit 121 acquires search conditions based on, for example, an input by an operator (step S101a).

[0080] The search conditions include, for example, one or more search keywords, and, if there are multiple search keywords, the relationship between the search keywords.

[0081] The search keyword is a keyword for searching for accounts in the SNS system 101 that may be used by the target person. The search keyword is, for example, one or more of the target person's full or partial name, date of birth, places visited, address, etc. The keyword may include, for example, words related to an event that the target person is likely to be associated with.

[0082] The relationship between search keywords includes combining search keywords with "and" and combining search keywords with "or".

[0083] The search unit 121 uses the search conditions acquired in step S101 to narrow down the SNS information managed by the SNS system 101 to one or more candidate accounts (step S101b).

[0084] For example, the search unit 121 narrows down the SNS information of one or more accounts that match the search criteria as candidate accounts. The SNS information of the accounts that match the search criteria is, for example, SNS information that includes the search keywords included in the search criteria in a relationship defined by the search criteria.

[0085] The search unit 121 acquires SNS information about the candidate accounts narrowed down in step S101b (step S101c).

[0086] The SNS information about the candidate account includes, for example, one or more of the user ID, profile information, sending information, receiving information, etc. about the candidate account. This information is as described above.

[0087] In detail, for example, the outgoing information about the candidate account is unspecified outgoing information sent from the candidate account to unspecified others, specific outgoing information sent from the candidate account to one or more specific accounts, etc. The received information about the candidate account is received information received by the candidate account from other accounts, etc.

[0088] 5 again, the first acquisition unit 122 acquires subject information including a plurality of first item information corresponding to a plurality of predetermined items (step S102). For example, the first acquisition unit 122 acquires the subject information based on an input from an operator.

[0089] The second acquisition unit 123 repeatedly executes steps S104 to S106 for each of the one or more candidate accounts narrowed down in step S101b (loop A; step S103).

[0090] The second acquisition unit 123 acquires candidate account information including a plurality of second item information corresponding to each of the plurality of items extracted from the SNS information acquired in step S101c (step S104).

[0091] For example, the second acquisition unit 123 acquires a plurality of pieces of second item information by extracting information corresponding to each of a plurality of items directly from the SNS information acquired in step S101c. The second acquisition unit 123 generates candidate account information including the acquired plurality of pieces of second item information. One or more pieces of second item information among the plurality of pieces of second item information may include the plurality of pieces of information extracted from the SNS information.

[0092] The first score acquisition unit 124 uses the subject information and candidate account information acquired in steps S102 and S104, respectively, to obtain a plurality of item scores for each of the plurality of items (step S105).

[0093] As described above, the item score is the degree of similarity between the first item information and the second item information regarding a common item. The method for calculating the degree of similarity may be determined appropriately.

[0094] For example, the score for each item may be set to a predetermined value depending on whether the first item information and the second item information for that item match. In this case, the score for each item may be, for example, "1.0" if they match and "0.0" if they do not match.

[0095] For example, the item score may be set to a value (match rate) according to the degree to which the first item information and the second item information match for that item. In this case, the item score may be set to a value within a predetermined range, such as "0.0" to "1.0," with a higher value corresponding to a higher degree of match. For example, if the second item information includes one piece of information, the match rate may be the degree to which that piece of information matches the first item information. For example, if the second item information includes multiple pieces of information, the match rate may be the proportion of the multiple pieces of information that match the first item information.

[0096] The item scores are not limited to the examples given here, and may be smaller values ​​as the degree of match increases, for example.

[0097] For example, when information corresponding to an item is represented by characters such as a name, the item score may be calculated according to the edit distance between the characters included in the first item information and the second item information. In detail, the item score may be calculated according to a formula or rule such that the score is "1.0" when the characters match (i.e., the edit distance is 0), and approaches "0.0" as the edit distance increases.

[0098] For example, if the information corresponding to an item is a name, the matching factor may be whether or not the names included in the first item information and the second item information match, or the degree to which the names match (match rate).Furthermore, for example, if the second item information includes multiple pieces of information, the match rate may be calculated by weighting using the edit distance between each of the multiple pieces of information and the characters included in the first item information.

[0099] For example, when information corresponding to an item is indicated by a point or location such as an address, the item score may be calculated according to the geographical distance to the point or location. In detail, for example, the item score may be calculated according to a formula such that the item score is "1.0" when the distance is within a predetermined range and approaches "0.0" as the distance increases. Alternatively, the item score may be calculated according to a predetermined rule such that the item score is "1.0" when the distance is within a predetermined range and gradually approaches "0.0" as the distance increases.

[0100] For example, when information corresponding to an item is expressed as a number, such as a date of birth, the item score may be calculated according to the difference between the numbers included in the first item information and the second item information. In detail, for example, the item score may be calculated according to a formula or rule such that the score is "1.0" when the numbers match, and approaches "0.0" as the difference in the numbers increases.

[0101] For example, if the second item information for a certain item cannot be obtained in step S104, the second item information for that item may be left blank. In this case, the item-specific score for that item may be a predetermined value (e.g., "0.5" or "0.0").

[0102] The method for calculating the scores for each item is not limited to these.

[0103] The second score obtaining unit 125 obtains an identity score using the scores for each of the plurality of items obtained in step S105 (step S106).

[0104] The identity score may be, for example, a simple average or a weighted average of the scores for multiple items. The weighted average may be calculated by applying multiple item weights corresponding to the multiple items to the multiple item scores calculated in step S105. In more detail, for example, the weighted average is the sum of values ​​obtained by multiplying the item weights and the item scores associated with the same item. The multiple item weights may be determined in advance, for example. However, the item weights are not limited to this.

[0105] When steps S104 to S106 are executed for each of one or more candidate accounts (loop A; step S103), the related account acquisition unit 126 identifies the target person's account from among the candidate accounts (step S107) and terminates the SNS information processing.

[0106] The related account acquisition unit 126 identifies the target person's account from among the candidate accounts based on the identity score calculated in step S106. For example, the related account acquisition unit 126 may identify the candidate account with the highest identity score as the target person's account.

[0107] In step S107, the related account acquisition unit 126 may further identify other accounts (related accounts) related to the identified target person's account.

[0108] An associated account is, for example, an account that frequently sends and receives information to and from the target account (e.g., at least a predetermined frequency, at least a predetermined number of times, etc.) or an account that has a predetermined relationship on SNS (e.g., a relationship in which posts can be viewed, a relationship in which replies can be made, etc.).

[0109] (Operations and Effects) As described above, according to this embodiment, the SNS information processing device 102 includes the first acquisition unit 122 , the second acquisition unit 123 , the first score acquisition unit 124 , and the second score acquisition unit 125 .

[0110] The first acquisition unit 122 acquires subject information that is information about the subject and includes a plurality of first item information corresponding to a plurality of items, respectively.

[0111] The second acquisition unit 123 acquires candidate account information, which is information regarding candidate accounts to be used by the target person, and includes multiple second item information corresponding to multiple items extracted from SNS information managed in the SNS system.

[0112] The first score acquisition unit 124 obtains an item score indicating the degree of similarity between the first item information and the second item information for each of the plurality of items.

[0113] The second score acquisition unit 125 uses the item-specific scores obtained for each of the multiple items to obtain an identity score indicating the possibility that the user of the candidate account is the target person.

[0114] This allows the account used by the target person on the SNS to be estimated using the plurality of second item information corresponding to the plurality of items extracted from the SNS information, thereby enabling the account used by the target person on the SNS to be estimated with high accuracy.

[0115] According to this embodiment, the identity score is a weighted average calculated by applying a plurality of item-specific weights corresponding to the plurality of items to the calculated item-specific scores.

[0116] This allows an appropriate identity score to be used to estimate the account that the subject uses on the SNS, thereby making it possible to accurately estimate the account that the subject uses on the SNS.

[0117] According to this embodiment, the SNS information includes at least one of unspecified outgoing information sent from the candidate account to unspecified others, specific outgoing information sent from the candidate account to one or more specific accounts, and received information received from other accounts.

[0118] This makes it possible to estimate the account used by the target person on the SNS using information contained in sent or received text such as posts, comments, replies, chats, etc. Therefore, it becomes possible to accurately estimate the account used by the target person on the SNS.

[0119] According to this embodiment, the plurality of items includes at least one of an address of a visited place and an address of a residential place.

[0120] This makes it possible to estimate the account that the subject uses on the SNS using geographical information such as the place of visit, place of residence, etc. Therefore, it becomes possible to accurately estimate the account that the subject uses on the SNS.

[0121] In this embodiment, an example will be described in which at least some of the item weights used to calculate the identity score are determined based on the content of the second item information. In this embodiment, for the sake of simplicity, descriptions that overlap with those in the first embodiment will be omitted as appropriate.

[0122] (Functional Configuration of the Second Score Acquisition Unit 225 According to Embodiment 2) The SNS information processing device according to this embodiment includes a second score acquisition unit 225 instead of the second score acquisition unit 125 according to Embodiment 1. The second score acquisition unit 225 is similar to the second score acquisition unit 125 according to Embodiment 1 in that it calculates an identity score using item-specific scores calculated for each of a plurality of items.

[0123] 7 is a diagram illustrating an example of the functional configuration of the second score acquisition unit 225 according to embodiment 2. The second score acquisition unit 225 includes a weight determination unit 225a and a score calculation unit 225b.

[0124] The weight determination unit 225a determines a plurality of item-specific weights. For example, the weight determination unit 225a determines at least some of the item-specific weights based on the content of second item information corresponding to items that are common to the item-specific weights. Note that the weight determination unit 225a may apply, for example, predetermined item-specific weights to the remaining part of the item-specific weights.

[0125] The score calculation unit 225b calculates the identity score using the item-specific scores and item-specific weights calculated for each of the multiple items.

[0126] (Example of SNS Information Processing According to Embodiment 2) Fig. 8 is a flowchart showing an example of SNS information processing according to embodiment 2. The SNS information processing according to this embodiment includes an identity score acquisition process (step S206) instead of the identity score acquisition process (step S106) according to embodiment 1.

[0127] For example, following steps S101 to S104 similar to those in the first embodiment, the second score acquisition unit 225 calculates an identity score using the scores for each of the items calculated in step S105 (step S206).

[0128] In step S206, the weight determining unit 225a determines weights for a plurality of items (step S206a).

[0129] For example, the weight determination unit 225a determines at least some of the multiple item weights based on the rarity of the content of the second item information corresponding to the item that is common to the item weight. For example, the higher the rarity, the larger the item weight that is determined.

[0130] 9 is a diagram showing an example of weights by item according to the second embodiment. The figure shows an example in which the items are "Name," "Date of Birth," and "Place of Residence." Since the first item information and the second item information match for each of "Name," "Date of Birth," and "Place of Residence," the similarity is "1.0." This similarity is calculated, for example, in step S105.

[0131] The item weight for the item "name" is a predetermined value of "1.0."

[0132] In the figure, an example is shown in which the item weights for the items "date of birth" and "place of residence" are determined based on the rarity of the content of the second item information.

[0133] In this example, the item weight for the item "date of birth" is calculated according to the rarity of people with common content (for example, a date of birth of "xx month △ day") at an address related to the subject (for example, "Y city, X prefecture" as a place of residence). In more detail, for example, assume that the population of "Y city, X prefecture" is P (people) and the number of people born on "xx month △ day" in "Y city, X prefecture" is D (people). In this case, the item weight for the item "date of birth" is calculated using the formula 1-(D / P), as shown in FIG. 9.

[0134] Here, the item-specific weights may be determined by further considering other attributes of the subject. In detail, for example, the weight determination unit 225a may determine the subject's age from the date of birth. Then, the weight determination unit 225a may determine the item-specific weight for the item "date of birth" by using the number of people in "Y city, X prefecture" who are the same age as the subject and were born on "XX month, △ day" instead of the above-mentioned D (people).

[0135] The item weight for the item "place of residence" is calculated, for example, according to the rarity of the address associated with the subject (for example, the place of residence, "Y City, X Prefecture"). In more detail, for example, assume that the population of "Y City, X Prefecture" is P (people). In this case, the item weight for the item "place of residence" is calculated by the formula 1 - (P / total population of Japan), as shown in FIG. 9.

[0136] The statistical data such as population used to determine the weight for each item may be stored in advance by the weight determination unit 225a, or may be obtained from an external device (not shown) via the network NT or the like.

[0137] The method for determining the weight for each item is not limited to this.

[0138] Referring again to Fig. 8, the score calculation unit 225b calculates the identity score using the item scores and item weights calculated in steps S105 and S206a (step S206b).

[0139] For example, the score calculation unit 225b calculates the product of the item-specific score and the item-specific weight for each item. In the example of FIG. 9, for example, the product for the item "Name" is 1.0 x 1.0. The product for the item "Date of Birth" is 1.0 x (1 - (D / P)). The product for the item "Place of Residence" is 1.0 x (1 - (P / Japan's total population)). The score calculation unit 225b calculates the sum of the products for these items (= 1.0 x 1.0 + 1.0 x (1 - (D / P)) + 1.0 x (1 - (P / Japan's total population))) as the identity score.

[0140] Referring to FIG. 8, when loop A (step S103) ends, step S107 is subsequently executed.

[0141] (Operations and Effects) As described above, according to this embodiment, at least some of the weights for each of the plurality of items are determined based on the content of the second item information.

[0142] This allows an appropriate identity score to be used to estimate the account that the subject uses on the SNS, thereby making it possible to accurately estimate the account that the subject uses on the SNS.

[0143] According to this embodiment, at least a portion of the multiple item weights is determined based on the rarity of the content of the second item information.

[0144] Generally, for example, if rare information matches between the first item information and the second item information, the SNS account from which the second item information was extracted is likely to belong to the subject. Therefore, by determining the item weight based on the rarity of the content of the second item information, an appropriate identity score can be used to estimate the SNS account used by the subject. Therefore, it is possible to accurately estimate the SNS account used by the subject.

[0145] (Embodiment 3) In the first embodiment, an example was described in which, when the second item information includes a plurality of pieces of information, the plurality of pieces of information is used to calculate item-specific scores. When the second item information includes a plurality of pieces of information, the plurality of pieces of information may be used as candidates for the second item information (second item candidate information), and information narrowed down (identified) from the plurality of second item candidate information may be used as the second item information. In the present embodiment, an example is described in which, when a plurality of pieces of second item candidate information are extracted from SNS information as candidates for the second item information, the second item information is estimated (identified) by statistically processing the plurality of second item candidate information. In the present embodiment, for the sake of simplicity, explanations that overlap with those of the first embodiment will be omitted as appropriate.

[0146] (Functional Configuration of Second Acquisition Unit 323 According to Embodiment 3) The SNS information processing device according to this embodiment includes a second acquisition unit 323 instead of the second acquisition unit 123 according to Embodiment 1. The second acquisition unit 323 is similar to the second acquisition unit 123 according to Embodiment 1 in that it acquires candidate account information including a plurality of pieces of second item information corresponding to a plurality of items extracted from SNS information managed by the SNS system 101.

[0147] 10 is a diagram illustrating an example of the functional configuration of the second acquisition unit 323 according to embodiment 2. The second acquisition unit 323 includes an extraction unit 323a and an estimation unit 323b.

[0148] The retrieval unit 323a retrieves a plurality of pieces of second item candidate information corresponding to a plurality of items, respectively, from the SNS information managed by the SNS system 101. The second item candidate information is information that is a candidate for the second item information.

[0149] The estimation unit 323b estimates second item information for each of the plurality of items using second item candidate information extracted from the SNS information.

[0150] For example, when a plurality of pieces of second item candidate information are extracted from the SNS information as candidates for second item information corresponding to a certain item, the estimation unit 323b statistically processes the plurality of pieces of second item candidate information to estimate the second item information corresponding to the item. Also, when a single piece of second item candidate information is extracted for a certain item, the estimation unit 323b may set the second item candidate information as the second item information.

[0151] (Example of SNS Information Processing According to Embodiment 3) FIG. 11 is a flowchart showing an example of SNS information processing according to embodiment 3. The SNS information processing according to this embodiment includes a candidate account information acquisition process (step S204) instead of the candidate account information acquisition process (step S104) according to embodiment 1.

[0152] For example, following steps S101 to S103 similar to those in embodiment 1, the second acquisition unit 323 acquires candidate account information including multiple second item information corresponding to each of the multiple items extracted from the SNS information acquired in step S101c (step S204).

[0153] In step S204, the extracting unit 323a extracts a plurality of second item candidate information corresponding to each of a plurality of items from the SNS information (step S204a).

[0154] For example, the retrieval unit 323a may retrieve the second item candidate information using a trained item-specific learning model corresponding to each of the multiple items. The item-specific learning model may retrieve the second item candidate information corresponding to each of the multiple items when SNS information is input. The item-specific learning model may retrieve information corresponding to the items directly from the SNS information without changing the information contained in the SNS information.

[0155] The item-specific learning model may be learned using SNS information for learning and correct answer data indicating the correct answer to the second item candidate information included in the SNS information for learning.

[0156] In more detail, for example, assume that the items are "name," "date of birth," and "place of residence." In this case, for example, the retrieval unit 323a uses a name model, a birth date model, and a place of residence model as item-specific learning models for retrieving "name," "date of birth," and "place of residence," respectively.

[0157] For example, the extraction unit 323a inputs the outgoing information and the received information into a name model, and extracts proper nouns from conversation histories with friends included in one or more of the outgoing information and the received information.

[0158] For example, the extraction unit 323a inputs the outgoing information and received information into a name model and extracts the date listed as the birthday from posts, conversation history, etc. included in one or more of the outgoing information and received information.

[0159] For example, the retrieval unit 323a inputs the outgoing information and the received information into a residence model and retrieves information indicating a place that is estimated to be the residence from a place name, an address, etc. included in one or more of the outgoing information and the received information. Furthermore, for example, the retrieval unit 323a may retrieve the residence using location information included in one or more of the outgoing information and the received information.

[0160] The estimation unit 323b estimates second item information for each of the plurality of items using the second item candidate information extracted in step S204a (step S204b).

[0161] For example, as in the above, it is assumed that the items are "name," "date of birth," and "place of residence," and multiple pieces of second item candidate information are extracted for each of them.

[0162] In detail, for example, assume that in step S204a, for the item "Name", the name "B" appears four times, the name "C" appears ten times, and the name "D" appears once from the SNS information. Of these pieces of second item candidate information, the estimation unit 323b estimates "C", which is most frequently included in the SNS information, as the second item information corresponding to "Name".

[0163] For example, suppose that in step S204a, for the item "Date of Birth," "March 7, 2001" is extracted once, "March 8, 2001" is extracted four times, and "March 9, 2001" is extracted once from the SNS information. Of these pieces of second item candidate information, the estimation unit 323b estimates "March 8, 2001," which is most frequently included in the SNS information, as the second item information corresponding to "Date of Birth."

[0164] For example, in step S204a, for the item "place of residence," it is assumed that, of all the location information, "Y City, X Prefecture" is extracted from the SNS information at a rate of 0.8, "W City, Z Prefecture" is extracted at a rate of 0.1, etc. The estimation unit 323b estimates that, of these pieces of second item candidate information, "Y City, X Prefecture," which is most frequently included in the SNS information, is the second item information corresponding to "place of residence."

[0165] Here, an example has been described in which, when there are multiple pieces of second item candidate information for a certain item, the multiple pieces of second item candidate information are statistically processed and the second item candidate information that is most frequently included in the SNS information is estimated as the second item information. The method of statistically processing multiple pieces of second item candidate information and estimating the second item information from the multiple pieces of second item candidate information is not limited to this.

[0166] Referring to FIG. 11, step S105 is then executed.

[0167] (Actions and Effects) As described above, according to this embodiment, when there are multiple pieces of second item candidate information extracted from SNS information as candidates for the second item information, the second item information is estimated by statistically processing the multiple pieces of second item candidate information.

[0168] This allows the second item information to be estimated without changing information such as wording in sentences included in the SNS information, thereby making it possible to accurately estimate the account used by the target person on the SNS.

[0169] (Variation 1) The second item information related to some of the multiple items may include information estimated from SNS information. This estimation may be performed using, for example, a machine learning model. Here, an example will be described in which the second item information related to the item "activity area" is estimated from SNS information using a machine learning model.

[0170] When the second acquisition unit 123 inputs the SNS information acquired in step S101c, it may acquire second item information related to the item "activity area" using an activity area estimation model that outputs an activity area estimated from the SNS information for the user of the candidate account.

[0171] In this case, the activity area estimation model is, for example, a machine learning model trained using a learning model including SNS information for training and activity areas. The activity area estimation model may output multiple activity areas as estimation results. In this case, the activity area estimation model may output, along with each activity area, the proportion of users of the candidate account who are active in each activity area (activity proportion). Furthermore, the activity area estimation model may output, along with each activity area, a reliability indicating the likelihood that users of the candidate account will be active in each activity area.

[0172] The second acquisition unit 123 may use the activity rate or the reliability to identify the second item information, such as by determining the activity area with the highest activity rate or reliability as the second item information. In addition, the activity rate or the reliability may be used as a weight for calculating the item-specific score.

[0173] By using such an estimation result as the second item information, it is possible to estimate the account used by the subject on the SNS by further using information other than the information extracted from the SNS information. Therefore, it is possible to accurately estimate the account used by the subject on the SNS.

[0174] (Embodiment 4) In this embodiment, an example will be described in which, when SNS information includes an image, the image is used to estimate the account used by a target person on SNS. The image may be included in, for example, one or more of profile information, sent information, received information, etc. The image may be a still image or a moving image. In this embodiment, an example will be described in which multiple items include items that use images (i.e., items related to images), such as facial images. In this embodiment, to simplify the explanation, explanations that overlap with those in embodiment 1 will be omitted as appropriate.

[0175] 12 is a diagram showing an example of the functional configuration of the SNS information processing device 402 according to embodiment 4. The SNS information processing device 402 according to this embodiment includes a first score acquisition unit 424 instead of the first score acquisition unit 124 according to embodiment 1. The first score acquisition unit 424 is similar to the first score acquisition unit 124 according to embodiment 1 in that it obtains an item-specific score for each of a plurality of items.

[0176] When the SNS information includes an image, the first score acquisition unit 424 further uses the image to obtain an item-specific score for each of the multiple items.

[0177] For example, the first score acquisition unit 424 may obtain the similarity between the first item information and the second item information corresponding to an item related to an image as the item-specific score for that item. This similarity is the similarity between images obtained by processing the images included in the first item information and the second item information. A general technique may be used to obtain the similarity between images.

[0178] For example, when an image included in the SNS information matches the content of at least one of the first item information and the second item information, the first score acquisition unit 424 calculates an item-specific score by adding a predetermined value to the similarity regarding an item that is common to at least one of the first item information and the second item information.

[0179] 13 is a flowchart showing an example of SNS information processing according to embodiment 4. The SNS information processing according to this embodiment includes an item-specific score acquisition process (step S405) instead of the item-specific score acquisition process (step S105) according to embodiment 1.

[0180] For example, steps S101 to S104 similar to those in the first embodiment are executed.

[0181] Similar to step S105 in embodiment 1, the first score acquisition unit 424 uses the subject information and candidate account information acquired in steps S102 and S104, respectively, to obtain multiple item-specific scores for each of the multiple items (step S405).

[0182] For items related to images, first item information and second item information including images may be acquired in steps S102 and S104, respectively.

[0183] In the item score acquisition process (step S405) according to this embodiment, if the SNS information includes an image, the first score acquisition unit 424 further uses the image to obtain an item score for each of the multiple items.

[0184] (Example of a Case Where the Image is a Facial Image) For example, assume that multiple items include "image of the person himself / herself."

[0185] The first score acquisition unit 424 obtains, as an item score, the degree of similarity between the images included in the first item information and the second item information related to the "personal image" acquired in steps S102 and S104, respectively.

[0186] (Example of a case where an image is an image taken at a certain location) For example, suppose that multiple items include "place of residence".

[0187] The first score acquisition unit 424 calculates the degree of similarity between the first item information and the second item information regarding “place of residence.” Furthermore, the first score acquisition unit 424 processes an image included in the SNS information to identify the area where the image was taken.

[0188] In detail, for example, the first score acquisition unit 424 calculates the score for each item based on whether or not information obtained from an image included in the SNS information corresponds to information included in at least one of the first item information and the second item information. The information obtained from the image includes, for example, the area where the image was taken (photography area), the time (photography time), etc.

[0189] More specifically, for example, when the area where the image was taken is included in at least one of the first item information and the second item information related to "place of residence," the first score acquisition unit 424 adds a predetermined value (e.g., 0.5, 1.0, etc.) to the obtained similarity to obtain the item-specific score. When the area where the image was taken is not included in either the first item information or the second item information related to "place of residence," the first score acquisition unit 424 sets the obtained similarity as the item-specific score without adding the predetermined value.

[0190] When the shooting time is used, the first score acquisition unit 424 may calculate the item-specific score based on, for example, whether the shooting time is within a time period included in at least one of the first item information and the second item information (e.g., a predetermined time range from the time of visiting the place of interest). For example, when the shooting time is within a time period included in at least one of the first item information and the second item information, the first score acquisition unit 424 may calculate the item-specific score by adding a predetermined value to the calculated similarity. When the shooting time is not within a time period included in at least one of the first item information and the second item information, the first score acquisition unit 424 may use the calculated similarity as the item-specific score.

[0191] (Example of a case where an image includes personal belongings) For example, assume that multiple items include "personal belongings."

[0192] For example, the first score acquisition unit 424 may obtain the similarity between the images included in the first item information and the second item information regarding the "belongings" acquired in steps S102 and S104, respectively, as an item-specific score.

[0193] For example, if one of the first item information and the second item information is text information indicating belongings and the other is an image, the first score acquisition unit 424 may calculate the score for each item based on whether the item indicated by the text information matches the item included in the image.

[0194] In addition, for example, if both the first item information and the second item information are text information indicating belongings, the first score acquisition unit 424 may calculate an item-specific score by adding a predetermined value (e.g., 0.5, 1.0, etc.) to the calculated similarity when an item indicated by text information included in at least one of the first item information and the second item information matches an item included in the image.

[0195] Following the process of step S405, the processes from step S106 onwards are executed, for example, as in the first embodiment.

[0196] (Operations and Effects) As described above, according to this embodiment, the plurality of items includes at least one of a face image, a place visited, and an address.

[0197] This allows the account used by the subject on the SNS to be estimated further using the image, thereby making it possible to more accurately estimate the account used by the subject on the SNS.

[0198] According to this embodiment, the item score for the item relating to the image is the degree of similarity between the images obtained by processing the images included in the first item information and the second item information.

[0199] This allows the account used by the subject on the SNS to be estimated by further using the similarity of the images, thereby making it possible to more accurately estimate the account used by the subject on the SNS.

[0200] Although the embodiments of the present invention have been described above with reference to the drawings, these are merely examples of the present invention, and various other configurations can also be adopted.

[0201] In addition, although the flowcharts used in the above description show multiple steps (processes) in a sequential order, the order of steps executed in each embodiment is not limited to the order shown. In each embodiment, the order of the steps shown in the drawings can be changed as long as it does not cause any problems in terms of content. Furthermore, the above-described embodiments can be combined as long as the content is not contradictory.

[0202] Some or all of the above embodiments may be described as, but are not limited to, the following notes: 1. An SNS information processing device comprising: a first acquisition means for acquiring subject information, which is information about a subject and includes a plurality of first items of information corresponding to each of a plurality of items; a second acquisition means for acquiring candidate account information, which is information about a candidate account to be used by the subject and includes a plurality of second items of information corresponding to each of the plurality of items extracted from SNS information managed in an SNS system; a first score acquisition means for calculating an item-specific score indicating the degree of similarity between the first items of information and the second items of information for each of the plurality of items; and a second score acquisition means for calculating an identity score indicating the possibility that a user of the candidate account is the subject, using the item-specific score calculated for each of the plurality of items. 2. The SNS information processing device described in 1., wherein the identity score is a weighted average calculated by applying a plurality of item-specific weights corresponding to each of the plurality of items to the calculated item-specific scores. 3. At least a portion of the item-specific weights are determined based on the content of the second item information. 2. The SNS information processing device described in any one of 1. to 5., wherein: 4. At least a part of the weights for each of the plurality of items is determined based on the rarity of the content of the second item information. 5. The SNS information processing device described in any one of 1. to 4., wherein the SNS information includes at least one of unspecified sent information sent from the candidate account to unspecified others, specific sent information sent from the candidate account to one or more specific accounts, and received information received from other accounts. 6. The SNS information processing device described in any one of 1. to 5., wherein, when there are multiple pieces of second item candidate information extracted from the SNS information as candidates for the second item information, the second item information is estimated by statistically processing the multiple pieces of second item candidate information. 7. The SNS information processing device described in any one of 1. to 6., wherein the multiple items include at least one of a face image, an address of a place visited, and an address of a place of residence.8. The SNS information processing device according to 7., wherein the item-specific score for the item related to an image is a similarity between the images obtained by processing the images included in the first item information and the second item information. 9. An SNS information processing system comprising: an SNS system that provides an SNS to a plurality of SNS users; and an SNS information processing device that processes SNS information managed by the SNS system, wherein the SNS information processing device comprises: a first acquisition means that acquires subject information that is information about a subject, the subject information including a plurality of first item information corresponding to each of a plurality of items; a second acquisition means that acquires candidate account information that is information about candidates for accounts to be used by the subject, the candidate account information including a plurality of second item information corresponding to each of the plurality of items extracted from the SNS information managed by the SNS system; a first score acquisition means that calculates an item-specific score indicating the degree of similarity between the first item information and the second item information for each of the plurality of items; and a second score acquisition means that calculates an identity score that indicates the possibility that a user of the candidate account is the subject, using the item-specific score calculated for each of the plurality of items. 10. An SNS information processing system comprising: an SNS system that provides an SNS to a plurality of SNS users; and the SNS information processing device described in any one of 1. to 8.. 11. An SNS information processing method in which one or more computers: acquire subject information that is information about a subject and includes a plurality of first information corresponding to each of a plurality of items; acquire candidate account information that is information about a candidate account to be used by the subject and includes a plurality of second item information corresponding to each of the plurality of items extracted from SNS information managed by the SNS system; calculate an item-specific score for each of the plurality of items indicating the degree of similarity between the first item information and the second item information; and use the item-specific score calculated for each of the plurality of items to calculate an identity score that indicates the possibility that the user of the candidate account is the subject. 12. The SNS information processing method described in 11., in which the identity score is a weighted average calculated by applying a plurality of item-specific weights corresponding to each of the plurality of items to the calculated item-specific scores.13. The SNS information processing method according to 12., wherein at least a portion of the weights for each of the plurality of items are determined based on the content of the second item information. 14. The SNS information processing method according to 13., wherein at least a portion of the weights for each of the plurality of items are determined based on the rarity of the content of the second item information. 15. The SNS information processing method according to any one of 11. to 14., wherein the SNS information includes at least one of unspecified sent information sent from the candidate account to unspecified others, specific sent information sent from the candidate account to one or more specific accounts, and received information received from other accounts. 16. The SNS information processing method according to any one of 11. to 15., wherein, when there is a plurality of second item candidate information extracted from the SNS information as candidates for the second item information, the second item information is estimated by statistically processing the plurality of second item candidate information. 17. The SNS information processing method according to any one of 11. to 16., wherein the plurality of items include at least one of a facial image, an address of a place visited, and an address of a place of residence. 18. The SNS information processing method according to 17., wherein the item-specific score for the item related to an image is a similarity between the images obtained by processing the images included in the first item information and the second item information. 19. A program causing one or more computers to execute the following: acquiring subject information that is information about a subject and includes a plurality of first item information corresponding to each of a plurality of items; acquiring candidate account information that is information about candidates for accounts to be used by the subject and includes a plurality of second item information corresponding to each of the plurality of items extracted from SNS information managed in an SNS system; calculating an item-specific score that indicates the degree of similarity between the first item information and the second item information for each of the plurality of items; and calculating an identity score that indicates the possibility that the user of the candidate account is the subject using the item-specific score calculated for each of the plurality of items. 20. A program causing one or more computers to execute the steps of 11. 21. A program for executing the SNS information processing method according to any one of 19. to 18. 22. A recording medium on which the program according to 19. or 20. is recorded.

[0203] This application claims priority based on Japanese Patent Application No. 2023-049471, filed on March 27, 2023, the disclosure of which is incorporated herein by reference in its entirety.

[0204] REFERENCE SIGNS LIST 100 SNS information processing system 101 SNS system 102, 402 SNS information processing device 111 SNS server 112 Terminal 121 Search unit 122 First acquisition unit 123, 323 Second acquisition unit 124, 424 First score acquisition unit 125, 225 Second score acquisition unit 126 Related account acquisition unit 225a Weight determination unit 225b Score calculation unit 323a Extraction unit 323b Estimation unit

Claims

1. a first acquisition means for acquiring subject information, the subject information being information about the subject and including a plurality of first item information corresponding to a plurality of items; a second acquisition means for acquiring candidate account information, which is information on candidates for accounts to be used by the target person, including a plurality of second item information corresponding to each of the plurality of items extracted from SNS information managed in the SNS system; a first score acquisition means for obtaining an item score indicating a degree of similarity between the first item information and the second item information for each of the plurality of items; and a second score acquisition means for acquiring an identity score indicating the possibility that the user of the candidate account is the target person, using the item-specific scores acquired for each of the plurality of items. SNS information processing device.

2. The identity score is a weighted average calculated by applying a plurality of item-specific weights corresponding to the plurality of items to the calculated plurality of item-specific scores. The SNS information processing device according to claim 1 .

3. At least some of the weights for each of the plurality of items are determined based on the content of the second item information. The SNS information processing device according to claim 2 .

4. At least a part of the weights for each of the plurality of items is determined based on the rarity of the content of the second item information. The SNS information processing device according to claim 3 .

5. The SNS information includes at least one of unspecified transmitted information transmitted from the candidate account to unspecified other people, specific transmitted information transmitted from the candidate account to one or more specific accounts, and received information received from other accounts. The SNS information processing device according to claim 2 .

6. When there are a plurality of pieces of second item candidate information extracted from the SNS information as candidates for the second item information, the second item information is estimated by statistically processing the plurality of pieces of second item candidate information. The SNS information processing device according to claim 5 .

7. The plurality of items include at least one of a face image, an address of a place visited, and an address of a place of residence. The SNS information processing device according to claim 2 .

8. The item score for the item related to the image is a similarity between the images obtained by processing the images included in the first item information and the second item information. The SNS information processing device according to claim 7 .

9. One or more computers acquiring subject information that is information about the subject and includes a plurality of first item information corresponding to a plurality of items, Acquire candidate account information, which is information about candidates for accounts to be used by the subject person, and includes a plurality of second item information corresponding to each of the plurality of items extracted from SNS information managed in the SNS system; calculating an item score indicating a degree of similarity between the first item information and the second item information for each of the plurality of items; Using the item-specific scores calculated for each of the plurality of items, an identity score is calculated that indicates the possibility that the user of the candidate account is the target person. SNS information processing method.

10. On one or more computers, acquiring subject information that is information about the subject and includes a plurality of first item information corresponding to a plurality of items, Acquire candidate account information, which is information about candidates for accounts to be used by the subject person, and includes a plurality of second item information corresponding to each of the plurality of items extracted from SNS information managed in the SNS system; calculating an item score indicating a degree of similarity between the first item information and the second item information for each of the plurality of items; A program for executing the process of calculating an identity score indicating the possibility that the user of the candidate account is the target person, using the item-specific scores calculated for each of the multiple items.