JP2017033250A - Information processing device, information processing method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing device, an information processing method and a program which match a plurality of terminals likely to be used by an identical user with each other.SOLUTION: An information processing device comprises: an acquisition section and a matching section. The acquisition section acquires plural pieces of log data which associates IP addresses, terminal IDs and time stamps. The matching section: identifies a candidate set including a plurality of terminals using the identical IP address on the basis of the plural pieces of log data; estimates a probability that an identical user uses the plurality of terminals included in the candidate set on the basis of the log data with respect to the IP address; and matches the plurality of terminals with each other on the basis of an estimation result.SELECTED DRAWING: Figure 1

Description

  Embodiments described herein relate generally to an information processing apparatus, an information processing method, and a program for matching a plurality of devices (hereinafter referred to as cross devices) used by a single user via a network.

  For example, an advertiser considers controlling information to a cross device by cross device matching. An information processing apparatus such as a server realizes cross-device matching based on log data that a user logs into a site or the like using the same ID or the like at each terminal.

  However, according to deterministic cross-device matching based on such an ID, there is a problem that it is difficult to match a device of a user who does not log in or a user who does not hold a login ID. Furthermore, according to such definite cross-device matching, there is a problem that a volume related to the cross device cannot be produced because a member who logs in with an ID or the like is an object.

JP 2012-133780 A

  In order to solve the above problems, an information processing apparatus, an information processing method, and a program that match a plurality of terminals that are likely to be used by the same user are provided.

  According to the embodiment, the information processing apparatus includes an acquisition unit and a matching unit. The acquisition unit acquires a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated with each other. The matching unit identifies a candidate set including a plurality of terminals using the same IP address based on the plurality of log data, and the same user of the plurality of terminals included in the candidate set based on the log data regarding the IP address Is used, and a plurality of terminals are matched based on the estimation result.

  According to the embodiment, the matching server can specify a plurality of terminals that have browsed advertisements or the like using the same IP address at the same time from the log data regarding the IP address. The matching server can estimate the plurality of identified terminals as being used by the same user and match the plurality of identified terminals.

FIG. 1 is a block diagram illustrating a configuration example of a service providing system according to an embodiment. FIG. 2 is a block diagram illustrating a configuration example of the matching server according to the embodiment. FIG. 3 is a diagram for explaining an operation example of matching estimation according to the embodiment. FIG. 4 is a flowchart illustrating an operation example of the matching server according to the embodiment.

Hereinafter, embodiments will be described with reference to the drawings.
FIG. 1 is a diagram illustrating a configuration example of a service providing system 10.
The service providing system 10 provides a predetermined service to the user terminal 4 or 5. For example, the service providing system 10 distributes data related to a service provided on an application such as a browser, a game application, a news application, or an SNS application executed on the user terminal 4 or 5.

  As shown in FIG. 1, the service providing system 10 includes a matching server 1, a server 2, a router 3, user terminals 4 and 5, and the like. The service providing system 10 connects the matching server 1, the server 2, the router 3, and the user terminals 4 and 5 to each other through a communication network 6 such as the Internet.

  The matching server 1 matches user terminals 4 or 5 used by the same user via a network. That is, the matching server 1 determines which terminal belongs to the use of the same user. The configuration or operation example of the matching server 1 will be described later.

  The server 2 provides a predetermined service to the user terminal 4 or 5 through the communication network 6. The server 2 also generates log data indicating that the user terminal 4 or 5 has accessed the server 2. The log data will be described later. For example, the server 2 generates log data based on access from the user terminal 4 or 5. The server 2 stores the generated log data.

  Here, it is assumed that the server 2 distributes the advertisement to the user terminal 4 or 5 through the communication network 6. The server 2 stores an advertisement provided by the advertiser in advance. The server 2 receives an advertisement request for requesting advertisement delivery from the user terminal 4 or 5. The server 2 transmits the advertisement to the user terminal 4 or 5 as a response to the advertisement request. Further, the server 2 may transmit a campaign ID for specifying an advertisement as an advertisement response to the user terminal 4 or 5.

  The advertisement distributed by the server 2 is configured by, for example, html, text, an image, or a combination thereof. The format and content of the advertisement distributed by the server 2 is not limited to a specific configuration. The server 2 may store a plurality of advertisements. The server 2 may update the advertisement as appropriate.

  The server 2 may provide e-commerce or a web server that provides a website. The server 2 may transmit predetermined data in response to a request from a smartphone application such as a game. The service provided by the server 2 is not limited to a specific configuration.

  The router 3 forms a local network. The router 3 connects a terminal (for example, the user terminal 4 or 5) in the local network to the communication network 6. That is, the router 3 relays data transmission / reception between the user terminal 4 or 5 and the communication network 6.

  The router 3 is assigned an IP address. The router 3 transmits data from a terminal (for example, the user terminal 4 or 5) in the local network formed by the router 3 to the communication network 6 as data from the assigned IP address. That is, in the communication network 6, access from a terminal (for example, the user terminal 4 or 5) in the local network formed by the router 3 is treated as access from an IP address assigned to the router 3. That is, access from a terminal (for example, user terminal 4 or 5) in the local network formed by the router 3 is treated as access from the same IP address.

  The user terminal 4 or 5 is a terminal possessed by the user. The user terminal 4 or 5 executes an application such as a browser based on a user operation (input). The user terminal 4 or 5 displays an advertisement distributed by the server 2 according to the operation of the application.

  The user terminal 4 or 5 includes a storage area 4a or 4b for storing a terminal ID, respectively. The storage area 4a or 4b is formed in, for example, a nonvolatile memory stored in the user terminal 4 or 5.

  The terminal ID is an ID for identifying the user terminal 4 or 5. The terminal ID may be stored when the user terminal 4 or 5 is manufactured. The terminal ID may be an ID generated as appropriate.

  For example, the terminal ID may be IDFA (Identification For Advisors) or Advertising ID. Further, the terminal ID may be an ID set in Cookie or the like. The terminal ID may be an ID (for example, an advertisement ID) issued by the server 2 for each terminal. Further, the terminal ID may be an ID generated by a script in the web site.

  The user terminal 4 or 5 may be, for example, a desktop PC, a notebook PC, a tablet PC, a smartphone, or a wearable terminal. The configuration of the user terminal 4 or 5 is not limited to a specific configuration.

  The communication network 6 connects the matching server 1, the server 2, and the router 3 so that they can communicate with each other. The communication network 6 is, for example, the Internet. The matching server 1, the server 2, and the router 3 access the communication network 6 by wire or wireless.

Next, the matching server 1 will be described.
FIG. 2 is a block diagram illustrating a configuration example of the matching server 1.

  In the configuration example illustrated in FIG. 2, the matching server 1 (information processing apparatus) includes a CPU 11, a ROM 12, a RAM 13, an NVM 14, a communication unit 15, an operation unit 16, a display unit 17, and the like as a basic configuration. These units are connected to each other via a data bus. The matching server 1 may include a configuration as necessary in addition to the configuration shown in FIG. 2 or may exclude a specific configuration.

  The CPU 11 has a function of controlling the overall operation of the matching server 1. The CPU 11 may include an internal cache and various interfaces. The CPU 11 implements various processes by executing a program (code) stored in advance in the internal memory, the ROM 12 or the NVM 14. The CPU 11 is, for example, a processor.

  Note that some of the various functions realized by the CPU 11 executing the program may be realized by a hardware circuit. In this case, the CPU 11 controls functions executed by the hardware circuit.

  The ROM 12 is a non-volatile memory in which a control program and control data are stored in advance. The control program and control data stored in the ROM 12 are incorporated in advance according to the specifications of the matching server 1. The ROM 12 stores, for example, a program (for example, BIOS) that controls the circuit board of the matching server 1.

  The RAM 13 is a volatile memory. The RAM 13 temporarily stores data being processed by the CPU 11. The RAM 13 stores various application programs based on instructions from the CPU 11. The RAM 13 may store data necessary for executing the application program, an execution result of the application program, and the like.

  The NVM 14 is a nonvolatile memory capable of writing and rewriting data. The NVM 14 is configured by, for example, a hard disk, SSD, EEPROM (registered trademark), flash memory, or the like. The NVM 14 stores a control program, an application, and various data according to the operation usage of the matching server 1.

  The NVM 14 includes a storage area 14a for storing log data. The log data will be described later.

  The communication unit 15 is an interface for transmitting / receiving data to / from the server 2 or the router 3 through the communication network 6. The communication unit 15 is an interface that supports LAN connection, for example.

  Various operation instructions are input to the operation unit 16 by the operator of the matching server 1. The operation unit 16 transmits operation instruction data input to the operator to the CPU 11. The operation unit 16 is, for example, a keyboard, a numeric keypad, and a touch panel.

  The display unit 17 is a display device that displays various information under the control of the CPU 11. The display unit 17 is, for example, a liquid crystal monitor. When the operation unit 16 is configured with a touch panel or the like, the display unit 17 may be formed integrally with the operation unit 16.

  Note that the matching server 1 and the server 2 may be integrated as one server.

Next, log data will be described.
The log data is a log of the user terminal 4 or 5 accessing the communication network 6 (for example, browsing an advertisement). The log data is stored in association with an IP address, a terminal ID, and a time stamp.

  The IP address is an IP address assigned to the user terminal 4 or 5. For example, the IP address is an IP address assigned to the router 3 forming the local network to which the user terminal 4 or 5 belongs.

The terminal ID indicates a terminal that has accessed the communication network 6.
The time stamp indicates the date and time when the user terminal 4 or 5 accesses the communication network 6.

  Here, the log data relates to advertisement distribution. That is, log data is generated when an advertisement is distributed to the user terminal 4 or 5.

  The log data may be generated from advertisement distribution, web browsing, e-commerce, or access related to a smartphone application. The log data may relate to any service as long as it has the above configuration.

  Note that the log data may include OS information indicating the OS operating on the user terminal 4 or 5 or the version of the OS.

  For example, the matching server 1 acquires log data from the server 2 and stores it in the storage area 14a. In addition, the matching server 1 may acquire log data from another interface (for example, an interface that supports USB) in accordance with the operation of the operator.

Next, functions realized by the CPU 11 will be described.
The CPU 11 has a function of acquiring a plurality of log data (acquisition unit). For example, the CPU 11 acquires log data from the storage area 14a. Further, the CPU 11 may acquire log data from an external device.

  The CPU 11 has a function of classifying IP addresses included in a plurality of acquired log data into clusters (classification unit). For example, the CPU 11 classifies IP addresses into clusters based on the characteristics of IP addresses. For example, the CPU 11 clusters the IP address based on whether it has the characteristics of an IP address used in a home router, an IP address used in a corporate router, or an IP address used in a mobile carrier. Classify into:

  For example, the CPU 11 determines the feature of the IP address based on the log data (feature determination unit). For example, the CPU 11 determines the characteristics of the IP address based on a time stamp corresponding to the IP address. For example, the CPU 11 determines characteristics of frequency (usage frequency) used for each time zone and each day of the week. For example, the CPU 11 generates a waveform indicating the usage frequency for each time zone and each day of the week, and determines the characteristics of the usage frequency based on the waveform.

For example, the CPU 11 determines the characteristics of the usage frequency as follows.
The CPU 11 has a low usage frequency on weekdays and holidays, and there is no significant difference in usage frequency between weekdays and holidays. It is determined that the characteristics of the IP address to be used are provided (model A).

  In addition, when the CPU 11 is frequently used on both weekdays and holidays, the holidays are more frequently used than weekdays, and the usage frequency is high around 12:00 on weekdays, the frequency of use of the IP address is used by the mobile carrier. It is determined that it has the characteristics of the IP address (model B).

  Further, the CPU 11 determines that the use frequency of the IP address has characteristics of the IP address used for the home router when the use frequency is high on weekday nights and the use frequency is higher on holidays than on weekdays. (Model C).

  Further, the CPU 11 determines that the use frequency of the IP address has characteristics of the IP address used for the enterprise router when the use frequency is very high during the weekday and the use frequency of the holiday is low (model D). ).

  The CPU 11 compares the model waveform corresponding to the models A to D with the waveform indicating the usage frequency of the IP address, and calculates a statistical distance from each model waveform. The CPU 11 determines the characteristics of the usage frequency based on the statistical distance.

  Note that the CPU 11 may determine the feature of the usage frequency using a model other than the models A to D. The model in which the CPU 11 determines the characteristics of the usage frequency is not limited to a specific model.

In addition, the CPU 11 determines the feature of the number of terminals using the IP address (density) as the feature of the IP address.
For example, when the number of terminals that use IP addresses exceeds 50, the CPU 11 determines that the IP address density has the characteristics of IP addresses used for carrier lines.

  Further, when the number of terminals using the IP address is more than 20 and less than 50, the CPU 11 determines that the IP address density has the characteristics of the IP address used for the corporate router.

  Further, when the number of terminals using the IP address is less than 10, the CPU 11 determines that the IP address density has the characteristics of the IP address used for the home router.

  Further, the CPU 11 determines the feature of the IP address based on how much the IP terminal is continuously used for the same terminal. That is, the CPU 11 uses the ratio of terminals used in both the last predetermined period (for example, the last 10 days) and the previous predetermined period (for example, 10 days from 20 days to 11 days ago) (continuation) Usage characteristics are determined.

  For example, when the ratio of terminals used in both periods is 5% or less, the CPU 11 determines that the continuous usage rate of the IP address has characteristics of the IP address used for the carrier line.

  In addition, when the ratio of terminals used in both periods is 30% or more and 50% or less, the CPU 11 determines that the continuous use rate of the IP address has the characteristics of the IP address used in the corporate router.

  In addition, when the ratio of terminals used in both periods is 70% or more, the CPU 11 determines that the continuous use rate of the IP address has characteristics of the IP address used in the home router.

  Further, the CPU 11 determines the feature of the number of log data per terminal (activity) as the feature of the IP address.

  For example, when the number of log data per terminal is 40 or less (that is, when the usage per user is low), the CPU 11 is characterized by the IP address activity amount used on the carrier line. It is determined that it is provided.

  When the number of log data per terminal is 41 or more and 70 or less, the CPU 11 determines that the activity amount of the IP address has the characteristics of the IP address used in the company router.

  When the number of log data items per terminal is 100 or more (that is, when the usage per user is large), the CPU 11 determines the IP address feature used by the home router. It is determined that it is prepared.

  The CPU 11 classifies IP addresses into clusters based on the above characteristics. Based on the features of the IP address, the CPU 11 classifies the IP addresses into clusters so that the features of the IP address are maximized by various statistical methods or machine learning. Note that the thresholds for the CPU 11 to determine the characteristics of the IP address are not limited to the above values. CPU11 may change a threshold suitably from a statistical viewpoint.

  Further, the CPU 11 determines the intended use of the IP address based on the feature of the IP address (use determining unit). For example, the CPU 11 adds the likelihoods of usages calculated from the features of the IP address, and determines the usages corresponding to the likelihood exceeding a predetermined threshold as the usages of the IP address. Further, the CPU 11 may calculate a certainty factor indicating the certainty of the intended use for each IP address.

  For example, the CPU 11 has a large amount of activity with respect to a certain IP address, and has a high density and features of an IP address used for a corporate router. However, the frequency of use is high on weekdays and holidays are higher than weekdays. If the characteristics of the IP address used for the home router also appear, classify it as an IP address used for the home router but calculate a low certainty (for example, classification = IP address used for the home router, Confidence = 0.55).

  Further, when the use of a certain IP address is concentrated on weekdays and the congestion is slightly high and the continuous usage rate is high, the CPU 11 classifies the IP address as an IP address used for a corporate router and calculates a high certainty ( For example, classification = IP address used for enterprise volume router, certainty = 0.98).

Note that the CPU 11 may perform fixed point observation of the trend or likelihood of each feature and automatically correct the classification method.
The method by which the CPU 11 classifies IP addresses into clusters is not limited to a specific method.

  Further, the CPU 11 has a function of selecting a cluster (first cluster) used for cross device matching (selection unit). For example, the CPU 11 selects an IP address that has a low density and a high continuous usage rate and a large amount of activity, and is used intensively in a predetermined time zone (for example, a weekday night or a holiday).

  The CPU 11 has a function of acquiring log data of the IP address of the selected cluster. For example, the CPU 11 acquires log data from the storage area 14a.

The CPU 11 has a function of estimating matching based on the acquired log data (matching unit).
When a plurality of terminal IDs frequently appear in log data with the same IP address in the same time zone, the CPU 11 estimates that the plurality of terminal IDs are terminal IDs of a plurality of terminals possessed by the same user.

FIG. 3 shows an operation example of matching estimation.
In the example illustrated in FIG. 3, the terminal A indicated by the terminal ID 123 and the terminal B indicated by the terminal ID 456 access the communication network 6 using the IP address 11: 22: 33: 44.

  When the same user uses terminal A and terminal B, it is highly likely that terminal A and terminal B are under the same network at the same time. Therefore, when the terminal A and the terminal B are under the same network at the same time (that is, accessed from the same IP address), the terminal A and the terminal B are likely to belong to the use of the same user.

  The CPU 11 estimates that combinations of terminals using the same IP address at the same time from the log data belong to the use of the same user, and based on the estimation result (based on the estimation result that the same user is highly likely to be used). Match the terminal.

  For example, the CPU 11 specifies a candidate set including two terminals using the same IP address based on the log data. The CPU 11 calculates a score indicating the possibility that the same user is using the candidate set.

  For example, the CPU 11 adds the score of the candidate set when both terminals of the candidate set use the same IP address within a predetermined time. For example, the CPU 11 adds a score according to an interval from when a terminal with a candidate set uses an IP address until another terminal uses the same IP address. For example, the CPU 11 adds 1.0 points when both terminals in the candidate set use the same IP address within one hour. Moreover, CPU11 adds 0.1 points | pieces, when it is within 8 hours exceeding 1 hour.

  Moreover, CPU11 adds the score of a candidate set, when both terminals of a candidate set are used in the same time slot | zone. For example, the CPU 11 divides the time zone (for example, Monday to Thursday PM22 to next morning AM7, weekday AM9 to PM17, and Friday PM22 to Monday AM7). CPU11 adds a score, when both terminals of a candidate set are used for the same time division. In addition, CPU11 may adjust a score according to the ratio etc. for which both terminals of a candidate set are used for the same time division.

  Moreover, CPU11 adds the score of a candidate set based on the time (duplicate use time) in which both terminals of a candidate set are used overlappingly. For example, the CPU 11 specifies the activation periods of both terminals based on log data including the terminal IDs of both terminals in the candidate set. For example, when the date and time indicated by the plurality of log data of the terminal are within a predetermined time, the CPU 11 specifies the period between the date and time indicated by the log data as the activation period of the terminal.

CPU11 specifies the length of the period when the starting period for every terminal overlaps with each other as overlapped use time.
For example, the CPU 11 adds the score of the candidate set when the overlapping usage time is longer than a predetermined threshold. Moreover, CPU11 may add a score according to duplication usage time. For example, the CPU 11 adds a larger score as the overlapping usage time increases. Further, the CPU 11 may add a score when the overlapping use time is larger than a predetermined threshold and both terminals do not use another IP address at a time other than the overlapping use time. For example, the CPU 11 adds a score when both terminals are activated one hour at a time, overlapping for 30 minutes and not using another IP address for 30 minutes that do not overlap.

  Further, when there is log data of both terminals of the candidate set at each IP address for different usage, the CPU 11 adds the scores of the candidate set. For example, the CPU 11 may add the score of the candidate set when there is an overlapping usage time at each IP address for different usage.

  Based on the score of the candidate set, the CPU 11 matches the two terminals included in the candidate set as terminals belonging to the use of the same user. For example, when the score of the candidate set exceeds a predetermined threshold, the CPU 11 matches two terminals included in the candidate set as terminals belonging to the use of the same user. That is, the CPU 11 specifies a candidate set used by the same user from the candidate set. The CPU 11 associates the matched terminals with each other and stores them in the NVM 14 or the like. Further, when the terminal A and the terminal B are matched and the terminal B and the terminal C are matched, the CPU 11 may match the terminals A to C as belonging to the use of the same person.

  The CPU 11 may generate a candidate set including three or more terminals and calculate the score.

Next, an operation example of each part of the service providing system 10 will be described.
First, an operation example of the user terminal 4 or 5 will be described. Here, an operation example of the user terminal 4 will be described as an example. The user terminal 5 performs the same operation.

  The user terminal 4 starts an application based on an operation from the user. Here, the application displays an advertisement provided from the server 2.

  When the application is activated, the user terminal 4 transmits an advertisement request including the terminal ID to the server 2 in accordance with the operation of the application. The advertisement request may include OS information indicating the OS or OS version running on the user terminal 4.

  The user terminal 4 receives an advertisement from the server 2 as a response to the advertisement request. When receiving the advertisement, the user terminal 4 displays the advertisement on a display unit or the like. When the advertisement is displayed, the user terminal 4 ends the operation.

  When the advertisement ID is a terminal ID, the user terminal 4 determines whether the advertisement ID is stored. If it is determined that the advertisement ID is not stored, the user terminal 4 transmits an advertisement request to the server 2 and stores the advertisement ID included in the response as the terminal ID. When it is determined that the advertisement ID is stored (for example, when the server 2 is accessed for the second time or later), the user terminal 4 transmits an advertisement request including the stored advertisement ID to the server 2.

Next, an operation example of the server 2 will be described.
The server 2 receives an advertisement request from the user terminal 4 or 5. When receiving the advertisement request, the server 2 transmits an advertisement corresponding to the advertisement request to the user terminal 4 or 5 as a response.

  Further, the server 2 generates log data. For example, the server 2 extracts the terminal ID from the advertisement request. Further, the server 2 acquires the IP address of the advertisement request source. Further, the server 2 generates a time stamp indicating the date and time when the advertisement request is received or the date and time when the advertisement is transmitted. The server 2 generates log data by associating the terminal ID, the IP address, and the time stamp. If the advertisement request includes OS information, the server 2 further stores the OS information in association with the log data.

  The server 2 may generate the log data after transmitting the advertisement to the user terminal 4 or 5, or may transmit the advertisement to the user terminal 4 or 5 after generating the log data.

  The server 2 stores the generated log data in a nonvolatile memory. Note that the server 2 may transmit log data to the matching server 1 as needed. Further, the server 2 may transmit log data to the matching server 1 at a predetermined interval.

Next, an operation example of the matching server 1 will be described.
FIG. 4 is a flowchart for explaining an operation example of the matching server 1.

  Here, it is assumed that the storage area 14a of the matching server 1 stores a plurality of log data.

First, the CPU 11 of the matching server 1 classifies the IP addresses included in the log data into clusters based on the log data stored in the storage area 14a (S11).
When the IP address is classified into clusters, the CPU 11 selects a cluster to be used for matching estimation from the classified clusters (S12).

When a cluster is selected, the CPU 11 acquires log data corresponding to the IP address of the selected cluster (S13).
When the log data is acquired, the CPU 11 estimates matching (S14).

  When the matching is estimated, the CPU 11 ends the operation.

  For example, the CPU 11 transmits two matched terminal IDs to the server 2.

When the server 2 receives the matched two terminal IDs, the server 2 stores them in a nonvolatile memory or the like.
The server 2 can provide related advertisements to the terminals indicated by the two matched terminal IDs.

Hereinafter, an operation example in which the server 2 distributes advertisements related to two matched terminals will be described. Here, it is assumed that terminal A and terminal B are matched.
First, the server 2 delivers a predetermined advertisement in response to an advertisement request from the terminal A.
When a predetermined advertisement is distributed to the terminal A, the server 2 stands by until an advertisement request from the terminal B is received.

  When the advertisement request is received from the terminal B, the server 2 extracts the terminal ID of the terminal B from the advertisement request. When the terminal ID of the terminal B is extracted, the server 2 searches for a terminal ID that matches the extracted terminal ID. Here, the server 2 finds the terminal ID of the terminal A. Note that the server 2 may acquire a terminal matched with the terminal B from the matching server 1.

  When the terminal ID of the terminal A is found, the server 2 acquires an advertisement related to the advertisement distributed to the terminal A. For example, when the terminal A is a smartphone and the terminal B is a PC, the server 2 acquires an advertisement for PC corresponding to the advertisement for smartphone distributed to the terminal A. Note that when both the terminal A and the terminal B are PCs (or smartphones), the server 2 may acquire the same advertisement as the advertisement distributed to the terminal A.

  When the advertisement is acquired, the server 2 delivers the acquired advertisement to the terminal B as a response to the advertisement request. When the advertisement is distributed, the server 2 ends the operation.

  The matching server configured as described above can classify IP addresses. In addition, the matching server can estimate the usage of the IP address. Therefore, the matching server can estimate the state in which the user uses the terminal according to the usage purpose of the IP address. For example, when the IP address used by the terminal is an IP address used for a home router, the matching server can recognize that the user of the terminal is using the terminal at home. The matching server can recognize that the user of the terminal is using the terminal during a company break when the IP address used by the terminal is an IP address used for a corporate router. . In addition, when the IP address used by the terminal is an IP address used for the mobile carrier, the matching server can recognize that the user of the terminal is using the terminal while moving on a train or the like.

  For example, in a service for a smart device (for example, a game application), the service provider can analyze whether the usage time (for example, play time) of the service changes for each user situation.

  The matching server can match a plurality of terminals belonging to the use of the same user. As a result, the advertisement server can distribute advertisements related to different terminals used by the same user based on the matching result. For example, the advertisement server can distribute a related advertisement to a PC of a smartphone user after distributing the advertisement to the smartphone. As a result, the advertisement server can effectively distribute the advertisement and prompt the user to convert effectively.

  In addition, a management server that provides a predetermined service (such as an EC site) can perform personalization (for example, design for each customer or presentation of a product) across terminals using the matching result. For example, even if the management server is not logged in with a common ID, personalization similar to personalization performed for service use on a PC can be applied to service use on a smartphone. As a result, the management server can improve the user experience.

  Further, the matching server matches a plurality of terminals based on the time stamp. When a plurality of terminals use the same address within a predetermined time, the matching server determines that there is a high possibility that the plurality of terminals belong to the use of the same user. As a result, the matching server can adjust the balance between the matching accuracy and the volume by adjusting the length of the time.

  Moreover, a matching server determines with a high possibility that a some terminal belongs to use of the same user, when a some terminal is used in the same time slot | zone. As a result, the matching server can improve the accuracy of matching when the time period during which the user uses a plurality of terminals is fixed.

  Further, the matching server determines that there is a high possibility that a plurality of terminals belong to the use of the same user when the overlapping use time is long. As a result, the matching server can improve matching accuracy when the user uses a plurality of terminals simultaneously.

  Further, the matching server determines that there is a high possibility that a plurality of terminals belong to the use of the same user when there is log data of a plurality of terminals at each IP address for different usage purposes. As a result, the matching server can match a plurality of terminals used in homes and businesses. Therefore, the matching server can prevent matching with a terminal used by the user's family.

  In addition, this invention is not limited to the said embodiment, In the implementation stage, it can change variously in the range which does not deviate from the summary. In addition, the embodiments may be appropriately combined as much as possible, and in that case, the combined effect can be obtained. Further, the above embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, the problem described in the column of the problem to be solved by the invention can be solved, and the effect described in the column of the effect of the invention Can be obtained as an invention.

  DESCRIPTION OF SYMBOLS 1 ... Matching server, 2 ... Server, 3 ... Router, 4 and 5 ... User terminal, 4a and 5a ... Storage area, 6 ... Communication network, 11 ... CPU, 14 ... NVM, 14a ... Storage area, 15 ... Communication part.

In addition, this invention is not limited to the said embodiment, In the implementation stage, it can change variously in the range which does not deviate from the summary. Further, the embodiments may be combined as appropriate as possible, and in that case, the combined effect can be obtained. Further, the above embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, the problem described in the column of the problem to be solved by the invention can be solved, and the effect described in the column of the effect of the invention Can be obtained as an invention.
The invention described in the scope of the claims at the beginning of the present application is added below.
[C1]
An acquisition unit that acquires a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated;
A candidate set including a plurality of terminals using the same IP address is identified based on the plurality of log data, and used by the same user of the plurality of terminals included in the candidate set based on the log data regarding the IP address. A matching unit that estimates the possibility of matching, and matches a plurality of terminals based on the estimation result,
An information processing apparatus comprising:
[C2]
The matching unit calculates a score indicating the possibility that the same user is using the candidate set based on the log data, and matches a plurality of terminals based on the score.
The information processing apparatus according to C1.
[C3]
The matching unit adds scores of the candidate set when a plurality of terminals of the candidate set use the same IP address within a predetermined time.
The information processing apparatus according to C2.
[C4]
The matching unit adds scores of the candidate set when a plurality of terminals of the candidate set are used in the same time zone;
The information processing apparatus according to C2 or 3.
[C5]
The matching unit adds the score of the candidate set based on a time when a plurality of terminals of the candidate set are used redundantly.
The information processing apparatus according to any one of C2 to C4.
[C6]
A classifying unit that classifies IP addresses included in the plurality of log data into a plurality of clusters based on characteristics of the IP addresses;
A selection unit for selecting a first cluster from the plurality of clusters;
Further comprising
The matching unit identifies a candidate set including a plurality of terminals using the same IP address based on log data regarding the IP address of the first cluster selected by the selection unit;
The information processing apparatus according to any one of C1 to C5.
[C7]
The classification unit classifies IP addresses into clusters based on at least one of IP address usage frequency, congestion, continuous usage rate, or activity amount.
The information processing apparatus according to C6.
[C8]
The classification unit determines the intended use of the IP address,
The matching unit adds the scores of the candidate set when there is log data of a plurality of terminals of the candidate set at different IP addresses for usage.
The information processing apparatus according to C6 or 7.
[C9]
Acquire a plurality of log data associating IP addresses, terminal IDs, and time stamps,
Identifying a candidate set including a plurality of terminals using the same IP address based on the plurality of log data;
Estimating the possibility that the same user of a plurality of terminals included in the candidate set is used based on the log data relating to the IP address, and matching the plurality of terminals based on the estimation result;
Information processing method.
[C10]
A code for causing a processor to acquire a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated;
A code for causing the processor to identify a candidate set including a plurality of terminals using the same IP address based on the plurality of log data;
The processor estimates the possibility that the same user of a plurality of terminals included in the candidate set is used based on the log data related to the IP address, and matches a plurality of terminals based on the estimation result;
A program comprising
[C11]
An acquisition unit that acquires a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated;
A feature determination unit that determines at least one feature of the use frequency, congestion, continuous usage rate, or activity amount of the IP address based on the plurality of log data;
A use determining unit that determines a use of the IP address based on the characteristics;
An information processing apparatus comprising:
[C12]
Acquire a plurality of log data associating IP addresses, terminal IDs, and time stamps,
Determining at least one characteristic of the usage frequency, congestion, continuous usage rate or activity amount of the IP address based on the plurality of log data;
Based on the characteristics, the usage of the IP address is determined.
Information processing method.
[C13]
A code for causing a processor to acquire a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated;
Code for causing the processor to determine at least one characteristic of the usage frequency, congestion, continuous usage rate, or activity amount of the IP address based on the plurality of log data;
A code for causing the processor to determine the intended use of the IP address based on the characteristics;
A program comprising

Claims (13)

An acquisition unit that acquires a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated;
A candidate set including a plurality of terminals using the same IP address is identified based on the plurality of log data, and used by the same user of the plurality of terminals included in the candidate set based on the log data regarding the IP address. A matching unit that estimates the possibility of matching, and matches a plurality of terminals based on the estimation result,
An information processing apparatus comprising: The matching unit calculates a score indicating the possibility that the same user is using the candidate set based on the log data, and matches a plurality of terminals based on the score.
The information processing apparatus according to claim 1. The matching unit adds scores of the candidate set when a plurality of terminals of the candidate set use the same IP address within a predetermined time.
The information processing apparatus according to claim 2. The matching unit adds scores of the candidate set when a plurality of terminals of the candidate set are used in the same time zone;
The information processing apparatus according to claim 2 or 3. The matching unit adds the score of the candidate set based on a time when a plurality of terminals of the candidate set are used redundantly.
The information processing apparatus according to any one of claims 2 to 4. A classifying unit that classifies IP addresses included in the plurality of log data into a plurality of clusters based on characteristics of the IP addresses;
A selection unit for selecting a first cluster from the plurality of clusters;
Further comprising
The matching unit identifies a candidate set including a plurality of terminals using the same IP address based on log data regarding the IP address of the first cluster selected by the selection unit;
The information processing apparatus according to any one of claims 1 to 5. The classification unit classifies IP addresses into clusters based on at least one of IP address usage frequency, congestion, continuous usage rate, or activity amount.
The information processing apparatus according to claim 6. The classification unit determines the intended use of the IP address,
The matching unit adds the scores of the candidate set when there is log data of a plurality of terminals of the candidate set at different IP addresses for usage.
The information processing apparatus according to claim 6 or 7. Acquire a plurality of log data associating IP addresses, terminal IDs, and time stamps,
Identifying a candidate set including a plurality of terminals using the same IP address based on the plurality of log data;
Estimating the possibility that the same user of a plurality of terminals included in the candidate set is used based on the log data relating to the IP address, and matching the plurality of terminals based on the estimation result;
Information processing method. A code for causing a processor to acquire a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated;
A code for causing the processor to identify a candidate set including a plurality of terminals using the same IP address based on the plurality of log data;
The processor estimates the possibility that the same user of a plurality of terminals included in the candidate set is used based on the log data related to the IP address, and matches a plurality of terminals based on the estimation result;
A program comprising An acquisition unit that acquires a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated;
A feature determination unit that determines at least one feature of the use frequency, congestion, continuous usage rate, or activity amount of the IP address based on the plurality of log data;
A use determining unit that determines a use of the IP address based on the characteristics;
An information processing apparatus comprising: Acquire a plurality of log data associating IP addresses, terminal IDs, and time stamps,
Determining at least one characteristic of the usage frequency, congestion, continuous usage rate or activity amount of the IP address based on the plurality of log data;
Based on the characteristics, the usage of the IP address is determined.
Information processing method. A code for causing a processor to acquire a plurality of log data in which an IP address, a terminal ID, and a time stamp are associated;
Code for causing the processor to determine at least one characteristic of the usage frequency, congestion, continuous usage rate, or activity amount of the IP address based on the plurality of log data;
A code for causing the processor to determine the intended use of the IP address based on the characteristics;
A program comprising