Data processing device, data processing method and data processing program

Abstract

A data processing device (10) comprises: a detection unit (11) that detects a change point of first data series having a first identifier and a change point of second data series having a second identifier; a calculation unit (12) that calculates an occurrence probability of the simultaneous occurrence of the change point of the first data series and the change point of the second data series; an association unit (13) that associates the first and second identifiers with each other as identifiers relating to a tracing target, on the basis of the occurrence probability; and a determination unit (14) that determines the type of a to-be-identified object having the associated first or second identifier, on the basis of the occurrence probability.

Description

[Technical field]

[0001] The present invention relates to a data processing device, a data processing method and a non-transitory computer readable medium. [Background technology]

[0002] Various methods have been proposed to acquire and trace a user's location. For example, there is a method in which radio waves emitted by a wireless terminal are observed by multiple sensors, and the user's location is estimated from the radio wave strength at the position of each sensor. There is also a method in which a camera is used to identify the face of a person captured in an image, and the movement history of that person is acquired by multiple cameras. A method that combines these multiple techniques has also been proposed.

[0003] Combining multiple methods is expected to have the effect of improving location accuracy and coverage, but it also requires matching data acquired by different types of sensors. For example, it becomes necessary to match a person whose face is identified using a camera with the identifier of a wireless terminal observed by a wireless sensor (e.g., the MAC address of a wireless LAN terminal).

[0004] As a related technique, for example, in a system for detecting the position of a human being, data observed by a plurality of types of sensors is associated based on the order of appearance time of the data from the plurality of types of sensors in Patent Document 1. Also, in an object recognition system, in Patent Document 2, a mobile terminal and a mobile object are associated based on a comparison between behavior data that detects the behavior of the mobile terminal and estimated history data of the mobile object that possesses the mobile terminal.

[0005] In addition, in Patent Document 3, a moving object tracking device determines the mutual approach state between moving objects based on the overlap of elliptical areas calculated as the destinations of each moving object. In Patent Document 4, an event log tracing device specifies the relationship between log data. In Patent Document 5, traffic change points are detected to analyze time series data of wireless communication services. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2003-111130 A [Patent Document 2] JP 2016-212675 A [Patent Document 3] JP 2011-180684 A [Patent Document 4] International Publication No. 2016 / 129093 [Patent Document 5] JP 2017-138708 A Summary of the Invention [Problem to be solved by the invention]

[0007] However, in the related technology, there is a risk that the tracing accuracy may decrease depending on the data associated with the tracing target and the identification target identified by the data.

[0008] In view of the above problems, an object of the present disclosure is to provide a data processing device, a data processing method, and a non-transitory computer-readable medium that are capable of improving tracing accuracy. [Means for solving the problem]

[0009] A data processing device according to the present disclosure includes a detection means for detecting a change point of a first data sequence having a first identifier and a change point of a second data sequence having a second identifier, a calculation means for calculating an occurrence probability that a change point of the first data sequence and a change point of the second data sequence occur synchronously, a correspondence means for matching the first identifier and the second identifier as identifiers related to a tracing target based on the occurrence probability, and a discrimination means for discriminating a type of identification target of the matched first identifier or second identifier based on the occurrence probability.

[0010] A data processing method according to the present disclosure detects change points in a first data sequence having a first identifier and change points in a second data sequence having a second identifier, calculates an occurrence probability that the change points in the first data sequence and the change points in the second data sequence occur synchronously, associates the first identifier and the second identifier as identifiers related to a tracing target based on the occurrence probability, and determines a type of identification target of the associated first identifier or second identifier based on the occurrence probability.

[0011] A non-transitory computer-readable medium according to the present disclosure is a non-transitory computer-readable medium having stored thereon a data processing program for causing a computer to execute a process of detecting a change point in a first data series having a first identifier and a change point in a second data series having a second identifier, calculating an occurrence probability that the change point in the first data series and the change point in the second data series occur synchronously, associating the first identifier with the second identifier as an identifier related to a tracing target based on the occurrence probability, and determining a type of identification target of the associated first identifier or second identifier based on the occurrence probability. Effect of the Invention

[0012] According to the present disclosure, it is possible to provide a data processing device, a data processing method, and a non-transitory computer-readable medium that are capable of improving tracing accuracy. [Brief description of the drawings]

[0013] [Figure 1] 1 is a configuration diagram showing an overview of a data processing device according to an embodiment; [Diagram 2] 1 is a configuration diagram showing a configuration example of a data corresponding device according to a first embodiment; [Diagram 3] 5 is a flowchart showing an operation example of the data corresponding device according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing a specific example of a first data sequence according to the first embodiment. [Diagram 5]FIG. 4 is a diagram showing a specific example of a first data sequence according to the first embodiment. [Figure 6] FIG. 4 is a diagram showing a specific example of a second data sequence according to the first embodiment. [Figure 7A] FIG. 4 is a diagram showing a method of detecting a change point of a second data sequence according to the first embodiment. [Figure 7B] FIG. 4 is a diagram showing a method of detecting a change point of a second data sequence according to the first embodiment. [Figure 7C] FIG. 4 is a diagram showing a method of detecting a change point of a second data sequence according to the first embodiment. [Figure 8] FIG. 4 is a diagram showing a specific example of a second data sequence according to the first embodiment. [Figure 9] FIG. 11 is a configuration diagram showing a configuration example of a data corresponding device according to a second embodiment. [Figure 10] 13 is a flowchart showing an example of the operation of the data corresponding device according to the second embodiment. [Figure 11] FIG. 2 is a diagram illustrating an example of a hardware configuration of a computer according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, an embodiment will be described with reference to the drawings. In the drawings, the same elements are denoted by the same reference numerals, and repeated explanations will be omitted as necessary.

[0015] (Review of related technologies) In a related tracing system, short-range wireless sensors are placed at gates on floors, and when a user passes through the gate, the short-range wireless sensors detect the wireless terminal carried by the user. Furthermore, after the user passes through the gate, a camera or other device is used to identify the user and detect his / her position, thereby tracing the user. In this case, the related tracing system associates the data of a person observed by a camera immediately after observing the wireless terminal at the gate with the data of the wireless terminal as data of the same person (trace target).

[0016] The present inventor has studied related tracing systems and found the following problem: In other words, in related tracing systems, when radio waves emitted by terminals with various characteristics such as notebook PCs (Personal Computers), tablets, smartphones, and smart watches, such as wireless LAN terminals, are used, if correlation is performed with other types of sensor data such as camera data, the accuracy of correlation and tracing is reduced.

[0017] The reason for this is that the observation target (detection target) includes devices such as laptops that are not necessarily carried by the user all the time. For example, if a user moves their laptop while it is still on a desk, etc., and moves it into the camera's shooting range, the related technology will perform incorrect association. Even if the association is correct, if the user's location is traced using a wireless device that is rarely carried and moved around, the location cannot be traced correctly, resulting in a problem of reduced tracing accuracy.

[0018] Therefore, in the following embodiment, when a user owns various types of wireless terminals, it becomes possible to accurately associate and trace different types of data.

[0019] (Outline of the embodiment) 1 shows an overview of a data processing device according to an embodiment. As shown in FIG 1, a data processing device 10 includes a detection unit 11, a calculation unit 12, a correspondence unit 13, and a discrimination unit 14.

[0020] The detection unit 11 detects a change point of a first data sequence having a first identifier and a change point of a second data sequence having a second identifier. The calculation unit 12 calculates an occurrence probability that a change point of the first data sequence and a change point of the second data sequence occur synchronously. The correspondence unit 13 associates the first identifier and the second identifier as identifiers related to the tracing target based on the occurrence probability. The discrimination unit 14 discriminates the type of the identification target of the associated first identifier or second identifier based on the occurrence probability.

[0021] With this configuration, it is possible to accurately match a first identifier in a first data sequence with a second identifier in a second data sequence, and it is also possible to determine the type of identification target of the matched first or second identifier, thereby improving tracing accuracy.

[0022] For example, in a system that combines face recognition using a camera and wireless positioning using wireless sensors to trace a user, a person identified by the camera can be automatically associated with a wireless device that the person owns. In particular, when using general-purpose wireless devices such as smartphones and notebook PCs, even in cases where the user does not necessarily carry the wireless device, it is possible to perform highly accurate association and identify the characteristics of the wireless device (such as whether it is always carried or sometimes placed on a desk).

[0023] (Embodiment 1) Hereinafter, the first embodiment will be described with reference to the drawings.

[0024] <Configuration of First Embodiment> An example of the configuration of a data correspondence device according to this embodiment is shown in Fig. 2. The data correspondence device 100 according to this embodiment is a device that associates data (identifiers) related to a traced user in a tracing system that traces the user's location. For example, the tracing system includes the data correspondence device 100 and a tracing device, and the tracing device traces the user based on the data associated by the data correspondence device 100. The tracing system makes it possible to instantly identify the location of a person in an office where free address systems are becoming more common, and to optimize the placement of things by tracing the location of a person in a factory, etc.

[0025] As shown in FIG. 2, data corresponding device 100 according to this embodiment includes data storage section 101, change point detection section 102, detection method selection section 103, identifier correspondence section 104, and terminal type determination section 105.

[0026] The data storage unit 101 stores data (identifiers) associated by the data corresponding device 100. Specifically, the data storage unit 101 stores data acquired by a plurality of types of sensors for each data type. A plurality of data for each data type may be referred to as a data series, and for example, the data storage unit 101 stores data of a first data series and data of a second data series.

[0027] The data type may be, for example, face identification data obtained by analyzing a camera image, or position data of a wireless terminal estimated by observing radio waves transmitted from the wireless terminal. The data stored by the data storage unit 101 may include, for example, the time when the data was acquired, and an identifier (also called an observation target ID) that uniquely identifies the target observed by the sensor. The observation target ID may be, for example, an ID that uniquely identifies a person obtained by face identification, or the MAC address of the wireless terminal.

[0028] The change point detection unit 102 calculates (detects) the time when a change appears in the characteristics of data for each data type stored in the data storage unit 101 (change point). The change point detection unit 102 calculates the change point for data including a plurality of observation target IDs for each data type. For example, the change point detection unit 102 detects the change point of the first data series and the change point of the second data series. The change point detection unit 102 detects the time when the state of the time series data in each data series changes as the change point. The state change of the time series data includes a change from a state where the time series data is not recorded to a state where it is recorded, and includes, for example, a change from a state where a person is not recognized from the camera image to a state where a person is recognized, when the time series data is a camera image. Note that the example shown here is merely an example, and any state in which the state of the time series data changes from one state to another may be used. For example, when the time series data is positioning data, the value of the time series data may be included from outside a specific range to within the range, such as the positioning data changing from outside the camera's shooting range to within the range.

[0029] It should be noted that the method of calculating the change points can be different for each data type, and the calculation is performed based on the method selected by the detection method selection unit 103. The change point detection unit 102 detects change points in the first data series based on a first detection method corresponding to the first data series, and detects change points in the second data series based on a second detection method corresponding to the second data series.

[0030] The detection method selection unit 103 selects a method for detecting change-points based on the data type, and selects a method for detecting change-points for each set of data (data series) to be analyzed. The detection method selection unit 103 selects a first detection method according to the type of the first data series, and selects a second detection method according to the type of the second data series.

[0031] The identifier correspondence unit 104 calculates (associates) a pair of corresponding observation target IDs based on the change points of data for each data type calculated by the change point detection unit 102. Here, calculating a corresponding observation target ID means, for example, calculating a pair of a person (identifier) ​​identified by face recognition and the MAC address of a wireless terminal owned by that person. In other words, the identifier correspondence unit 104 associates a first observation target ID of a first data sequence with a second observation target ID of a second data sequence as an identifier related to the person to be traced.

[0032] The identifier correspondence unit 104 is also a calculation unit that calculates the occurrence probability of a synchronous change point where the change points for each data type are synchronized. For example, the identifier correspondence unit 104 calculates the occurrence probability of a synchronous change point between a change point of a first data series and a change point of a second data series, and associates a first observation target ID of the first data series with a second observation target ID of the second data series based on the calculated occurrence probability. The occurrence probability is the probability that a change point of the first data series and a change point of the second data series occur within a predetermined time range. The occurrence probability may be calculated by aggregating the synchronous change points within a predetermined time range using any statistical method. For example, the identifier correspondence unit 104 associates a first observation target ID with a second observation target ID according to a comparison result between the occurrence probability and a first threshold value.

[0033] The terminal type determination unit 105 determines (discriminates) the type of the observation target (e.g., wireless terminal) of the associated observation target ID based on the occurrence probability of the synchronization change point calculated by the identifier correspondence unit 104. The terminal type determination unit 105 determines the type of the identification target of either the first observation target ID of the associated first data sequence or the second observation target ID of the associated second data sequence. Among the first observation target ID and the second observation target ID, the unit 105 determines the type of an ID including multiple types such as a wireless terminal. For example, the terminal type determination unit 105 compares the occurrence probability with a second threshold value, and determines the type of the wireless terminal based on the comparison result. For example, the terminal type determination unit 105 determines, based on the occurrence probability, whether the type of the terminal linked to the observation target ID is a terminal that is frequently carried such as a smartphone (a mobile terminal that is carried) or a terminal that is relatively often left on a desk such as a notebook PC (a non-mobile terminal that is not carried).

[0034] <Operation of the First Embodiment> Fig. 3 is a diagram showing an example of the operation of the data corresponding device according to the present embodiment. As shown in Fig. 3, first, the change point detection unit 102 acquires data of a first data series (first data) from the data accumulation unit 101 (step S1001). In this example, the first data series will be described as data in which a face is identified by a specific camera. For the sake of explanation, the two data series are referred to as the first data series and the second data series, but unless otherwise specified, either data series may be the first data series or the second data series (the same applies to the first identifier and the second identifier, etc.).

[0035] FIG. 4 is an example of the first data series acquired in step S1001. As shown in FIG. 4, the first data series in this example is a data group having information on time, camera ID, and identification ID. That is, in the first data series, data in which the time, camera ID, and identification ID are associated with each other are arranged in a chronological order. The time is the time when face identification was performed, and is described as the number of seconds elapsed since 00:00 on January 1, 1970. In this example, the data is obtained from an image at a 1-second interval and face identification was performed, and if a face is identified, data is recorded at a 1-second interval, and if a face is not identified due to reasons such as the face not being captured by the camera, data is not recorded. The camera ID is an identifier that uniquely identifies the camera used for face identification. The identification ID (first identifier) ​​is an ID that uniquely identifies a person identified by face identification. For example, the first identifier can also be said to be an identifier that indicates an identification target identified by a predetermined identification process from data for generating the first data series. In this example, the identification process is a process of performing image recognition of an identification target (a person's face) from an image captured by a camera.

[0036] Next, the change-point detection unit 102 detects a change-point in the first data based on the detection method selected by the detection method selection unit 103 (step S1002). Here, since the first data series is face identification data from the camera, the detection method selection unit 103 selects a change-point detection method for face identification data. The algorithm selected here is an algorithm that detects the time when face identification begins or ends as a change point. In this case, the change in the state of the data detected as a change point can also be said to be a change in the identification state of the identification target by the identification process. For example, the change in the identification state is a change in whether or not the identification target has been identified by the identification process (between an identified state and a non-identified state).

[0037] For example, according to the data in FIG. 4, faces are continuously identified (every second) before the time "1601379163", and then face identification is not performed until the time "1601379311", so the time "1601379163" is detected as a change point. For example, the change point detection unit 102 detects the time when the identification ID changes in the first data series as a change point. Specifically, the identification ID is "U1" until the time "1601379163", and the identification ID changes to "U2" at the time "1601379311", so the time "1601379163" is determined as a change point. Note that the change point may have a type. That is, the change point detection unit 102 may detect the change point of the data and determine the change type of the change point according to the change in the state of the data. Here, since the face with the identification ID "U1" changes to an unidentifiable state at the change-point at time "1601379163," the type of the change-point is set to "departure."

[0038] In addition, in the data of FIG. 4, before the time "1601379563", face recognition is not performed, and after that, face recognition is continuously performed. Therefore, the time "1601379563" is detected as a change point and its type is "approach". In addition, when focusing on the time "1601379311", face recognition is not performed before or after that, so a transition from a non-recognized state to a recognized state occurs in one period, and then a transition immediately occurs again to a non-recognized state. This is also detected as a change point. Its type is "passage". FIG. 5 is a diagram showing an example of a change point detected from the first data series of FIG. 4.

[0039] Next, the change-point detection unit 102 acquires data of the second data series (second data) from the data accumulation unit 101 (step S1003). In this example, the second data series is described as position data of a wireless terminal estimated using a wireless sensor.

[0040] FIG. 6 is an example of the second data series acquired in step S1003. As shown in FIG. 6, the second data series in this example is a data group having information on time, location, and MAC address. That is, in the second data series, data associating time, location, and MAC address are arranged in chronological order. The time is the time when the location data was generated, and is described as the number of seconds elapsed since 00:00 on January 1, 1970, as in FIG. 4. In this example, the location data is generated at 10-second intervals. The MAC address (second identifier) ​​is an identifier that uniquely identifies the wireless terminal whose location has been estimated. It can also be said that the second data series includes detection data (the location of the second identifier) ​​associated with the second identifier. Note that only one type of MAC address is described here for the sake of explanation, but in reality, data of multiple MAC addresses may be included.

[0041] Next, the change-point detection unit 102 detects a change-point in the second data based on the detection method selected by the detection method selection unit 103 (step S1004). Here, the detection method selection unit 103 selects a change-point detection method based on the first data series and the second data series. That is, a change-point detection method for the positioning data series that depends on the position of the camera indicated by the camera ID of the first data series is adopted.

[0042] 7A to 7C are diagrams showing a method for detecting change points in a second data series according to this embodiment. In FIG. 7A to FIG. 7C, a dotted circle with camera C1 at the center is the shooting range R1 of camera C1. Assume that camera C1 is located at a position (10,25) (unit: meters), and the shooting range R1 is 5 m from C1. Also, a solid circle represents a time series transition of radio positioning data (position P1) for a terminal with a certain MAC address (here, AA:BB:CC:DD:EE:FF). As with the first data series, three types of change points, "approach", "departure", and "passage", are detected for the second data series as well.

[0043] In this case, the change in the state of the data detected as the change point is a change in the detected data (position). The change point detection unit 102 detects the time when the position of the second data series changes as the change point, and determines the type of change of the change point according to the change in position. The change in the state of the data detected is a change in position relative to the boundary of the shooting range R1 of the camera C1 (between inside the shooting range and outside the shooting range). The shooting range R1 can also be said to be an area from which data for generating the first data series can be acquired.

[0044] FIG. 7A shows a method for detecting a change point of "approach", where the time when the position P1 of the radio positioning moves from outside the shooting range R1 of the camera C1 into the range R1 is detected as a change point. For example, referring to the example of FIG. 6, at the time "1601379130", the position changes from (10,10) (position outside R1) to (10,28) (position within R1), which is a change point of "approach". FIG. 7B shows a method for detecting a change point of "departure", where the time when the position P1 of the radio positioning moves from within the shooting range R1 of the camera C1 out of the range R1 is detected as a change point. Referring to the example of FIG. 6, at the time "1601379170", the position changes from (10,28) to (10,10), which is a change point of "departure". FIG. 7C shows a method for detecting a change point of "passage". In Fig. 7C, the data of position P1 from radio positioning is not directly within the shooting range R1 of camera C1, but when the data of position P1 is interpolated with a straight line or the like, it crosses the shooting range R1 of camera C1. Fig. 8 is a diagram showing an example of a change point detected from the second data series in Fig. 6.

[0045] Next, the identifier correspondence unit 104 extracts, as a synchronization change point, a change point occurring within a predetermined time difference from among the change points of the first data series and the change points of the second data series (step S1005). If the change points have a type, only change points of the same type are extracted. Here, the predetermined time difference is 10 seconds. With reference to the examples of Figs. 5 and 8, for example, the data at time "1601379163" in Fig. 5 and the data at time "1601379170" in Fig. 8 are both change points of "leaving" and the time difference is within 10 seconds, so they are detected as a synchronization change point. At this time, it is assumed that the identifier ID "U1" and the MAC address "AA:BB:CC:DD:EE:FF" are synchronized.

[0046] Next, the identifier correspondence unit 104 calculates the occurrence probability of a synchronization change point, and compares the calculated occurrence probability with a first threshold value (step S1006). The occurrence probability of a synchronization change point is, for example, the ratio of the total number of synchronization change points including the change points of the first data series and the change points of the second data series within a predetermined time range to the total number of change points of the detected first data series. Specifically, the total number of detected change points is counted for the identifier ID "U1" in the first data series, and it is assumed that it is N1. After that, the total number of synchronization change points for the identifier ID "U1" of the first data series and the MAC address "AA:BB:CC:DD:EE:FF" of the second data series is counted, and it is assumed that it is N2. In this case, the occurrence probability of a synchronization change point is calculated as N2 / N1.

[0047] If the calculated occurrence probability exceeds the first threshold (or is equal to or greater than the first threshold) (Yes in step S1007), the identifier correspondence unit 104 links the identifiers of the detected synchronization change points (step S1008). That is, it associates the identifier of the first data with the identifier of the second data. In this example, it is determined that a person with identifier ID "U1" owns a wireless terminal with MAC address "AA:BB:CC:DD:EE:FF". If the occurrence probability is equal to or less than the first threshold (No in step S1007), it ends the process without associating the identifiers.

[0048] In this embodiment, one identifier of the first data sequence is linked to one identifier of the second data sequence, but multiple identifiers may be linked. That is, when multiple second identifiers are included in the second data sequence, the occurrence probability of a synchronization change point may be calculated for each second identifier, and the first identifier may be associated with multiple second identifiers according to the comparison result between each calculated occurrence probability and the first threshold. For example, it may be determined that a person with identifier ID "U1" owns a wireless terminal with MAC address "AA:BB:CC:DD:EE:FF" and a wireless terminal with MAC address "AA:BB:CC:DD:EE:F0". In this way, it is possible to handle the case where one person owns multiple wireless terminals.

[0049] In step S1008, after determining that the person with identifier ID "U1" owns a wireless terminal with MAC address "AA:BB:CC:DD:EE:FF", the terminal type determination unit 105 compares the occurrence probability of a synchronization change point calculated in step S1006 with a second threshold value (step S1009).

[0050] If it is determined that the occurrence probability is greater than the second threshold (or equal to or greater than the second threshold) (Yes in step S1010), the terminal type determination unit 105 determines that the type of the terminal corresponding to the identifier is the first type (step S1011). In this example, the type of the wireless terminal is determined to be a mobile terminal of a type that is always carried, such as a smartphone. Also, if it is determined that the occurrence probability is equal to or less than the second threshold (or smaller than the second threshold) (No in step S1010), the terminal type determination unit 105 determines that the type of the terminal corresponding to the identifier is the second type (step S1012). In this example, the type of the wireless terminal is determined to be a non-mobile terminal of a type that is not always carried, such as a notebook computer or desktop computer. When a plurality of second identifiers are associated with the first identifier, the terminal type may be determined for each second identifier. When the data corresponding device 100 associates the identifiers of each data and determines the terminal type of the identifier, the trace system traces the user's location using the data of the associated identifier based on the determined terminal type.

[0051] <Advantages of the First Embodiment> As described above, in this embodiment, the focus is on the change points of data in the data series, the synchronized change points are statistically tallied, and the identifiers are associated with each other based on the probability of the synchronization occurring. Therefore, for example, even if the user does not always carry the wireless terminal, it is possible to prevent incorrect association. In addition, since the association is performed by the type of change point, multiple change points can be judged in an integrated manner, improving the accuracy of the association.

[0052] In addition, in this embodiment, two thresholds are used for the synchronization occurrence probability. This is to determine the type of terminal owned by the user. For example, there are types of terminals, such as smartphones and smart watches, which are always carried by the user, and terminals such as notebook PCs, which are often moved while left on a desk. Terminals that are always carried by the user have a high probability of synchronization, while terminals that are not always carried by the user have a lower probability of synchronization than terminals that are always carried. Therefore, by setting the threshold of the synchronization change point in two stages, it is possible to distinguish between the two.

[0053] According to this embodiment, not only can multiple types of data originating from the same user be linked with high accuracy, but the attributes of the data (such as data from a terminal that is always carried) can also be determined at the same time.

[0054] In this embodiment, only one threshold is set for determining the terminal type, but multiple thresholds may be set. For example, if the occurrence probability exceeds the highest threshold (third threshold), it may be determined to be a smart watch (a mobile terminal with higher mobility), if the occurrence probability is equal to or less than the third threshold and equal to or more than the second threshold, it may be determined to be a smartphone (a mobile terminal with lower mobility), and otherwise it may be determined to be a notebook computer (a non-mobile terminal). In this way, it is possible to classify the terminal types in more detail.

[0055] In addition, in the present embodiment, an example of determining the terminal type using a threshold has been described, but it is not necessary to use a threshold, and a continuously changing attribute value of the terminal may be calculated. For example, the probability that a user moves while carrying a terminal may be calculated from the occurrence probability of a synchronous change point. In this way, it becomes possible to classify the second data series more finely.

[0056] (Embodiment 2) Next, a second embodiment will be described. In this embodiment, when a first identifier (e.g., a user name) of a first data sequence (e.g., a camera) is associated with a plurality of second identifiers (e.g., MAC addresses) of a second data sequence (e.g., radio positioning data), which data of the second data sequence is to be used is determined based on the attribute (e.g., a terminal carried at all times) of each second identifier of the second data sequence. When tracking the user's position, the accuracy of tracking the user is higher when the position of a terminal carried at all times, such as a smartphone or a smartwatch, is used than when information of a terminal such as a notebook PC or a desktop PC that is often left on a desk and the user gets up from the desk is used.

[0057] An example of the configuration of data corresponding device 100 according to this embodiment will be described with reference to Fig. 9. As shown in Fig. 9, data corresponding device 100 according to this embodiment includes data storage section 101, change point detection section 102, detection method selection section 103, identifier correspondence section 104, terminal type determination section 105, and location estimation section 116. Note that since the configuration other than location estimation section 116 is the same as that of embodiment 1, description thereof will be omitted.

[0058] The location estimation unit (data selection unit) 116 selects location information to be used for tracing from a second data series having multiple location information (data of a second identifier) ​​linked to a specific identifier (first identifier) ​​of the first data series, based on the terminal type determined by the terminal type determination unit 105.

[0059] Next, an example of the operation of the data correspondence device according to the present embodiment will be described with reference to FIG. 10. The operations from step S1001 to step S1012 are the same as those in the first embodiment, and therefore detailed description will be omitted. In this example, it is assumed that the MAC addresses "AA:BB:CC:DD:EE:FF" and "AA:BB:CC:DD:EE:F0", which are identifiers of the second data series that is positioning data, are linked to the identifier ID "U1" in the first data series in the operations up to step S1008. It is also assumed that the terminal type of "AA:BB:CC:DD:EE:FF" is determined to be a smartphone (first type) and the terminal type of "AA:BB:CC:DD:EE:F0" is determined to be a notebook PC (second type) in the operations up to step S1012.

[0060] Thereafter, the location estimation unit 116 selects which identifier (MAC address) of the second data sequence to use for tracing, based on the terminal type determined in steps S1011 and S1012 (step S1113). In this example, it is determined to use the location data of "AA:BB:CC:DD:EE:FF" determined to be a smartphone (first type). When the data corresponding device 100 selects the location data, the tracing system uses the selected location data to trace the user's location.

[0061] As described above, in this embodiment, when the same user owns multiple terminals, the type of terminal is identified and, based on that, it is determined which terminal's location information to use, making it possible to track the user with high accuracy.

[0062] The present disclosure is not limited to the above-described embodiment, and can be modified as appropriate without departing from the spirit and scope of the present disclosure.

[0063] Each configuration in the above-described embodiment may be configured by hardware or software, or both, and may be configured by one piece of hardware or software, or may be configured by multiple pieces of hardware or software. Each device and each function (processing) may be realized by a computer 20 having a processor 21 such as a CPU (Central Processing Unit) and a memory 22 which is a storage device, as shown in Fig. 11. For example, a program for performing the method (data processing method) in the embodiment may be stored in the memory 22, and each function may be realized by the processor 21 executing the program stored in the memory 22.

[0064] These programs can be stored and provided to a computer using various types of non-transitory computer readable media. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROM (Read Only Memory), CD-R, CD-R / W, and semiconductor memory (e.g., mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (random access memory)). The programs may also be provided to a computer by various types of transitory computer readable media. Examples of transitory computer readable media include electric signals, optical signals, and electromagnetic waves. The transitory computer readable media can provide the programs to a computer via a wired communication path such as an electric wire and an optical fiber, or via a wireless communication path.

[0065] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-mentioned embodiments. Various modifications that can be understood by a person skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

[0066] A part or all of the above-described embodiments can be described as, but is not limited to, the following supplementary notes. (Appendix 1) detection means for detecting a change point of a first data sequence having a first identifier and a change point of a second data sequence having a second identifier; a calculation means for calculating an occurrence probability that a change point of the first data sequence and a change point of the second data sequence occur synchronously; a correspondence means for associating the first identifier with the second identifier as identifiers related to the tracing target based on the occurrence probability; A discrimination means for discriminating a type of the identification target of the associated first identifier or the associated second identifier based on the occurrence probability; A data processing device comprising: (Appendix 2) the detection means detects a change point of the first data sequence based on a first detection method corresponding to the first data sequence, and detects a change point of the second data sequence based on a second detection method corresponding to the second data sequence; 2. A data processing device according to claim 1. (Appendix 3) a detection method selection means for selecting the first detection method in accordance with a type of the first data sequence and selecting the second detection method in accordance with a type of the second data sequence, 3. A data processing device according to claim 2. (Appendix 4) the detection means detects a time at which a data state changes in the first or second data series as a change point; 4. A data processing device according to any one of claims 1 to 3. (Appendix 5) The detection means determines a change type of the change point according to a change in the state of the data. 5. A data processing device according to claim 4. (Appendix 6) the first or second identifier is an identifier indicating an identification target identified by a predetermined identification process from data for generating the first or second data sequence, the change in the state of the data is a change in the identification state of the identification target by the identification process; 6. A data processing device according to claim 4 or 5. (Appendix 7) The change in the identification state is a change in whether or not the identification target has been identified by the identification process. 7. A data processing device according to claim 6. (Appendix 8) The identification process is a process of image-recognizing the identification target from an image captured by a camera. 8. A data processing device according to claim 6 or 7. (Appendix 9) the first and second data series include sensing data associated with the first and second identifiers, The change in the state of the data is a change in the sensed data. 6. A data processing device according to claim 4 or 5. (Appendix 10) the detection data is data indicating a position of an identification target of the first or second identifier, The change in the state of the data is a change in the position of the sensor data relative to a boundary of a predetermined area. 10. A data processing device according to claim 9. (Appendix 11) The occurrence probability is a probability that a change point of the first data sequence and a change point of the second data sequence occur within a predetermined time range. 11. A data processing device according to any one of claims 1 to 10. (Appendix 12) The occurrence probability is a ratio of a total number of synchronization change points including a change point of the first data sequence and a change point of the second data sequence within a predetermined time range to a total number of change points of the detected first data sequence. 12. A data processing device according to claim 11. (Appendix 13) The corresponding means, when the occurrence probability is greater than a first threshold value, associates the first identifier with the second identifier. 13. A data processing device according to any one of claims 1 to 12. (Appendix 14) The discrimination means discriminates a type of the identification target of the first identifier or the second identifier according to a result of comparing the occurrence probability with a second threshold value. 14. A data processing device according to any one of claims 1 to 13. (Appendix 15) When the occurrence probability is greater than the second threshold, the determination means determines that the identification target of the first identifier or the second identifier is a mobile device. 15. A data processing device according to claim 14. (Appendix 16) the second data sequence includes a plurality of the second identifiers; The correspondence means corresponds the first identifier to the plurality of second identifiers. 16. A data processing device according to any one of claims 1 to 15. (Appendix 17) a data selection means for selecting data of a second identifier to be used for tracing from the data of the plurality of associated second identifiers based on the determined terminal type; 17. A data processing device according to claim 16. (Appendix 18) Detecting a change point of a first data series having a first identifier and a change point of a second data series having a second identifier; calculating an occurrence probability that a change point of the first data series and a change point of the second data series occur synchronously; Corresponding the first identifier to the second identifier as identifiers related to the tracing target based on the occurrence probability; determining a type of an identification target of the associated first identifier or the associated second identifier based on the occurrence probability; Data processing methods. (Appendix 19) Detecting a change point of a first data series having a first identifier and a change point of a second data series having a second identifier; calculating an occurrence probability that a change point of the first data series and a change point of the second data series occur synchronously; Corresponding the first identifier to the second identifier as identifiers related to the tracing target based on the occurrence probability; determining a type of an identification target of the associated first identifier or the associated second identifier based on the occurrence probability; A non-transitory computer-readable medium having stored thereon a data processing program for causing a computer to carry out a process. [Explanation of symbols]

[0067] 10 Data Processing Device 11 Detection section 12 Calculation section 13. Correspondence Department 14 Discrimination part 20. Computers 21 Processors 22 Memory 100 Data-capable devices 101 Data Storage Unit 102 Change point detection unit 103 Detection method selection unit 104 Identifier correspondence part 105 Terminal type determination unit 116 Position estimation part

Claims

1. A detection means for detecting a change point of a first data sequence having a first identifier capable of identifying a tracing object and a change point of a second data sequence having a second identifier for tracing the tracing object; a calculation means for calculating an occurrence probability that a change point of the first data series and a change point of the second data series occur synchronously; a correspondence means for associating the first identifier with the second identifier as identifiers related to the tracing target when the occurrence probability is greater than a first threshold value; a determination means for determining that a type of the identification object of the associated second identifier is a mobile device when the occurrence probability is greater than a second threshold value; a tracing means for tracing the tracing target based on the determined type; A data processing device comprising:

2. the detection means detects a change point of the first data sequence based on a first detection method corresponding to the first data sequence, and detects a change point of the second data sequence based on a second detection method corresponding to the second data sequence; 2. A data processing apparatus according to claim 1.

3. a detection method selection means for selecting the first detection method in accordance with a type of the first data sequence and for selecting the second detection method in accordance with a type of the second data sequence, 3. A data processing apparatus according to claim 2.

4. the detection means detects a time at which a state of data changes in the first or second data series as a change point; A data processing device according to any one of claims 1 to 3.

5. The detection means determines a change type of the change point according to a change in the state of the data.

5. A data processing apparatus according to claim 4.

6. the first or second identifier is an identifier indicating an identification target identified by a predetermined identification process from data for generating the first or second data sequence, the change in the state of the data is a change in the identification state of the identification target by the identification process; 6. A data processing device according to claim 4 or 5.

7. The change in the identification state is a change in whether or not the identification target has been identified by the identification process.

7. A data processing apparatus according to claim 6.

8. The identification process is a process of image-recognizing the identification target from an image captured by a camera.

8. A data processing device according to claim 6 or 7.

9. the first or second data series includes sensing data associated with the first or second identifier, The change in the state of the data is a change in the sensed data.

6. A data processing device according to claim 4 or 5.

10. the detection data is data indicating a position of an identification target of the first or second identifier, The change in the state of the data is a change in the position of the sensor data relative to a boundary of a predetermined area.

10. A data processing apparatus according to claim 9.

11. the occurrence probability is a probability that a change point of the first data series and a change point of the second data series occur within a predetermined time range; A data processing apparatus according to any one of claims 1 to 10.

12. the occurrence probability is a ratio of a total number of synchronization change points including a change point of the first data sequence and a change point of the second data sequence within a predetermined time range to a total number of change points of the detected first data sequence, A data processing apparatus according to claim 11.

13. the second data sequence includes a plurality of the second identifiers; The correspondence means corresponds the first identifier to the plurality of second identifiers. A data processing apparatus according to any one of the preceding claims.

14. a data selection means for selecting data of a second identifier to be used for tracing from the data of the plurality of associated second identifiers based on the determined type; 14. A data processing apparatus according to claim 13.

15. A method for detecting a change point of a first data sequence having a first identifier capable of identifying a tracing object and a change point of a second data sequence having a second identifier for tracing the tracing object, calculating an occurrence probability that a change point of the first data series and a change point of the second data series occur synchronously; If the occurrence probability is greater than a first threshold, the first identifier and the second identifier are associated with each other as identifiers related to the tracing target; If the occurrence probability is greater than a second threshold, determining that the type of the identification object of the associated second identifier is a mobile device; tracing the tracing target based on the determined type; Data processing methods.

16. A method for detecting a change point of a first data sequence having a first identifier capable of identifying a tracing object and a change point of a second data sequence having a second identifier for tracing the tracing object, calculating an occurrence probability that a change point of the first data series and a change point of the second data series occur synchronously; If the occurrence probability is greater than a first threshold, the first identifier and the second identifier are associated with each other as identifiers related to the tracing target; If the occurrence probability is greater than a second threshold, determining that the type of the identification object of the associated second identifier is a mobile device; tracing the tracing target based on the determined type; A data processing program for causing a computer to execute processing.

Images