JP6089764B2 - Activity status processing apparatus and activity status processing method - Google Patents

Description

  The present invention relates to a technique for acquiring user activity status information.

  Currently, a system that provides various services suitable for each user by collecting user information that can represent various actions of the user has been developed. For example, Patent Literature 1 below recommends a product that is likely to be preferred by the user while updating information related to the user (user profile) by collecting the user's access history to the WEB page as user behavior information. A method is proposed. The following Patent Document 2 proposes a method of separating the user's actions by collecting the user's image, sound, and contact state by a sensor mounted on the robot, and causing the robot to perform output in accordance with the user.

  In addition to the above-described system, sensor data such as acceleration, angular velocity, and heart rate is collected from a mobile terminal equipped with various sensors, and the user's movement status (behavioral state) such as walking, running, or riding a train is identified. Technology has also been proposed.

  However, such a technique using a portable terminal with limited available power requires a device for suppressing the power consumption of the portable terminal. For example, the process of identifying the user's movement status and the process of transmitting data to the server consume the power of the mobile terminal.

  Therefore, the following Patent Document 3 groups predetermined terminals or terminals located in the vicinity, measures a position by GPS (Global Positioning System), and communicates to a server only for a specific terminal of each group. We have proposed a method to reduce the power consumption of the terminal.

International Publication No. 2003-0117137 JP 2006-088251 A JP 2010-258906 A

  However, the proposed method of Patent Document 3 cannot be applied to an apparatus or system that identifies a user's activity status. Since each user located in the vicinity does not always perform the same activity, the information acquired by the specific terminal determined by the proposed method can accurately identify the activity status of other terminals in the same group. It is not possible. For example, even in the same building or the same floor, each user often takes different actions.

  This invention is made | formed in view of such a situation, and provides the technique which acquires the activity condition of each user accurately, suppressing power consumption or communication amount.

  Each aspect of the present invention employs the following configurations in order to solve the above-described problems.

  The first aspect relates to an activity status processing apparatus. The activity status processing device according to the first aspect is based on a plurality of activity status information indicating at least activity statuses of a plurality of users identified using data obtained by a plurality of user devices, and users having similar activity statuses are identified with each other. A classification unit that classifies the plurality of user devices into at least one activity group by grouping, a selection unit that selects at least one representative user device from the plurality of user devices belonging to the activity group, and a selection unit And an instruction unit for instructing other user devices belonging to the activity group other than the representative user device selected in the above to suppress specific processing related to the activity status information.

  The second aspect relates to an activity status processing method executed by at least one computer. The activity status processing method according to the second aspect is based on a plurality of activity status information indicating at least activity statuses of a plurality of users identified using data obtained by a plurality of user devices. By grouping, the plurality of user devices are classified into at least one activity group, and at least one representative user device is selected from the plurality of user devices belonging to the activity group, and other than the selected representative user device Instructing other user devices belonging to the activity group to suppress specific processing related to the activity status information.

  As another aspect of the present invention, there may be a program that causes at least one computer to execute the activity status processing method according to the second aspect, or a computer-readable recording medium that records such a program. It may be. This recording medium includes a non-transitory tangible medium.

  According to each said aspect, the technique which acquires the activity condition of each user accurately can be provided, suppressing power consumption or communication amount.

It is a figure which shows notionally the structural example of the activity condition processing apparatus which concerns on embodiment of this invention. It is a figure which shows notionally the structural example of the activity condition collection system in 1st Embodiment. It is a figure which shows notionally the process structural example of the activity condition collection server in 1st Embodiment. It is a figure which shows the example of the information stored in an information storage part. It is a figure which shows the example of the parameter | index corresponding to a place attribute. It is a figure which shows the example of the information regarding an activity group. It is a sequence chart which shows the operation example of the activity condition collection system in 1st Embodiment. It is a figure which shows notionally the process structural example of the activity condition collection server in 2nd Embodiment. It is a figure which shows the example of the disassembly space | interval of an activity group. It is a figure which shows the example of the disassembly space | interval and reclassification space | interval of an activity group. It is a sequence chart which shows the operation example of the activity condition collection system in 2nd Embodiment. It is a figure which shows the example of the activity condition log | history of a user apparatus (0001) and a user apparatus (0002). It is a figure which shows the example of a display of an activity condition. It is a figure which shows the example of a display of an activity condition.

  Embodiments of the present invention will be described below. In addition, embodiment mentioned below is an illustration and this invention is not limited to the structure of the following embodiment.

  FIG. 1 is a diagram conceptually showing a configuration example of an activity status processing apparatus 100 according to an embodiment of the present invention. As shown in FIG. 1, the activity status processing apparatus 100 is similar to an activity status based on a plurality of activity status information indicating at least the activity status of a plurality of users identified using data obtained by a plurality of user devices. A grouping unit 101 that classifies the plurality of user devices into at least one activity group, and a selection for selecting at least one representative user device from the plurality of user devices belonging to the activity group An instruction unit 103 for instructing other user devices belonging to the activity group other than the representative user device selected by the selection unit 102 to suppress specific processing related to the activity status information; Have.

  The activity status processing apparatus 100 has, for example, the same hardware configuration as the activity status collection server 5 in the detailed embodiment described later, and the program is processed in the same manner as the activity status collection server 5. A processing unit is realized. However, the hardware configuration of the activity status processing device 100 is not limited.

  The activity status processing method according to the embodiment of the present invention is executed by at least one computer such as the activity status processing device 100 described above. The activity status processing method groups users having similar activity statuses based on a plurality of activity status information indicating at least activity statuses of a plurality of users identified using data obtained by a plurality of user devices. The plurality of user devices are classified into at least one activity group, at least one representative user device is selected from the plurality of user devices belonging to the activity group, and the activity group other than the selected representative user device is selected. Instructing other user devices belonging to the control of specific processing related to the activity status information. However, the steps included in the activity status processing method may be executed sequentially in any order, or may be executed simultaneously.

  Here, a user's activity condition shows a user's movement condition, for example. The movement status of the user indicates stop, walking, running, stairs up / down, train, car, bicycle ride, and the like. Such a movement situation is identified by existing technology based on data acquired from a sensor carried by a user such as an acceleration sensor or a vibration sensor. The user activity status is not limited to the user movement status, but may indicate the user activity itself such as “reading” or “during soccer”. In addition, the above-described user activity status may be identified from a combination of supplementary information such as the user's position and a location attribute corresponding to the position, and the user's movement status. For example, the user activity status “reading” may be estimated from the place attribute “library” and the user movement status “stop”. Also, the user activity status “during soccer” may be estimated from the place attribute “soccer field” and the user movement status “running”.

  Such activity status information is generated by the user device, the activity status processing device 100, and other devices using data obtained by the user device. The data obtained by the user device is date / time data, sensor data obtained by a sensor as described above, position data obtained by GPS, and the like. That is, the user activity means a user's behavior or behavior that can be identified using data obtained by the user device. In addition, the user device refers to acquiring data that can identify the user's activities, such as a device carried by the user, a device mounted on the user's vehicle, and a device installed at a place visited by the user. It is a device that can.

  In the present embodiment, based on such activity status information, a plurality of user devices for the plurality of users are classified into at least one activity group by grouping users with similar activity statuses. As a result, user devices used by users with similar activity status are classified into the same activity group. In the present embodiment, at least one representative user device is selected for each activity group thus grouped according to the similarity of the user activity status, and other user devices belonging to the activity group other than the representative user device are selected. In response to this, the suppression of the specific processing related to the activity status information is instructed. The specific process related to the activity status information includes at least one of a process of uploading data for generating the activity status information, a process of generating the activity status information, and the like.

  Since users of user devices belonging to the same activity group are likely to perform similar activities, activity status information of other user devices other than the representative user device belonging to the same activity group is data obtained from the representative user device. It can be replaced with activity status information generated from That is, for other user devices belonging to the activity group other than the representative user device, even if the specific processing related to the activity status information is suppressed, the specific processing related to the activity status information only for the representative user device By executing this, it is possible to acquire the activity status information of all user devices belonging to the activity group. Further, by suppressing the specific processing of other user devices other than the representative user device, it is possible to reduce the power consumption of the other user devices or the amount of communication from the other user devices.

  Therefore, according to the present embodiment, it is possible to accurately acquire the activity status of each user while reducing power consumption or communication traffic.

  Hereinafter, the details of the above-described embodiment will be described. Hereinafter, the first to fourth embodiments will be exemplified as detailed embodiments. Each of the following detailed embodiments is an example when the above-described activity status processing apparatus 100 and activity status processing method are applied to an activity status collection server.

[First Embodiment]
〔System configuration〕
FIG. 2 is a diagram conceptually illustrating a configuration example of the activity status collection system 1 in the first embodiment. The activity status collection system 1 in the first embodiment includes an activity status collection server 5 and a plurality of user devices 3. The activity status collection server 5 and the plurality of user devices 3 are connected via the network 2 so that they can communicate with each other. The network 2 is a public network such as the Internet, a WAN (Wide Area Network), a LAN (Local Area Network), a wireless communication network, or the like. In the present embodiment, the communication mode between the activity status collection server 5 and each user device 3 is not limited.

  The activity status collection server 5 is a so-called computer and includes, for example, a CPU (Central Processing Unit) 10, a memory 11, an input / output interface (I / F) 12, a communication device 13 and the like that are connected to each other via a bus 15. . The memory 11 is a random access memory (RAM), a read only memory (ROM), a hard disk, a portable storage medium, or the like. The input / output I / F 12 can be connected to a user interface device such as a display device (not shown) or an input device (not shown). The communication device 13 communicates with each user device 3 via the network 2. In the present embodiment, the hardware configuration of the activity status collection server 5 is not limited. The activity status collection server 5 may be at least one of the following user devices 3.

  The user device 3 is a so-called computer such as a mobile phone, a smart phone, a mobile PC (Personal Computer), an in-vehicle device of a car navigation system, and has the above-described hardware configuration. Further, the user device 3 may be a human sensor, a surveillance camera device, a tag reader, a card reader, an NFC (Near Field Communication) receiver, or the like that is installed in a certain place and recognizes a user who has visited the place. Good. Further, the user device 3 may be a built-in device such as a toy or a health device that has an opportunity to come into contact with a person.

  In the present embodiment, the user device 3 generates user activity status information and transmits the activity status information to the activity status collection server 5. In the present embodiment, when the user apparatus 3 receives an instruction to suppress specific processing related to the activity status information from the activity status collection server 5, the user device 3 generates the activity status information and sends the activity status information to the activity status collection server 5. The transmission process is stopped as the specific process. The user apparatus 3 that has received the instruction to suppress the specific process may stop only one of the generation process and the transmission process. Further, the user device 3 may stop a part of the specific process without completely stopping the specific process.

  Here, in addition to the user activity status as described above, the activity status information is at least one supplement that can indicate the user activity status, such as the user location, the location attribute corresponding to the location, the date and time, and the like. Information may be indicated. The attribute of the place corresponding to the user's position is, for example, a store, an entertainment facility, an office, a station, an intersection, or the like. Also, holidays, holidays, New Years, Christmas, and the like may be indicated by activity status information because they may affect the user's activities. When the activity status information indicates the user's movement status, the user device 3 may include a sensor and a processing unit that identifies the user's movement status from sensor data obtained from the sensor. . In this case, a well-known method may be used as a method for identifying the movement status of the user. Here, an example of a well-known method for identifying a user's movement status using sensor data will be given.

  The user device 3 acquires sensor data for a certain period (time window) from the current time with respect to the sensor data, and performs statistical processing such as dispersion and averaging, frequency analysis by Fourier transform, and the like on the sensor data. To obtain a feature value for identifying the movement state. And the user apparatus 3 acquires the movement condition with respect to the said feature-value using the identification model which shows the relationship between the movement condition and feature-value which were built beforehand. The identification model is constructed using a widely known method such as a neural network, SVM (Support Vector Machine), Bayesian network, or decision tree. In addition, the neural network and the Bayesian network may output not only the movement situation but also the likelihood for the movement situation. Further, the user device 3 may prepare an identification model for each movement situation, and may select the movement situation having the highest likelihood based on the likelihood output from each identification model.

  Moreover, when a user's position is shown by activity status information, the user apparatus 3 may have a GPS receiver for acquiring the user's position. Moreover, when the location attribute of a user's position is shown by activity status information, the user apparatus 3 may hold | maintain the map information for specifying a location attribute. The present embodiment does not limit the specific content of the user activity status information generated by the user device 3. Further, the present embodiment does not limit the hardware configuration of the user device 3.

[Processing configuration]
FIG. 3 is a diagram conceptually illustrating a processing configuration example of the activity status collection server 5 in the first embodiment. Hereinafter, the activity status collection server 5 is simply referred to as the server 5. As shown in FIG. 3, the server 5 in the first embodiment includes an information acquisition unit 21, a calculation unit 22, an information storage unit 23, a classification unit 24, a selection unit 25, an instruction unit 26, and the like. Each of these processing units is realized, for example, by executing a program stored in the memory 11 by the CPU 10. In addition, the program may be installed from a portable recording medium such as a CD (Compact Disc) or a memory card or another computer on the network via the input / output I / F 12 and stored in the memory 11. Good.

  The information acquisition unit 21 acquires user activity status information from each user device 3 together with a device ID that can identify the user device 3. The information acquisition unit 21 stores the acquired user activity status information and the device ID in the information storage unit 23 in association with each other. In addition, the information acquisition part 21 may acquire the said activity status information directly from the user apparatus 3, and may acquire via another apparatus. Further, when the information acquisition unit 21 acquires the activity status information from the representative user device 3 selected as the representative of the activity group as described above, the information acquisition unit 21 uses the acquired activity status information as the device ID of the representative user device 3. Are stored in the information storage unit 23 and also stored in the information storage unit 23 as entries related to other user devices 3 of the activity group.

  The user device 3 that is the acquisition destination of the activity status information is the representative user device 3 or the user device 3 that is not classified into the activity group. As described above, the user device 3 that has received the instruction to suppress the specific process related to the activity status information from the server 5 does not become the acquisition destination of the activity status information. The information acquisition unit 21 may voluntarily collect the device ID and the user activity status information from each target user device 3, or information transmitted spontaneously by each target user device 3. May be obtained.

  The information storage unit 23 stores the device ID acquired by the information acquisition unit 21 and the user activity status information in an associated state. FIG. 4 is a diagram illustrating an example of information stored in the information storage unit 23. FIG. 4 shows an example in which the activity status information of the user indicates date and time, movement status, position, and place attribute. According to the example of FIG. 4, the device ID, date / time, movement status, position, and location attribute are stored in the information storage unit 23 in association with each other.

  The information storage unit 23 may hold a history of some information indicated by the user activity status information. For example, the information storage unit 23 may hold a history of the user activity status indicated by the activity status information. In this case, the information storage unit 23 can also be called a history holding unit.

  The calculation unit 22 calculates the similarity of the activity status between the users based on the user activity status information acquired by the information acquisition unit 21. The similarity of the activity status calculated by the calculation unit 22 indicates the high possibility that the users are performing the same activity. For example, the calculation unit 22 calculates the similarity as follows. However, the method for calculating the similarity of the activity status between users is not limited to the following example.

  The calculation unit 22 calculates the degree of similarity of activity status between users using the degree of coincidence of activity status history between users held in the information storage unit 23. For example, the calculation unit 22 acquires a history of activity statuses related to two users for a predetermined period from the information storage unit 23, and sets the ratio of the number of activity statuses to the number of activity statuses for each user to match the activity status history. Calculate as degrees. The calculation unit 22 can set the calculated coincidence as the similarity.

  In addition, when the activity status information indicates at least one of the user position and the user location attribute, the calculation unit 22 adds the closeness of the distance between users calculated from the user location in addition to the degree of coincidence of the activity status history described above. Furthermore, the similarity of the activity status between users may be calculated by further using at least one of the index of difficulty of change of the user activity status determined in advance with respect to the user location attribute. Since users who are located close to each other and are in places where the activity status is difficult to change are likely to be performing the same activity, the calculation unit 22 sets a high degree of similarity between such users. To do.

  FIG. 5 is a diagram illustrating an example of an index corresponding to a location attribute. The calculation unit 22 may hold in advance an index (weight) that makes it difficult for the user's activity status to change with respect to the user's location attribute as shown in FIG. In the example of FIG. 5, a large weight (index value) is set for the location attribute where the user's activity status hardly changes.

  The classification unit 24 classifies the plurality of user devices 3 into at least one activity group by grouping users based on the similarity calculated by the calculation unit 22. For example, the classification unit 24 uses the reciprocal of the degree of similarity calculated by the calculation unit 22 as distance information to cluster the users, and groups users who are within a predetermined distance from the center of each cluster. The user devices 3 of the users are classified into one activity group. For the clustering, a well-known cluster analysis method such as a K-average method or an NN (Nearest Neighbor) method is used. However, this embodiment does not limit the clustering method itself. Thereby, the user apparatus 3 of the user who exists outside the predetermined distance from the center of each cluster is not classified into the activity group. Thereby, among all the user apparatuses 3, the user apparatus 3 which belongs to an activity group and the user apparatus 3 which does not belong to an activity group may be mixed.

  The selection unit 25 selects at least one representative user device 3 from among the plurality of user devices 3 belonging to the activity group. The selection unit 25 may select a plurality of representative user devices 3. In the present embodiment, the selection method of the representative user device 3 by the selection unit 25 is not limited. The selection unit 25 may select the representative user device 3 at random, or may select the user device 3 closest to the center of the cluster associated with the activity group as the representative user device 3. Further, the selection unit 25 may select the user device 3 with the smallest change in activity status and the user device 3 with the largest weight of the location attribute.

  The classification unit 24 and the selection unit 25 may hold the results of classification into activity groups and selection of the representative user device 3 as data shown in FIG. FIG. 6 is a diagram illustrating an example of information on activity groups. According to FIG. 6, the device ID, the activity group ID, and the representative information are held in association with each other. The activity group ID is information that can identify an activity group. Also, “YES” is set in the representative information of the entry including the device ID of the representative user device 3, and “NO” is set in the representative information of the entry including the device ID of the user device 3 that is not the representative. Yes. Information on such activity groups is also referred to by the information acquisition unit 21 described above.

  The instruction unit 26 instructs the other user devices 3 belonging to the activity group other than the representative user device 3 selected by the selection unit 25 to suppress the specific process related to the activity status information. For example, the instruction unit 26 transmits a message indicating an instruction to suppress the specific process to the target user device 3. The instruction unit 26 may disassemble the activity group at an arbitrary timing, and notify the user device 3 that has instructed the suppression of the specific process of the cancellation of the suppression of the specific process. Furthermore, the instruction unit 26 may instruct another processing unit to reclassify into an activity group at an arbitrary timing.

[Operation example]
Hereinafter, the activity status processing method in the first embodiment will be described with reference to FIG. FIG. 7 is a sequence chart showing an operation example of the activity status collection system 1 in the first embodiment.

  First, it is assumed that the user devices 3 (# 1) to 3 (#n) shown in FIG. 7 are not yet classified into the activity group. At this time, each user apparatus 3 identifies the user's activity status using existing technology based on the acquired data, and generates activity status information indicating at least the identified activity status (S31). At this time, each user apparatus 3 may include date and time information, position information, location attributes, and the like in the activity status information.

  Each user device 3 transmits the generated activity status information to the server 5 together with the device ID of each user device 3 (S32).

  The server 5 acquires the activity status information and the device ID transmitted from each user device 3 (S51). The server 5 confirms whether or not the user device 3 specified by the device ID is a representative of the activity group. Here, since the user device 3 specified by the device ID is not yet classified into the activity group, the server 5 holds the device ID and the activity status information in association with each other.

  The server 5 calculates the similarity of the activity status between the users based on the plurality of activity status information acquired in (S51) (S52). For example, the server 5 calculates the similarity of the activity status for all combinations of two users among all users of the user devices 3 (# 1) to 3 (#n). Further, the server 5 calculates the ratio of the number of the same activity status to the number of the activity status for each user based on the history of the activity status of the two users for a predetermined period, and this ratio is used as the similarity between the two users. Also good. Further, in addition to the ratio, the server 5 is provided with an index of difficulty in changing the user activity state that is determined in advance with respect to the proximity of the distance between users calculated from the user's position and the user's location attribute. You may calculate the similarity degree of the activity condition between users using at least one further.

  The server 5 classifies the users 3 (# 1) to (#n) into at least one activity group by grouping users based on the similarity calculated in (S52) (S53). Specifically, the server 5 uses the reciprocal of the similarity as the distance information, clusters the users, groups the users existing within a predetermined distance from the center of each cluster, and user devices of the grouped users Classify 3 into one activity group. Here, it is assumed that user devices 3 (# 1) to 3 (#n) are classified into one activity group.

  The server 5 selects at least one representative user device 3 from the user devices 3 belonging to each activity group (S54). Here, it is assumed that the user device 3 (# 1) is selected as a representative of the activity group generated in (S53).

  The server 5 relates to the activity status information for the other user devices 3 (# 2) to (#n) belonging to the activity group other than the representative user device 3 (# 1) selected in (S54). An instruction to suppress the specific processing to be performed is given (S55). According to the above example, the user device 3 other than the user device 3 (# 1) is instructed to suppress the specific process.

Each user device 3 that has received an instruction to suppress the specific process other than the representative user device 3 (# 1) stops the activity status information generation process and the activity status information transmission process (S33).
On the other hand, the user apparatus 3 (# 1) that has not received the instruction to suppress the specific process repeats generation (S31) and transmission (S32) of the activity status information at an arbitrary timing.

  When the server 5 acquires the activity status information and the device ID transmitted from the user device 3 (S56), it confirms whether or not the user device 3 (# 1) specified by the device ID is a representative of the activity group. Here, since the user device 3 (# 1) is a representative of the activity group, the server 5 holds the device ID and the activity status information acquired in (S56) in association with each other and also stores the activity status information in the activity Copying is performed for the number of (#n) from other user apparatuses 3 (# 2) belonging to the group (S57). The server 5 holds the copied activity status information together with the device IDs of the other user devices 3 (# 2) to (#n). If the user device 3 identified by the device ID acquired in (S56) is not a representative of the activity group (does not belong to the activity group), (S57) is not performed.

  Although not shown in FIG. 7, the server 5 disassembles the activity group generated in (S53) at any timing after (S57), and instructs generation of activity status information and restart of transmission.

[Operation and Effect of First Embodiment]
As described above, in the first embodiment, the similarity of activity status between users is calculated based on the activity status information obtained from each user device 3, and the users are clustered using the reciprocal of the similarity as distance information. By doing so, users having similar activity statuses are grouped, and the user devices 3 of the grouped users are classified into one activity group. And the specific process regarding the activity status information of the user devices 3 other than the representative user device 3 in the activity group is suppressed.

  As described above, according to the first embodiment, since the similarity of the activity status is quantified, various well-known cluster analysis methods can be used. As a result, users with similar activity status are appropriately grouped together. can do. Therefore, according to 1st Embodiment, since the possibility that the user of the user apparatus 3 which belongs to the same activity group will perform the same activity increases, the activity status information of the representative user apparatus 3 is used as another user apparatus in the same activity group. Even if it is used as the activity status information 3, it is possible to prevent a decrease in accuracy of the activity status of each user apparatus 3.

  Furthermore, in the user devices 3 other than the representative user device 3 belonging to the same activity group, the generation processing and transmission processing of the activity status information are suppressed, so that the power consumption of the user device 3 can be reduced. The amount of communication from the device 3 to the server 5 can also be reduced.

[Second Embodiment]
In the first embodiment described above, the disassembly timing of the activity group and the release timing of the suppression of the specific process related to the activity status information are arbitrary. The server 5 in the second embodiment determines the dismantling timing of the activity group. Hereinafter, the activity status collection system 1 in the second embodiment will be described with a focus on the contents different from the first embodiment. In the following description, the same contents as those in the first embodiment are omitted as appropriate.

[Processing configuration]
FIG. 8 is a diagram conceptually illustrating a processing configuration example of the server 5 in the second embodiment. As shown in FIG. 8, the server 5 in the second embodiment further includes a timing determination unit 81 in addition to the processing unit in the first embodiment. The timing determination unit 81 is realized, for example, by executing a program stored in the memory 11 by the CPU 10.

  The timing determination unit 81 determines the disassembly timing of the activity group to which the representative user device 3 belongs based on at least the activity status information of the representative user device 3 in the activity status information acquired by the information acquisition unit 21. For example, the timing determination unit 81 sets the activity group so that the disassembly interval becomes longer as the activity status of each user included in the activity group is less likely to change, and the disassembly interval becomes shorter as the activity status is more likely to change. Determine the disassembly timing.

  FIG. 9 is a diagram illustrating an example of an activity group disassembly interval. The timing determination unit 81 may hold a table as shown in the example of FIG. 9 in advance, and may determine the disassembly timing of the activity group by referring to this table. In this case, for example, the timing determination unit 81 specifies an entry that matches the activity status information of the representative user device 3 of each activity group in the table, and sets the disassembly interval of the entry as the disassembly timing of each activity group. be able to. Moreover, the timing determination part 81 determines the disassembly timing based on activity status information for every user apparatus 3, respectively, and determines statistical values, such as those minimum values or an average value, in the disassembly timing of each activity group Also good.

  The timing determination unit 81 further determines a reclassification timing after the disassembly timing based on at least the activity status information of the representative user device 3 in the activity status information acquired by the information acquisition unit 21. May be. For example, the timing determination unit 81 has a longer interval between dismantling and reclassification as the activity status of each user included in the activity group is less likely to change, and a shorter interval as the activity status is more likely to change. Next, the reclassification timing of the activity group is determined.

  FIG. 10 is a diagram illustrating an example of an activity group disassembly interval and a reclassification interval. The timing determination unit 81 may hold a table as illustrated in the example of FIG. 10 in advance, and may determine the disassembly timing and reclassification timing of the activity group by referring to this table. In this case, for example, the timing determination unit 81 specifies an entry that matches the activity status information of the representative user device 3 of each activity group in the table, and sets the disassembly interval and reclassification interval of the entry for each activity group disassembly. Timing and reclassification timing can be set. Further, the timing determination unit 81 determines the reclassification timing based on the activity status information for each user device 3, and determines the statistical value such as the minimum value or the average value as the reclassification timing of each activity group. May be.

  In the above example, the elapsed time (interval) from the timing corresponding to the generation of the activity group is used as the disassembly timing, but an absolute time indicating the time may be used.

  The classification unit 24 disassembles each activity group at the disassembly timing determined by the timing determination unit 81. The dismantling of the activity group is to eliminate the classification state into the activity group, and is realized, for example, by deleting the information regarding the activity group as shown in the example of FIG. The classification unit 24 sets the selection completion time as the generation timing of the activity group for the activity group in which the classification of the user device 3 is completed and the selection of the representative user device 3 is completed by the selection unit 25, and the generation timing thereof The dismantling timing is detected by monitoring the elapsed time from.

  The classification unit 24 performs reclassification into an activity group for the user devices 3 that do not belong to the activity group immediately after the disassembly timing. In addition, when the reclassification timing is determined together with the disassembly timing, the classification unit 24 detects the reclassification timing determined by the timing determination unit 81 by monitoring the elapsed time from the disassembly timing. Reclassify to the activity group again at the reclassification timing.

  When the above absolute time is used for the disassembly timing and the reclassification timing, the classification unit 24 detects the disassembly timing and the reclassification timing based on the time information obtained from the clock of the server 5. May be.

  When the activity group is reclassified by the classification unit 24, the selection unit 25 reselects at least one representative user device from among a plurality of user devices belonging to the reclassified activity group.

  The instruction unit 26 instructs the user devices 3 other than the representative belonging to the activity group dismantled by the classification unit 24 to cancel the suppression of the specific process related to the activity status information. When the instruction unit 26 is reclassified after the activity group is disassembled, the instruction unit 26 applies to user devices other than the representative user device selected by the selection unit 25 in the activity group reclassified by the classification unit 24. Instruct the suppression of specific processing.

[Operation example]
The activity status processing method in the second embodiment will be described below with reference to FIG. FIG. 11 is a sequence chart showing an operation example of the activity status collection system 1 in the second embodiment. In FIG. 11, steps similar to those in FIG. 7 are denoted by the same reference numerals as in FIG.

  When the server 5 selects at least one representative user device 3 for each activity group (S54), the server 5 determines the disassembly timing of each activity group (S111). The server 5 determines the disassembly timing of each activity group by referring to a table as shown in each example of FIG. 9 or 10 based on at least the activity status information of the representative user device 3 of each activity group. To do.

  The server 5 monitors the dismantling timing for each activity group. For example, with respect to each activity group in which the representative user device 3 is selected in (S54), the server 5 monitors whether the elapsed time from the time corresponding to (S54) is the disassembly timing. When the disassembly timing of the activity group arrives (S112), the server 5 instructs the user devices 3 other than the representative belonging to the activity group to cancel the suppression of the specific process (S113).

  The user device 3 other than the representative device that is in the process of suppressing the specific process receives the release instruction and restarts the activity status information generation process and the activity status information transmission process (S114). Thereafter, when the reclassification timing arrives (not shown), the server 5 executes the processing after (S52) again.

[Operations and effects in the second embodiment]
In the second embodiment, the disassembly timing of the activity group is determined based on the user activity status information. In the activity group in which the user's activity is difficult to change, it is set for a long time until it is demolished, and in the activity group in which the user's activity is easily changed, until it is demolished, Set short.

  Therefore, according to the second embodiment, since the classification and disassembly into the activity group is controlled according to the activity status information of the user, the activity status of each user in the activity group is similar while the activity group exists. Therefore, the accuracy of the activity status information of each user in the activity group can be maintained as a whole.

  For example, when an activity group in which user devices 3 on a train are classified is dismantled at a timing corresponding to arrival at a station, a certain amount of time until each user device 3 in the activity group settles on the next activity. May take time. In such a case, even if classification into an activity group is performed immediately after dismantling, there is a possibility that it cannot be classified appropriately. Therefore, in the second embodiment, the reclassification timing of the activity group is determined based on the activity status information of the user, and the user device 3 is classified at the reclassification timing after the activity group is disassembled. Therefore, according to 2nd Embodiment, since it can classify | categorize appropriately also in the above cases, the precision of the activity condition information of each user in an activity group can be maintained as a whole.

[Third Embodiment]
The server 5 in the third embodiment selects at least one representative user device 3 from among the plurality of user devices 3 belonging to the activity group based on the state information of the user device 3. Hereinafter, the activity status collection system 1 according to the third embodiment will be described with a focus on the content different from the first embodiment and the second embodiment. In the following description, the same contents as those in the first embodiment and the second embodiment are omitted as appropriate.
[Processing configuration]
The server 5 in the third embodiment has the same processing configuration as that of the first embodiment shown in FIG. 3 or the second embodiment shown in FIG.

  The information acquisition unit 21 further acquires state information of each user device 3. The state information of the user device 3 is information regarding the user device 3 that affects the accuracy of the user activity status information, or information indicating the accuracy. For example, the status information is related to the user device 3 such as the remaining battery level, CPU usage rate, memory usage, number of application activations, communication speed, etc. The information includes information on sensors and GPS mounted on the user device 3 such as the usage status of the device 3, sensor accuracy, sampling frequency, number of satellites used, and the like. The information acquisition unit 21 can also be called a state information acquisition unit. The information acquisition unit 21 may acquire the state information of the user device 3 only from the user device 3 classified into the activity group.

  The selection unit 25 selects at least one representative user device 3 suitable as a representative of the activity group from among the plurality of user devices 3 belonging to the activity group based on the state information acquired by the information acquisition unit 21. The user device 3 suitable as a representative of the activity group is a user device 3 that can continuously generate and transmit highly accurate activity status information. Therefore, for example, the user device 3 having indexes indicated by the state information, such as a low battery remaining amount, a high CPU usage rate, a large memory usage amount, a large number of application activations, and a low sensor accuracy, is representative. It becomes difficult to be chosen.

  The selection unit 25 calculates an evaluation value for each user device 3 from the state information, and selects the user device 3 having the highest evaluation value as a representative. For example, when the status information indicates the remaining battery level and the number of application activations, the selection unit 25 may calculate the evaluation value based on the product of the remaining battery level and the inverse of the number of application activations. The evaluation value of the user apparatus 3 for the remaining battery level (80%) and the number of application activations (5) is calculated as 16 (= 80 * 1/5), and the remaining battery level (30%) and the number of application activations (3) The evaluation value of the user device 3 is calculated as 10 (= 30 * 1/3), and the evaluation value of the user device 3 for the remaining battery level (20%) and the number of application activations (2) is 10 (= 20 *). 1/2). In this case, the selection unit 25 selects the user devices 3 having the remaining battery level (80%) and the number of application activations (5) as representatives.

[Operation example]
The activity status processing method according to the third embodiment is the same method as the first embodiment shown in FIG. 7 or the second embodiment shown in FIG. Each user device 3 further transmits state information (S32), and the server 5 further acquires state information from each user device 3 (S51). In (S54), the server 5 selects the representative user device 3 using the state information. However, the server 5 may acquire the state information of the user device 3 only from the user device 3 classified into the activity group. In this case, after the completion of (S53), the server 5 may acquire the state information of the user device 3 only from the user device 3 classified into the activity group.

[Operation and Effect of Third Embodiment]
A certain amount of processing load and communication load are applied to the generation and transmission of user activity status information. Therefore, when the same user device 3 is always selected as a representative, the load is concentrated on the user device 3, and it may be difficult to operate only a specific user device 3, or battery consumption may be severe. is there. Moreover, since device performance and sensor performance differ for every user apparatus 3, there exists a possibility that the identification accuracy of an activity condition may differ. Therefore, as described above, in the third embodiment, the state information of each user device 3 is collected, and the representative user device 3 is selected based on the state information. Therefore, according to the third embodiment, the user device 3 that can continuously generate and transmit high-accuracy activity status information can be selected as a representative, and consequently the activity status of each user device 3 in the activity group. Information can be acquired accurately and reliably.

[Fourth Embodiment]
The server 5 in the fourth embodiment excludes the specific user device 3 from the classification target for the activity group in advance. Hereinafter, the activity status collection system 1 according to the fourth embodiment will be described focusing on the contents different from the first embodiment to the third embodiment. In the following description, the same contents as those in the first to third embodiments are omitted as appropriate.
[Processing configuration]
The server 5 in the fourth embodiment has the same processing configuration as that of the first embodiment shown in FIG. 3 or the second embodiment shown in FIG.

  The classification unit 24 acquires filter information including a device ID (identification information) of at least one user device 3 among the plurality of user devices 3, and classifies the user device 3 specified by the filter information into an activity group. Exclude from the target. The filter information may be stored in advance by the classification unit 24 or may be acquired from another device. The filter information may include the device ID of the user device 3 to be classified instead of the device ID of the user device 3 to be excluded. In this case, the classification unit 24 classifies only the user device 3 specified by the filter information into the activity group.

[Operation example]
The activity status processing method in the fourth embodiment is the same method as the first embodiment shown in FIG. 7 or the second embodiment shown in FIG. In (S53), the server 5 classifies the user devices 3 other than the user device 3 indicated by the filter information into activity groups.

[Operation and Effect of Fourth Embodiment]
Sales representatives who are active outside the office and frequently move often use a variety of transportation such as trains, taxis, and buses, and sometimes move on foot. Such a user device 3 of a specific user is unlikely to have a user device 3 of another person who performs a similar activity. Therefore, in the fourth embodiment, by setting the user device 3 of such a specific user in the filter information in advance, the user device 3 is excluded from the classification target for the activity group. Therefore, according to the fourth embodiment, the similarity of each user in the activity group can be increased, and consequently the accuracy of the activity status information of each user in the activity group can be maintained as a whole.

  Examples will be given below to describe the above-described embodiment in more detail. However, the above-described embodiments are not limited to the following examples.

  For example, the user device 3 includes an acceleration sensor, and uses the sensor data obtained by the acceleration sensor to identify the movement status of the user. Moreover, the user apparatus 3 has a GPS receiver, and acquires position information indicated by longitude and latitude. Moreover, the user apparatus 3 acquires map information from another server etc., and acquires a location attribute from the map information and its position information. The user device 3 transmits activity status information indicating the user's movement status, date and time, position (latitude and longitude), and location attribute to the server 5. The user device 3 may display the activity status information.

It is assumed that the information acquisition unit 21 of the server 5 stores the activity status information shown in FIG. 4 by acquiring such activity status information. Based on the activity status information as shown in FIG. 4, the information acquisition unit 21 calculates the similarity of the activity status between users using the following equation. In the following formula, “*” indicates multiplication, and exp () indicates an exponential function. a is a coefficient. a may be set to 1.
Activity status similarity = g * h * f
g = exp (−a * distance between users)
h = Location attribute weight of one user * Location attribute weight of the other user f = Activity status history matching degree

  Here, a method of calculating the similarity between the device ID (0001) and the device ID (0002) will be described with reference to the examples of FIGS. FIG. 12 is a diagram illustrating an example of activity status history of the user device 3 (0001) and the user device 3 (0002).

The calculation unit 22 calculates the inter-user distance as 0.1 from the position information shown in FIG. Here, when a in the above formula is 1, the calculation unit 22 calculates g as 0.9.
Since both are present in the store from the location attribute shown in FIG. 4, the calculation unit 22 uses the weight 0.5 shown in FIG. 5 and sets h to 0.25 (= 0.5 * 0. 5) is calculated.
The calculating unit 22 calculates the ratio f (= 0.83 = 5/6) of the number of matching activities (5) to the number of activities (6) in the illustrated period from the activity status history shown in FIG.

  As a result, the calculation unit 22 sets the similarity of the activity status between the user with the device ID (0001) and the user with the device ID (0002) to 0.19 (= 0.9 * 0.25 * 0.83). calculate. The calculation unit 22 calculates the similarity in this way for each combination of two users.

  The classification unit 24 uses the reciprocal of the degree of similarity corresponding to each combination between users as distance information, clusters the users, groups users existing within a predetermined distance from the center of each cluster, and performs grouping. The user device 3 of the user is classified into one activity group. For the clustering, a well-known K-average method, NN method, or the like is used. The classification unit 24 may not classify the user devices 3 of users who exist outside a predetermined distance from the center of each cluster into activity groups.

[Modification]
In the above-described first to fourth embodiments, the user device 3 generates the activity status information. However, the user device 3 may acquire some data of the activity status information and some data that is the basis of the activity status information, and transmit this data to the server 5. For example, the user device 3 may send the user's position and vibration data (acceleration data) to the server 5, and the server 5 may acquire the location attribute from the user's position and identify the movement status from the vibration data. . In addition, the user device 3 may transmit the data to a computer other than the server 5, the computer may identify the movement status based on information from the user device 3, and may transmit the identified movement status to the server 5. . Each embodiment will not restrict | limit the main body which produces | generates activity condition information, if the user apparatus 3 transmits some data and the server 5 can acquire activity condition information.

  The server 5 may further include an information providing unit that provides the activity status information of each user stored in the information storage unit 23 to each user device 3. In this case, the information providing unit transmits display data indicating the current activity status of the user to each user device 3. The information providing unit may transmit display data indicating the activity status of all users or other users to each user device 3.

  FIG. 13A and FIG. 13B are diagrams showing examples of displaying the activity status. The user device 3 displays a screen as shown in the example of FIG. 13A or 13B based on the display data provided from the information providing unit. In the example of FIG. 13A, the current activity status information (movement status, location attribute, time) of the user of the user device 3 is displayed, and the position of the user is shown on the map. In the example of FIG. 13B, the information stored in the information storage unit 23 is displayed in a list, the activity status information of the user of the user device 3 selected therein is displayed in a balloon, and the positions of a plurality of users are shown on the map. It is. Also, a history of user activity status information may be displayed.

  In addition, in the some flowchart used by the above-mentioned description, although several process (process) is described in order, the execution order of the process performed by each embodiment is not restrict | limited to the order of the description. In each embodiment, the order of the illustrated steps can be changed within a range that does not hinder the contents. Moreover, each above-mentioned embodiment and each modification can be combined in the range with which the content does not conflict.

1 Activity Status Collection System 2 Network 3 User Device 5 Activity Status Collection Server (Server)
10 CPU
DESCRIPTION OF SYMBOLS 11 Memory 21 Information acquisition part 22 Calculation part 23 Information storage part 24 Classification part 25 Selection part 26 Instruction part 81 Timing determination part 100 Activity status processing apparatus 101 Classification part 102 Selection part 103 Instruction part

Claims (17)

By grouping users having similar activity statuses based on a plurality of activity status information indicating at least the activity statuses of a plurality of users identified using data obtained by the plurality of user devices, A classifier for classifying into at least one activity group;
A selection unit for selecting at least one representative user device from a plurality of user devices belonging to the activity group;
An instruction unit that instructs other user devices belonging to the activity group other than the representative user device selected by the selection unit to suppress a specific process related to the activity status information;
An activity status processing apparatus. A timing determination unit that determines disassembly timing of an activity group to which the representative user device belongs, based on at least activity state information of the representative user device;
Further comprising
The classification unit dismantles an activity group to which the representative user device belongs at a disassembly timing determined by the timing determination unit,
The instructing unit instructs a user device other than the representative user device belonging to the activity group to be dismantled to cancel the suppression of the specific process.
The activity status processing apparatus according to claim 1. The timing determination unit further determines reclassification timing after the disassembly timing based on at least the activity status information of the representative user device,
The classification unit performs the classification again at the reclassification timing determined by the timing determination unit,
The selection unit reselects at least one representative user device from among a plurality of user devices belonging to the activity group reclassified by the classification unit,
The instruction unit instructs the user device other than the representative user device selected by the selection unit in the activity group to be reclassified by the classification unit to suppress specific processing.
The activity status processing apparatus according to claim 2. A status information acquisition unit for acquiring status information of each user device;
Further comprising
The selection unit selects at least one representative user device from a plurality of user devices belonging to the activity group based on the state information acquired by the state information acquisition unit;
The activity status processing apparatus according to any one of claims 1 to 3. The classification unit obtains filter information including identification information of at least one user device among the plurality of user devices, and excludes the user device specified by the filter information from the target to be classified into the activity group;
The activity status processing apparatus according to any one of claims 1 to 4. The classification unit includes:
Based on the plurality of activity status information, including a calculation unit for calculating the similarity of activity status between users,
Classifying the plurality of user devices into the at least one activity group by clustering the plurality of users using the reciprocal of the similarity calculated by the calculation unit as distance information;
The activity status processing apparatus according to any one of claims 1 to 5. A history holding unit for holding a history of the activity status of the plurality of users indicated by the plurality of activity status information;
Further comprising
The calculation unit calculates the degree of similarity of activity status between users using the degree of coincidence of activity status history between users held in the history holding unit.
The activity status processing apparatus according to claim 6. The plurality of activity status information further indicates at least one of a position of the plurality of users and a location attribute of the plurality of users,
The calculation unit includes a proximity index between users calculated using the location information of the plurality of users, and an index of difficulty in changing a user's activity status determined in advance with respect to the location attributes of the plurality of users. Using at least one of them to calculate the similarity of activity status between users;
The activity status processing apparatus according to claim 7. In an activity status processing method executed by at least one computer,
By grouping users having similar activity statuses based on a plurality of activity status information indicating at least the activity statuses of a plurality of users identified using data obtained by the plurality of user devices, Into at least one activity group,
Selecting at least one representative user device from a plurality of user devices belonging to the activity group;
Instructing other user devices belonging to the activity group other than the selected representative user device to suppress specific processing related to the activity status information.
Activity status processing method. Determining the disassembly timing of the activity group to which the representative user device belongs based on at least the activity status information of the representative user device;
Dismantling the activity group to which the representative user device belongs at the determined dismantling timing;
Instructing user devices other than the representative user device belonging to the dismantled activity group to cancel the suppression of the specific process,
The activity status processing method according to claim 9, further comprising: Further determining a reclassification timing after the disassembly timing based on at least the activity status information of the representative user device;
The classification is performed again at the determined reclassification timing,
Reselecting at least one representative user device from a plurality of user devices belonging to the reclassified activity group;
Instructing user devices other than the reselected representative user device in the reclassified activity group to suppress specific processing.
The activity status processing method according to claim 10, further comprising: Obtaining the status information of each user device;
Further including
The selection of the representative user device selects at least one representative user device from a plurality of user devices belonging to the activity group based on the acquired state information.
The activity status processing method according to any one of claims 9 to 11. Obtaining filter information including identification information of at least one user device among the plurality of user devices;
Excluding the user device identified by the filter information from the target to be classified into the activity group;
The activity status processing method according to claim 9, further comprising: Based on the plurality of activity status information, calculate the similarity of the activity status between users,
Clustering the plurality of users using the reciprocal of the calculated similarity as distance information;
Further including
The classification classifies the plurality of user devices into the at least one activity group based on the result of the clustering.
The activity status processing method according to any one of claims 9 to 13. The history holding unit holds the history of the activity status of the plurality of users indicated by the plurality of activity status information.
Further including
The calculation of the degree of similarity calculates the degree of similarity of the activity status between users using the degree of coincidence of the activity status history between users held in the history holding unit.
The activity status processing method according to claim 14. The plurality of activity status information further indicates at least one of a position of the plurality of users and a location attribute of the plurality of users,
The calculation of the similarity is based on the closeness of the distance between users calculated using the location information of the plurality of users, and the difficulty of changing the user's activity status determined in advance with respect to the location attributes of the plurality of users. Using at least one of the indicators to calculate the similarity of activity status between users;
The activity status processing method according to claim 15.   The program which makes an at least 1 computer perform the activity condition processing method of any one of Claim 9 to 16.

Images