JP6638265B2 - Information providing device, program - Google Patents

Description

The present invention relates to an information providing technique, and more particularly, to an information providing apparatus and a program for providing information on a user group.

  There is a system that detects position information of a terminal device held by a user and provides appropriate information to the terminal device according to the position information. For example, it is determined whether or not the current location of the mobile terminal is the user's living area, and local recommendation information is provided accordingly (for example, see Patent Literature 1). Further, position information of a plurality of portable terminal devices is detected, portable terminals having common position information over a predetermined time interval are grouped, and optimal information is provided to the group (for example, see Patent Document 2). .

JP 2009-210400 A JP 2004-320217 A

  The technology described in Patent Literature 2 groups terminal devices (users) based on the position information of the terminal devices, extracts preference information common to the terminal devices (users) in the group, and matches information that matches the preference information. Is provided to each terminal device. That is, information matching the common part of the tastes and interests of the users in the group is provided. However, the decision making in the actual group is not always performed based on the common part of the taste and interest of each user. In many cases, group decisions are made based on the preferences and interests of a particular user, which can be said to be the leader of the group. For example, when a restaurant to be entered by five users is determined, a restaurant is often determined based on a specific user's preference instead of a restaurant based on a common part of the five people's preferences regarding cooking.

  Which users have the greatest influence is on a case-by-case basis; for example, the oldest person in the group, the most knowledgeable about cooking, and the most financially beneficial are the leaders There are cases. In addition, a group having a strong influence is not limited to one person, and a plurality of people may have a strong influence. As described above, in a real group, the influence of each user on decision making is not equal, and is often biased toward a specific user. However, the power relationship (influence) between the users in the group has not been sufficiently considered, and it has been difficult to provide useful information to the group with high accuracy.

  The present invention has been made in view of such a situation, and an object of the present invention is to provide a technique for providing information in consideration of the influence of users in a group.

In order to solve the above-described problems, an information providing apparatus according to an aspect of the present invention obtains position information of a subject that is a user or a terminal device, and based on time-series data of position information related to a plurality of subjects, The influence of the subjects in the group on each of the subjects included in the group formed by the group forming unit based on the time series data of the location information based on the group forming unit forming the group including the subject. Is calculated as a feature action amount indicating a degree of matching with a predetermined action pattern that is determined to be high, and a point calculation unit that calculates points for each subject based on the feature action amount, and a point calculated by the point calculation unit is high. An item selection unit that selects an item to be presented to the group so that the subject is prioritized. Predetermined behavioral patterns in point calculating unit, when the group is joined after divided into a plurality, a pattern mainly from the time division in the direction of movement after the confluence is moving.

Another embodiment of the present invention also relates to an information providing device . This apparatus obtains position information of a subject which is a user or a terminal device, and, based on time-series data of position information relating to a plurality of subjects, forms a group including a plurality of subjects based on a time series data; Based on the time series data, the degree of matching of a predetermined behavior pattern in which the influence of the subject in the group is considered to be high for each of the plurality of subjects included in the group formed by the group forming unit is indicated. A point calculation unit that calculates a characteristic action amount and calculates a point for each subject based on the characteristic action amount, and an item to be presented to the group so that the subject with a higher point calculated by the point calculation unit is prioritized. And an item selection unit that selects The predetermined action pattern in the point calculation unit is a pattern in which, when the subject stops, the other subjects also stop following the stop.

Yet another embodiment of the present invention also relates to an information providing device . This apparatus obtains position information of a subject which is a user or a terminal device, and, based on time-series data of position information relating to a plurality of subjects, forms a group including a plurality of subjects based on a time series data; Based on the time series data, the degree of matching of a predetermined behavior pattern in which the influence of the subject in the group is considered to be high for each of the plurality of subjects included in the group formed by the group forming unit is indicated. A point calculation unit that calculates a characteristic action amount and calculates a point for each subject based on the characteristic action amount, and an item to be presented to the group so that the subject with a higher point calculated by the point calculation unit is prioritized. And an item selection unit that selects The point calculation unit calculates a point for one subject included in the group based on previously calculated points corresponding to other subjects included in the group.
Yet another embodiment of the present invention also relates to an information providing device. The apparatus includes a group forming unit configured to acquire position information of a subject that is a user or a terminal device, and form a group including a plurality of subjects based on time-series data of the position information related to the plurality of subjects. The initial value of the point of each subject at the time when the group was formed in the part is set to zero, and for a predetermined period after the time, based on the time-series data of the position information, the position information of one subject included in the group, It is determined whether or not the position information of another subject included in the group matches any of a plurality of types of patterns relating to the position information of two or more subjects. A point calculation unit that calculates a point for each subject included in the group by adding a feature action amount according to the type to a point of one subject, and a point calculation unit An item selection unit that selects an item to be presented to the group based on a value obtained by adding the preference level for the item of each subject or the preference level for the item attribute while weighting the points of each subject calculated in the above. , Is provided.

  Note that any combination of the above-described components and any conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, and the like are also effective as embodiments of the present invention.

  ADVANTAGE OF THE INVENTION According to this invention, the information which considered the influence of the user in a group can be provided.

FIG. 1 is a diagram illustrating a configuration of an information providing system according to a first embodiment of the present invention. FIG. 2 is a diagram illustrating a configuration of the information providing apparatus of FIG. 1. 3 is a flowchart illustrating an information providing procedure performed by the information providing apparatus of FIG. 2. FIG. 3 is a diagram illustrating a data structure of a database stored in a position information management unit in FIG. 2. FIG. 3 is a diagram illustrating a data structure stored in a group forming unit in FIG. 2. FIG. 3 is a diagram illustrating a data structure of a database stored in a group information processing unit in FIG. 2. 3 is a flowchart illustrating a procedure for updating a leader point by the information providing apparatus of FIG. 2. FIG. 3 is a diagram illustrating an outline of a process in a point calculation unit in FIG. 2. FIGS. 9A and 9B are diagrams illustrating an outline of another process in the point calculation unit in FIG. FIG. 3 is a diagram illustrating a data structure of a database stored in a user information storage unit in FIG. 2. FIG. 3 is a diagram illustrating a data structure of a database stored in a provision information storage unit in FIG. 2. FIG. 3 is a diagram illustrating a data structure of information transmitted from a transmission unit in FIG. 2. FIG. 2 is a diagram illustrating a configuration of a terminal device of FIG. 1. FIG. 14 is a diagram illustrating a screen displayed on a display unit 52 in FIG. 13. FIG. 6 is a diagram illustrating a configuration of an information providing apparatus according to a second embodiment of the present invention. FIG. 16 is a diagram illustrating a data structure of a database stored in a user information storage unit in FIG. 15. FIG. 16 is a diagram illustrating a data structure of another database stored in the user information storage unit in FIG. 15. FIG. 14 is a diagram illustrating a data structure of another database stored in the user information storage unit according to the third embodiment of the present invention. FIG. 14 is a diagram illustrating a configuration of an information providing apparatus according to a fifth embodiment of the present invention. FIG. 14 is a diagram illustrating a screen displayed on a display unit according to a fifth embodiment of the present invention. FIG. 15 is a diagram illustrating another screen displayed on the display unit according to the fifth embodiment of the present invention. FIG. 15 is a diagram illustrating another screen displayed on the display unit according to the fifth embodiment of the present invention. FIG. 14 is a diagram illustrating a data structure of a database stored in a point storage unit according to Embodiment 5 of the present invention. FIG. 15 is a diagram illustrating another screen displayed on the display unit according to the fifth embodiment of the present invention. FIG. 15 is a diagram illustrating another screen displayed on the display unit according to the fifth embodiment of the present invention. FIG. 15 is a diagram illustrating another screen displayed on the display unit according to the fifth embodiment of the present invention. FIG. 14 is a diagram illustrating a data structure of a database stored in a group information processing unit according to a fifth embodiment of the present invention. FIG. 14 is a diagram illustrating a data structure of a database stored in a group information processing unit according to a sixth embodiment of the present invention. 15 is a flowchart illustrating a procedure for correcting a leader point by the information providing apparatus according to the sixth embodiment of the present invention.

(Example 1)
Before describing the present invention in detail, an overview will first be given. Embodiment 1 Embodiment 1 of the present invention relates to an information providing system including a plurality of terminal devices and an information providing device connectable thereto. A group is formed by the users having each of the plurality of terminal devices acting together. When making a decision in a group, for example, when deciding a restaurant to enter for a meal, the preference of a user having a large influence in the group may be reflected. Therefore, the information providing apparatus according to the present embodiment forms a group using the position information of a plurality of terminal devices, and estimates the influence of each user in the group. Further, the information providing device provides optimal information to the group based on the estimated influence. In this way, information based on the position information is provided to the group with sufficient accuracy.

  Next, information provided in the present embodiment will be described. The information providing device provides information on the item to the group. The “item” is a term indicating the general object used by the user, and is, for example, a store, a product, a service, a Web page, digital content, or the like. In the present embodiment, the item is a store such as a restaurant. That is, in the present embodiment, the information providing device provides the group with information on stores such as restaurants. The information providing device may provide the information on the item in response to the information providing request from the group, or may provide the information to the group at an appropriate timing regardless of the presence or absence of the information providing request. . The above is the summary description.

  FIG. 1 shows a configuration of an information providing system 100 according to the first embodiment of the present invention. The information providing system 100 includes an information providing device 10, a network 12, a base station device 14, and a first terminal device 16a, a second terminal device 16b, a third terminal device 16c, and an N-th terminal device 16n collectively referred to as a terminal device 16. Including.

  The terminal device 16 corresponds to a wireless communication system and connects to a base station device 14 described later. Examples of the wireless communication system include a mobile phone system, a wireless LAN (Local Area Network) system, a business wireless system, and the like, but are not limited thereto. The terminal device 16 is a mobile phone, a smartphone, a tablet terminal, a notebook PC (Personal Computer), a wireless device, or the like, and is configured to be easily carried by a user. The terminal device 16 has a function of acquiring position information (latitude, longitude, and the like) by a GPS (Global Positioning System), a wireless LAN, or the like.

  At one end, the base station apparatus 14 corresponds to the same wireless communication system as the terminal apparatus 16 and can connect to the terminal apparatus 16. The base station device 14 connects the network 12 at the other end. The network 12 is connected to a base station device 14. The network 12 may be anything, for example, a wired network, a wireless network, or a combination thereof.

  The information providing device 10 is connected to a network 12. With such a configuration, the terminal device 16 can communicate with the information providing device 10 via the base station device 14 and the network 12. Therefore, the terminal device 16 transmits the acquired position information to the information providing device 10 via the base station device 14 and the network 12. The information providing device 10 forms a group of the terminal devices 16 (users) based on the received position information, and selects information suitable for the group. The information providing device 10 transmits information to the terminal device 16 via the network 12 and the base station device 14. Although the processing in the information providing apparatus 10 will be described later, in the following, the case where the term "subject" meaning the user or / and the terminal device 16 and the term "subject identifier" meaning the user ID or / and the terminal ID is used There is.

  FIG. 2 shows a configuration of the information providing apparatus 10. The information providing device 10 includes a receiving unit 20, a position information managing unit 22, a group information processing unit 24, an item selecting unit 26, a user information storing unit 28, a provided information storing unit 30, a transmitting unit 32, and a control unit 34. The group information processing unit 24 includes a group formation unit 40 and a point calculation unit 42.

  This configuration can be realized in terms of hardware by the CPU, memory, or other LSI of any computer, and can be realized in terms of software by a program loaded in the memory. The functional blocks are drawn. Therefore, it will be understood by those skilled in the art that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof. For example, a distributed processing system in which the functions of each unit constituting the information providing apparatus 10 are shared by a plurality of computers may be constructed. In this case, distributed processing may be performed in which different computers are assigned to respective functions, such as assigning the group information processor 24 to the first computer and assigning the item selector 26 to the second computer. May be configured to have all the processing units of the information providing apparatus 10.

  The operation of the information providing system 100 having the above configuration will be described. FIG. 3 is a flowchart illustrating an information providing procedure by the information providing apparatus 10. To briefly describe the outline of the process, first, the group forming unit 40 forms a group (S100). Next, the point calculation unit 42 calculates a leader point (S110), and normalizes the leader point (S120). Next, the item selection unit 26 calculates a preference level for the item attribute of the group (S130), and calculates a recommendation level (recommendation score) of the item (store) (S140). Then, the item selecting unit 26 selects the items (stores) in descending order of the recommendation level (S150). The details of each step will be described below together with the description of each unit shown in FIG.

  The receiving unit 20 periodically receives the position information, the terminal ID, and the date and time information from each terminal device 16 via the base station device 14 and the network 12. The receiving unit 20 outputs the position information, the terminal ID, and the date and time information to the position information management unit 22. The position information management unit 22 inputs the position information, the terminal ID, and the date and time information from the reception unit 20. The position information management unit 22 stores the position information, the terminal ID, and the date and time information for each terminal device 16 in association with each other. Note that the reception unit 20 may not be configured to receive the date and time information from the terminal device 16 and the location information management unit 22 may be configured to generate the date and time information based on the location information and the date and time when the terminal ID was acquired. FIG. 4 shows the data structure of the database stored in the position information management unit 22. Position information corresponding to various date and time information is stored for each of the plurality of terminal IDs. This corresponds to storing time-series data of position information relating to a plurality of subjects. As described above, the subject indicates the user or the terminal device 16. The position information includes at least information on latitude and longitude, but may also include information on altitude, moving direction, acceleration, and the like.

  Next, the processing of each unit will be described with reference to the flowchart of FIG. In step S100, the group forming unit 40 forms a group including a plurality of subjects based on the time-series data of the location information on the plurality of subjects stored in the location information management unit 22. A group is equivalent to grouping entities that are presumed to be acting (moving) together. That is, the group forming unit 40 determines which terminal device 16 belongs to the group, and forms a group. More specifically, the group forming unit 40 uses the data stored in the position information management unit 22 to determine whether a plurality of terminal devices 16 exist within a predetermined distance for a predetermined time or more. , The terminal devices 16 are determined to be one group. The predetermined time is set, for example, to 1 to 10 minutes. The predetermined distance is set, for example, to 1 to 50 m.

  Here, when the date and time recorded in the location information management unit 22 is slightly different for each terminal ID, it is desirable to correct the latitude and longitude according to the difference between the date and time between the terminal IDs. For example, the date and time of the first terminal device 16a whose terminal ID is “A” is “2015/03/20 10:11:25” and the same corresponding to the second terminal device 16b whose terminal ID is “B” When there is no position information of the date and time, two pieces of position information of the date and time adjacent to “2015/03/2010 10:11:25” are extracted from the position information of the second terminal device 16b. If the two pieces of position information are, for example, “2015/03/20 10:11:00” and “2015/03/20 10:11:26”, the two pieces of position information are interpolated and the second terminal The latitude and longitude of “2015/03/20 10:11:25” of the device 16b are calculated. In this example, a position that internally divides the position at “2015/03/20 10: 111: 00” and the position at “2015/03/20 10:11:26” into 25: 1 is calculated. In addition, the interpolation of the position information is not limited to such linear interpolation, and a non-linear interpolation process using three or more pieces of position information may be performed.

  Here, one method of performing group determination in the group forming unit 40 will be described with reference to FIG. FIG. 5 shows a data structure stored in the group forming unit 40. This is because, during a certain time period (for example, 2015/1/1 10:00 to 10:03), the first terminal device 16a to the eighth terminal device 16h existing in an area of a predetermined range and the respective terminal devices Shows the average distance (m). In addition, the terminal ID of the first terminal device 16a is indicated as “A”, the terminal ID of the second terminal device 16b is indicated as “B”, the terminal ID of the third terminal device 16c is indicated as “C”, The terminal ID of the fourth terminal device 16d is indicated as "D". Similarly, the terminal ID of the eighth terminal device 16h is indicated as “H”. The average distance is calculated by calculating the distance between any two terminal devices 16 present in a certain area at predetermined time intervals (for example, every second) during a predetermined length of time (for example, 3 minutes), It is a value obtained by calculating their average.

  The group forming unit 40 applies a rule to the data when the average distance between the terminals is equal to or less than 10 m (first threshold). As a result, the first terminal device 16a to the fifth terminal device 16e belong to the first group, the sixth terminal device 16f and the seventh terminal device 16g belong to the second group, and the eighth terminal device 16h belongs to the third group. . Note that the group forming unit 40 sets the minimum value of the number of terminals forming one group, and does not have to form a group when the number is less than the number. For example, when the condition of at least three terminal devices 16 is applied, only the first group is formed in the data example shown in FIG. Further, the group forming unit 40 may set the maximum value of the number of terminals forming one group. Further, the group forming unit 40 may calculate the maximum value of the distance instead of the average value of the distance between the terminals in the predetermined period, and form a group using the maximum value. Further, the group forming unit 40 may calculate a variance or a standard deviation indicating the magnitude of the fluctuation of the distance between the terminals, and form a group using the variance or the standard deviation.

  For example, when the average value of the distance between the terminals during the predetermined period is equal to or less than the first threshold value and the standard deviation of the distance between the terminals is equal to or less than the third threshold value, The group may be determined. By using the standard deviation, the probability of erroneously recognizing the terminal device 16 that is not performing a group action as a group is reduced. Further, when the number of candidate terminals forming one group is large to some extent, the group forming unit 40 sets the first threshold value even if the distance between the group candidate terminals exceeds the first threshold value. If the number of distances that exceed the distance is equal to or less than a predetermined number and the maximum value of the distance is equal to or less than the second threshold, processing such as grouping may be performed.

  For example, in the group determination process, when the number of terminal devices 16 that are candidates for a certain group is equal to or more than a predetermined number (for example, seven), the following process is performed. Here, there are ten terminal devices 16 that are candidates for a certain group, and the distance between one terminal device 16 (hereinafter, referred to as “determination target terminal”) and the other nine terminal devices 16 is set. Is the maximum value of 18 m, the second largest distance is 10 m, and the maximum value of the distance between the other nine terminal devices 16 is 9 m. In this case, the number of distances exceeding the first threshold value (for example, 10 m) for group determination is one, this number is equal to or less than a predetermined number (for example, two), and the maximum value of the distance (18 m) Is less than or equal to the second threshold value (for example, 20 m), the group forming unit 40 determines that the determination target terminal also belongs to the group. Note that the group determination process is not limited to the above-described method. For example, the group forming unit 40 applies various known cluster analysis methods to the distance between terminals, and associates the formed clusters with the groups. It may be performed using a process such as causing

  The group forming unit 40 generates a new group ID each time one group is formed, and stores the new group ID in the storage unit in the group information processing unit 24 as a database shown in FIG. FIG. 6 shows the data structure of the database stored in the group information processing unit 24. The database associates a group ID, a terminal ID (user ID), and a leader point (Rp), and is also called a group information table. The leader point is a numerical value obtained by estimating the influence of the user in the group based on the behavior of the user in the group. The greater the value, the greater the influence of the user (terminal device 16). Although not shown in FIG. 6, the date and time when the group was formed (group start date and time) may be recorded in the database. Furthermore, the date and time when the group was canceled (group end date and time) may be recorded in the database. For example, the group ID, the terminal ID (user ID), the leader point (Rp), the group start date and time, and the group end date and time may be stored as a group information table in association with each other.

  FIG. 6 shows a state in which a group having a group ID “G1” and a group having a group ID “G2” are formed. The example shown in this figure shows a state in which “G2” is formed after “G1” is formed and resolved. The group with the group ID “G1” includes a first terminal device 16a, a second terminal device 16b, and a third terminal device 16c. The respective leader points are “20”, “50”, and “30”. The leader point calculation method will be described later. The group with the group ID “G2” includes the first terminal device 16a, the third terminal device 16c, the sixth terminal device 16f, and the seventh terminal device 16g, and is an initial state in which the group is formed. As described above, in the initial state in which the group is formed, the leader point of each user is “0”. Note that the leader point stored in the group information table is updated as needed until one group action ends. Furthermore, after one group action is completed, it is stored in the same value state without being updated. The term "temporary point" may be used to represent a leader point that is in an updated state.

  The point calculation unit 42 estimates the influence of the users in the group on the decision making using the time-series data of the position information stored in the position information management unit 22. Specifically, in step S110, the point calculation unit 42 calculates an influence value (leader point) as an index indicating the influence of each user. In other words, the point calculation unit 42 analyzes the time-series data of the position information of each terminal ID, and determines that each of the plurality of subjects included in the group has a high influence on the subjects in the group. A characteristic action amount indicating the degree of matching with the action pattern is calculated. Further, the point calculation unit 42 calculates or updates a leader point (temporary point) for each subject based on the characteristic action amount. Hereinafter, this processing will be described in detail with reference to the flowchart of FIG.

  The point calculation unit 42 calculates a leader point such that the higher the ranking determined by the position of the subject, the higher the value. Specifically, in step S200, the point calculation unit 42 defines that the subject is located in a predetermined order (order) from the head of the group based on the moving direction of the group as a first predetermined action pattern. Then, the number of times the subject is positioned in a predetermined order is calculated as the characteristic action amount. Typically, the point calculation unit 42 counts the user's preceding frequency. The precedence frequency is the number of times that the user is located at the head of the group with reference to the moving direction of the group.

  The outline of the process in step S200 will be described with reference to FIG. In the example shown in this figure, the first terminal device 16a, the second terminal device 16b, and the third terminal device 16c are moving in the direction of the arrow. The point calculation unit 42 compares the terminal ID and the latitude / longitude information stored in the position information management unit 22 with, for example, the latest and predetermined number (for example, 100 times) of latitude / longitude information from the latest, so that the group Specifies the user located at the head of the traveling direction. In addition, the point calculation unit 42 counts the number of times (preceding frequency) at the preceding position for each terminal ID. Further, the point calculation unit 42 calculates an influence value (leader point) indicating the influence of the user using the preceding frequency. For example, out of the position information measured 100 times, it is assumed that the number of times the third terminal device 16c, the first terminal device 16a, and the second terminal device 16b are at the top is 60 times, 30 times, and 10 times, respectively. In this case, the leader points of the third terminal device 16c, the first terminal device 16a, and the second terminal device 16b are “60”, “30”, and “10”. Alternatively, when the position information of each terminal device 16 is updated every predetermined time (for example, every 5 seconds), the terminal device 16 located at the head of the group is specified at each update timing, and its leader point is set to a predetermined value. Processing such as increasing the number (for example, “1”) may be performed.

  As described above, the characteristic action amount may be used as it is as a leader point, or a value obtained by multiplying the characteristic action amount by a predetermined coefficient may be used as the leader point. Such processing is based on the finding that when a person moves in a group, the user who walks at the head of the group tends to have the greatest influence. In particular, when the group is moving while searching for an appropriate store for a certain purpose, such a tendency is likely to appear more remarkably. The leader point can be said to be an index indicating the influence of the group user when searching and selecting an item such as a store or a product. The point calculation unit 42 adds the calculated leader point (leader point related to the leading frequency) to the leader point of the group information table, and updates the group information table.

  Note that the point calculation unit 42 may count the number of times the user is positioned in a predetermined order, such as the second from the top or the third from the top, without being limited to the case where the user is positioned at the top. Further, the point calculation unit 42 may calculate the characteristic action amount such that the higher the rank of the position from the top, the larger the value. That is, the point calculation unit 42 may set the leader point to a larger value as the position ranking from the head is higher. This corresponds to adding a leader point whose value has been changed according to the order (order) from the top. For example, if it is located at the second position from the head, a leader point such as 50% of the position at the top is added, and if it is located at the third position from the beginning, a leader point such as 20% is added.

  Further, the point calculation unit 42 may determine the position order from the head used for the leader point calculation according to the number of subjects included in the group. For example, when the number of members of the group is small (when the number is less than the first predetermined number), only the first position from the top is considered, and when the number of members of the group is medium (the number is equal to or more than the first predetermined number, and If the number is less than the second predetermined number), the first to second positions are considered. When the number of members in the group is large (when the number is equal to or more than the second predetermined number), the third position from the top is considered.

  The point calculation unit 42 may calculate the leader point using the distance moved at the head of the group instead of the preceding frequency as the characteristic action amount and using the characteristic action amount. For example, within a predetermined time, the first terminal device 16a moves 500 m with the first position, the second terminal device 16b moves 300 m with the first position, and the third terminal device 16c is the first position. It is assumed that the user has moved 100 m in In this situation, assuming that the movement distance in meters is the characteristic action amount, the characteristic action amounts are “500”, “300”, and “100”. The point calculation unit 42 may set a value obtained by multiplying the characteristic action amount by “0.1” as a leader point, and calculate each of the leader points as “50”, “30”, or “10”. That is, the point calculation unit 42 may calculate the leader point such that the longer the distance moved to the head, the larger the value.

  The point calculation unit 42 may calculate the leader point using the length of time during which each terminal device 16 is located at the head of the group as the characteristic action amount, and using the characteristic action amount. For example, in 10 minutes (predetermined time), the time when the first terminal device 16a was at the head is 2 minutes, the time when the second terminal device 16b was at the head is 7 minutes, It is assumed that the time at which the terminal device 16c is located at the top is one minute. In this situation, when the time in minutes is a feature amount, the feature action amounts are “2”, “7”, and “1”. The point calculation unit 42 may set a value obtained by multiplying the characteristic action amount by “10” as a leader point, and calculate each leader point as “20”, “70”, or “10”. That is, the point calculation unit 42 may calculate the leader point such that the longer the time at the head is, the larger the value is. Further, the point calculation unit 42 may calculate the leader point by arbitrarily combining the characteristic action amounts regarding the number of times, time, and distance at the head of the group.

  Next, in step S210, when one group is divided into a plurality of groups and then merges, the point calculation unit 42 determines a pattern in which the subject has moved from the time of division in the direction of movement after the merge into a second predetermined behavior pattern. And the number of times of merging after division is calculated as the characteristic action amount. At that time, the point calculation unit 42 determines whether the small groups (including one user) have joined or joined, based on the moving direction of the group after the joining. That is, the point calculation unit 42 determines that the small group that has moved in the same direction after the merge from the time of division is the merged side, and adds the leader points of the users of the merged small group.

  Here, when a plurality of grouped terminal devices 16 take different actions and then merge again, the terminal ID that precedes the merged group has an influence on the group. Therefore, the point calculation unit 42 adds the leader point to this. FIGS. 9A and 9B show an outline of another processing in the point calculation unit 42. In FIG. 9A, the first terminal device 16a, the second terminal device 16b, and the third terminal device 16c move in different directions. In FIG. 9B, following FIG. 9A, the first terminal device 16a and the second terminal device 16b join the third terminal device 16c, and the three terminal devices 16 move in the same direction. . When detecting such an action pattern, the point calculation unit 42 sets a leader point of the merged terminal device 16 (the third terminal device 16c in this example) to a predetermined value (for example, “30”) for each merge. Is added.

  That is, the point calculation unit 42 updates the group information table by adding the leader point related to the merging action to the leader point of the group information table. If the small group on the merged side is composed of a plurality of users, the point calculation unit 42 may add the leader points to all of them, or may add the leader point to a predetermined rank (for example, The addition may be limited to the users located at the first and second positions. In addition, the point calculation unit 42 may change the leader points to be added according to the order from the head of the small group on the merged side. For example, the point calculation unit 42 adds leader points such as “30” to the first user, “20” to the second user from the top, and “10” to the third user from the top.

  The point calculation unit 42 may calculate the leader point based on the number of subjects included in the small group when the group is divided into a plurality. That is, the point calculation unit 42 may calculate the value of the leader point related to the merging action according to the number of users who have taken another action (the number of users who have joined). Specifically, the point calculation unit 42 calculates the value of the leader point related to the merging action such that the larger the number of persons who have taken another action (the larger the number of small groups that have joined), the larger the value. . For example, the point calculation unit 42 has “20” when two people join after taking another action, and “30” when three people join after taking another action, and four people took another action. If they join later, a leader point such as "40" is added.

  The point calculation unit 42 may calculate the value of the leader point related to the merging action according to the number of persons on the merging side. Specifically, the point calculation unit 42 calculates a leader point having a smaller value as the number of persons on the merged side increases, and adds the leader point to each of the users on the merged side. For example, the point calculation unit 42 calculates “50” when the number of persons joined is one, “30” when the number of persons is two, and “3” when three persons are three. Each person adds a leader point such as "10". In addition, when there are a plurality of users on the merging side, the point calculation unit 42 may add the leader point related to the merging action to the user located at the top of the users. In other words, the point calculation unit 42 may calculate a leader point having a smaller value as the number of persons on the merging side increases, and may add the leader point related to the merging action to the user located at the head. For example, the point calculation unit 42 determines “50” when the number of persons on the merging side is one, “40” when two persons are joined, and “40” when the number of persons joined is three. Alternatively, a leader point such as "30" may be added to the person located at the top of the list.

  Next, in step S220, when the subject stops, the point calculation unit 42 defines a pattern in which another subject also stops following the stop as a third predetermined action pattern, and determines the number of times of movement stop as a characteristic action. Calculate as quantity. The stop following the stop means that immediately after a certain terminal device 16 (for example, the first terminal device 16a) stops moving, another terminal device 16 (a terminal device 16 other than the first terminal device 16a in the group) stops moving. This is the situation where the movement has been stopped. This situation is likely to be a situation in which the user who has stopped moving first finds a store or the like that he is interested in, stops there, and is considering entering the store. For this reason, the point calculation unit 42 adds a predetermined value (“for example,“ 20 ”) to the leader point of the terminal device 16 that has stopped moving first for each stop of movement of another terminal device 16.

  The point calculation unit 42 may calculate the leader point based on the number of other subjects that have stopped following the stop. Specifically, the point calculation unit 42 calculates a leader point having a larger value as the number of persons who stop later increases. For example, the point calculation unit 42 adds a leader point such as “20” when two people stop, “30” when three people stop, and “40” when four people stop. In addition, if there are a mixture of users who have stopped afterwards and users who have continued to move without stopping, the point calculation unit 42 calculates the leader point related to the stopping action according to the proportion of users who have stopped afterwards. It may be calculated. For example, when the number of stopped users is Ns and the number of users who continue moving is Nc, the point calculation unit 42 calculates the ratio of the number of stopped users as Ns ÷ (Ns + Nc). Further, the point calculation unit 42 sets “40” when the ratio is 80% or more, “30” when the ratio is 60% or more and less than 80%, “20” when the ratio is 40% or more and less than 60%, If it is less than 40%, a leader point such as "0" is added. As described above, the point calculation unit 42 may calculate the leader point having a larger value as the ratio of the number of stopped persons increases.

  The point calculation unit 42 determines whether or not the time-series data of the position information of each terminal device 16 matches at least one of the first to third predetermined behavior patterns. Alternatively, in addition to the deterministic determination, the determination may be made stochastically. In that case, the point calculation unit 42 uses a probability model that receives time-series data of position information as input and outputs a probability value corresponding to a predetermined behavior pattern of each terminal device 16.

  For example, in the process for the first predetermined action pattern in step S200, the point calculation unit 42 determines that the two terminal devices 16 from the head of the group are substantially side by side during a predetermined period (for example, one minute). The probability that the first terminal device 16a is the first place is calculated as “0.75” or the like. At that time, the point calculation unit 42 calculates the probability that the second terminal device 16b is the first place, such as “0.25”. The case where they are substantially horizontally arranged is, for example, a case where it is not clear which is the first place, a case where the first place and the second place of the group are frequently switched within a short time, and the state is not stable.

  In addition, for example, in the processing for the third predetermined action pattern in step S220, the point calculation unit 42 determines that the first terminal device 16a stops moving and the other terminal devices 16 stop moving although the moving speed is significantly reduced. When the vehicle moves forward slowly without performing, the probability of matching with the third predetermined action pattern is calculated as “0.2” or the like. The point calculation unit 42 may use the calculated probability value as it is as the characteristic action amount, or may use a value obtained by multiplying the probability value by a predetermined coefficient (for example, 100 times) as the characteristic action amount.

  As described above, the characteristic action amount is a degree (time, time, time, etc.) in which the time-series data of the position information of a certain subject matches a predetermined behavior pattern in which the influence of the subject in the group is determined to be high Distance, probability). Furthermore, the leader point is the sum total of the characteristic action amount of each subject during a predetermined period (for example, from the time when the group is formed to the time when the group is dissolved). It has the nature which becomes.

  Returning to the description of the flowchart of FIG. In step S120, the point calculation unit 42 normalizes the leader point (influence value) of each terminal ID. For example, when there are three users in the group, the leader points of each user are normalized such that the total value of the three leader points is “1.0” or “100”. Alternatively, normalization may be performed such that the leader point maximum value of the members in the group becomes a predetermined value (for example, “100”). The normalization process is performed at a predetermined time interval (for example, once a minute), every time the group moves a predetermined distance (for example, 100 m), or stops the group from moving for a predetermined time (for example, 30 seconds). It is executed every time. The information providing apparatus 10 may be executed at a timing when an information providing request is received from the terminal device 16 of the group. The point calculation unit 42 reads the leader point of terminal ID = i (i = 1 to N) from the group information table, where N is the number of users (terminal devices 16) belonging to the processing target group, and calculates the total value of the leader points. calculate. The point calculation unit 42 may set a value obtained by dividing the leader point by the total value as the normalized leader point. The point calculation unit 42 stores the group ID, the terminal ID, and the calculated normalized leader point in a storage unit in the point calculation unit 42 in association with each other. By storing the normalized leader point in an area different from the database storing the leader point shown in FIG. 6, even if the group action is continuing and the updating of the leader point of each terminal device 16 is continued. The information providing process can be performed by calculating the normalized reader at an arbitrary timing. However, it is also possible to omit this normalization processing and use a leader point that has not been normalized in the following processing. Hereinafter, the leader point of the user belonging to the processing target group is set to P [i] (i = 1 to N). In the following description, it is assumed that P [i] is normalized.

  The item selection unit 26 selects an item to be presented to the group so that the subject having the higher leader point calculated by the point calculation unit 42 has priority. To this end, in step S130, the item selection unit 26 adds the degree of preference for the item attribute of the user corresponding to each subject to the group while weighting the leader point of each subject calculated by the point calculation unit 42. Derives the degree of preference for the item attribute of. Here, the item attribute is attribute information indicating the property of the item. For example, when expressing the characteristics of a restaurant, the food genre, the size of the store (large, medium, small, etc.), the type of location (downtown, suburbs, etc.), and the type of store atmosphere (luxury, common people, etc.) Target, lively, quiet, etc.). In the present embodiment, a dish genre is used as an item attribute, but processing using another item attribute is also possible. In addition, one item attribute (for example, Japanese food) generally corresponds to a plurality of items (for example, restaurant-1, restaurant-2, restaurant-3). The process of step S130 will be specifically described. First, the item selection unit 26 acquires, from the user information storage unit 28, preference information indicating a preference for a cooking genre for each terminal ID in the group.

  The user information storage unit 28 stores, for each terminal ID for identifying a terminal device or each user ID for identifying a user, preference information of the user. FIG. 10 shows a data structure of a database (user preference database) stored in the user information storage unit 28. In this embodiment, it is assumed that one user occupies and uses one terminal device 16 like a mobile phone / smartphone, and the terminal ID and the user ID are substantially equal. In the present embodiment, the preference information of the user who uses the terminal device 16 is managed for each terminal ID, but one terminal device 16 is used by a plurality of users by using both the terminal ID and the user ID. It is of course possible to deal with As shown in the drawing, the user's preference information (preference degree) for the cooking genre is shown, and the numerical value (point) is the number of visits to stores of each genre and the degree of preference for each genre. In the example shown in this figure, normalization is performed so that the sum of the preference degrees of the respective terminal IDs becomes a constant value. However, the present invention is not limited to this. Good.

If the total number of cooking genres is M and the degree of preference for the cooking genre j (j = 1 to M) of the terminal device i (terminal ID = i) is F [i] [j], the item selection unit 26 The preference level A [j] for the cooking genre j is calculated as follows. In other words, the preference of the user having the larger leader point is more strongly reflected in the preference of the group.

  The item selection unit 26 selects an item based on the preference for the item attribute (the preference for the cooking genre) in the group derived in step S130. For this purpose, in step S140, the item selection unit 26 calculates a recommendation degree (recommendation score) of each item (store) for the group. At that time, the item selection unit 26 refers to the database stored in the provision information storage unit 30. The provided information storage unit 30 stores a store ID for identifying a store, a store name, a genre of food provided by the store, position information (latitude and longitude) of the store, an address, and contact information (telephone number, mail address). , Etc.), a description introducing the store, and information associated with the evaluation points are stored. FIG. 11 shows a data structure of the store database stored in the provision information storage unit 30. The evaluation score is a numerical value of the reputation of the store. Like a word-of-mouth site on the Internet, it may be created by voting / posting by a large number of users, or may be scored by a culinary expert or critic. In the example of FIG. 11, a score out of 100 is used. Further, one store may correspond to a plurality of cooking genres. For example, a store may support both "French" and "Italian."

To be more specific, the total number of stores (recommended candidates) registered in the store database is K, and the evaluation score of store k (k = 1 to K) is R [k]. , The degree of recommendation S [k] (k = 1 to K) of the store k with respect to the group is calculated as follows. Here, δ (k, j) is a function that returns “1” when the shop k corresponds to the cooking genre j, and returns “0” when the shop k does not. That is, in the processing according to the formulas (1) and (2), the degree of recommendation for each item in the group is calculated by adding the degree of preference to the item attribute of each of the members constituting the group while weighting it with the leader point. It can be said that it is processing. Note that the stores for which the degree of recommendation is calculated may be narrowed down according to the current position of the group, or the distance between the current position of the group and the store may be reflected in the degree of recommendation of the store.

  Then, in step S150, the item selection unit 26 selects a predetermined number of stores from the stores registered in the store database in the descending order of the recommendation degree S [k] (k = 1 to K). Alternatively, a store whose recommendation degree is equal to or more than a predetermined value may be selected. The item selected by the item selection unit 26 is also referred to as a “recommended item”. As described above, the item selection unit 26 selects an item based on the leader point calculated by the point calculation unit 42. For example, the preference information is shown in FIG. 10, and among the position information measured 100 times, the number of times the third terminal device 16c was at the head is 60 times, the first terminal device 16a is 30 times, and the second terminal device is 30 times. Assume that 16b is 10 times. The influence value of each terminal ID is as follows: the first terminal device 16a = 0.3, the second terminal device 16b = 0.1, and the third terminal device 16c = 0.6. As a result, the group's preference for Japanese food is “5 × 0.3 + 2 × 0.1 + 2 × 0.6 = 2.9”. When other food genres are similarly calculated, French = 1.9, Italian = 4.9, ethnic = 2.4, and fast food = 3.9. In this group, the preference of Italian is the highest.

  Here, the recommendation level of the store is calculated using the leader points of all the users in the group, but another method may be used. For example, the item selection unit 26 may select one user having the highest leader point in the group, and use the preference of the user as the preference of the group. That is, the item selection unit 26 may determine recommendation information (information of recommended shops) using preference information of one of the most influential users. In addition, the item selection unit 26 may select a predetermined number of users in descending order of the leader points in the group, and determine recommendation information based on the leader points of the selected users.

In the above description, the item (store) is selected after calculating the group preference for the item attribute (culinary genre), but the group preference for the item (store) is calculated without calculating the group preference for the item attribute. It may be calculated directly. For example, instead of the user preference database shown in FIG. 10, store-specific preference data (not shown, which is associated with the user ID (i), the shop ID (k), and the degree of preference Fs [i] [k]) is stored. Item preference data) is stored in the user information storage unit 28. Then, instead of Expressions (1) and (2), Expression (3) can be used to calculate the recommendation degree S [k] (k = 1 to K) of the store k.

  That is, the degree of recommendation for each item in the group is calculated by adding the degree of preference for each item of the members constituting the group while weighting it with the leader point. Here, when the preference degree of the terminal i for the store k is not registered, Fs [i] [k] = 0. R [k] is an evaluation point of store k registered in the store database, but the recommendation degree S [k] may be calculated without using this. That is, the recommendation degree may be calculated with R [k] = 1 in equation (3).

  The method using this formula (3) can calculate the recommendation degree with high accuracy when the degree of preference of the terminal device 16 of the group is registered for many stores. If the different preference data is small, the accuracy of the recommendation degree decreases. In addition, there is a restriction that the store is not a recommendation target unless the store has a registered preference of at least one person in the group. In some cases, there is a problem that the number of recommended items cannot be sufficiently obtained. In the above-described method of calculating the group preference level for the item attribute, a larger number of stores can be set as recommendation candidates, and a sufficient number of recommended items can be easily obtained.

  Further, the item selection unit 26 may determine whether the group has visited a store or the like described in the recommendation information, and reflect that information in the next recommendation information. Whether the group has visited a store or the like described in the recommendation information is determined by determining whether the position information of each terminal device 16 of the group substantially matches the position information of the store for a predetermined time (for example, 5 minutes) or more. Can be determined by If the group acts in accordance with the recommendation information, such a recommendation is likely to be accepted by the group in the future, so that the next or subsequent recommendation information includes the same information or similar information as this time. Process to make it easier to enter.

  Specifically, when updating the user preference information F as shown in FIG. 10, the item selecting unit 26, when visiting the store based on the recommendation information, increases the number of visits compared to other visits. Add a number. For example, in the example shown in FIG. 10, a process of increasing the numerical value by “1” for one store visit or a process of increasing the numerical value by “2” when visiting a store in accordance with the recommendation information may be performed. In addition to making the score of the corresponding cooking genre larger than usual, store-specific preference data (not shown) in which the terminal ID, the store ID, and the degree of preference are associated is stored in the user information storage unit 28, and the item selection is performed. The unit 26 may perform processing such as increasing the preference level of the store (store ID) visited by the user (terminal ID) according to the recommendation information, as compared to the normal process. Further, the item selection unit 26 may perform a process of reflecting the store-specific preference data when creating the next recommendation information. In addition, when the group does not visit a store or the like described in the recommendation information, the item selecting unit 26 performs a process of reducing the degree of preference of the store-specific preference data, and uses the process to recommend and visit the last time. A process that makes it difficult for the store that has not been included in the next recommendation information may be performed.

  The control unit 34 controls the entire information providing device 10. The control unit 34 also causes the transmitting unit 32 to transmit information (recommended information) on the item (recommended item) selected by the item selecting unit 26. Specifically, necessary data on recommended items is read from the store database shown in FIG. 11 to create recommended information, and the transmitting unit 32 transmits the recommended information. The transmitting unit 32 transmits the recommendation information to the group terminal devices 16 via the network 12 and the base station device 14. This transmission (recommended information provision) may be performed periodically by the information provision device 10, or may be transmitted as a response to an information provision request received from at least one terminal device 16 in the group. . When the information providing apparatus 10 transmits periodically, for example, the information may be transmitted at a predetermined time interval (for example, once a minute) during a period in which a group is formed, or the group may be transmitted at a predetermined distance. (For example, 100 m). Alternatively, the provision of information may be stopped by detecting that the group has been canceled. Further, when the group stops for a long time at a position corresponding to the latitude and longitude of the store registered in the store database, it can be determined that the store has entered a certain store, and thus the information provision may be stopped. In addition, since the information providing apparatus 10 can accurately grasp the number of terminals (the number of terminals) of the group, the information providing apparatus 10 combines the data indicating the availability of the store to provide recommendation information. Whether or not may be presented. Alternatively, the item selection unit 26 may select a store to be recommended, and provide recommendation information based on the store, limited to stores that all the groups can enter. FIG. 12 shows a data structure of information transmitted from the transmission unit 32. As shown in the figure, information of recommended stores (recommended information) is configured in a ranking format, and recommendation order, store ID, recommendation level (recommended score), store name, store description, current vacancy status, store position information ( It is composed of information such as latitude, longitude, address, contact information (telephone number, mail address, etc.), and distance from the current position of the terminal device 16.

  FIG. 13 shows a configuration of the terminal device 16. The terminal device 16 includes an acquisition unit 50, a display unit 52, a control unit 54, a transmission unit 56, and a reception unit 58. The acquisition unit 50 has a positioning function using a GPS, a wireless LAN, or the like, and periodically acquires the position information of the user or the terminal device 16 by measuring the position of the terminal device 16. The position information is indicated by latitude and longitude. The acquisition unit 50 sequentially outputs the position information to the control unit 54.

  The control unit 54 sequentially inputs the position information from the acquisition unit 50. The control unit 54 controls the timing of communication in the transmission unit 56 and the reception unit 58. The control unit 54 sequentially outputs the position information to the transmission unit 56 when the transmission timing comes. The transmitting unit 56 transmits the position information sequentially input from the control unit 54 to the information providing device 10 via the base station device 14 and the network 12. Since such position information is sequentially transmitted, this corresponds to transmitting time-series data of the position information. Further, as described above, the terminal ID and the date and time information are transmitted together with the position information. Note that, as described above, the terminal device 16 need not transmit the date and time information.

  The receiving unit 58 receives information (recommended information) on the item selected in the information providing device 10 that has transmitted the time-series data of the positional information in the transmitting unit 56 from the information providing device 10 via the base station device 14 and the network 12. I do. The information providing apparatus 10 may periodically transmit the recommendation information to the terminal device 16. Also, data requesting the recommendation information (recommendation information request) is transmitted from the transmission unit 56 to the information providing apparatus 10, and as a response, the information providing apparatus 10 transmits the recommendation information, and the reception unit 58 receives the recommendation information. Is also good. The receiving unit 58 outputs the received information to the control unit 54. The control unit 54 causes the display unit 52 to display information (recommended information) on the item received by the receiving unit 58. The display unit 52 displays a screen as shown in FIG. As shown in the figure, information such as a recommended rank, a recommended rank, a store name, a store description, a current vacancy status, an address, and a distance from the current position are displayed for each recommended store. Here, the recommendation rank is information that roughly indicates the size of the recommendation level, and is information that the control unit 54 has created based on the recommendation level. For example, when the recommendation degree is a numerical value in the range of 0 to 10, a store with a recommendation degree of 8 or more is “★★★”, a store with a recommendation degree of 6 or more and less than 8 “★★”, and a recommendation degree of 6 The recommended rank may be determined using a predetermined rule such as “★” for stores less than. Note that the information providing apparatus 10 may transmit the recommended rank. Instead of the recommendation rank, the recommendation degree may be displayed as it is. It should be noted that the store ID included in the recommendation information is not particularly meaningful information for the user, and may or may not be displayed. For example, when the user requests detailed information of a specific store, by transmitting the store ID of the store specified by the user from the terminal device 16 to the information providing device 10, processing can be performed efficiently. In the example shown in the figure, a "map display" button and a "contact" button are displayed for each recommended shop. When the user presses the "map display" button, the location information (latitude, longitude) or address of the store is transmitted to a map application built into the terminal device 16 or an external map providing service, and the store is displayed on a map. You. When the user presses the “contact” button, the user can contact the store via a call application, a mail application, an external communication service, or the like built in the terminal device 16. Further, as shown in this figure, the name of the area where the group is located or the number of people in the group may be displayed, such as "restaurant list recommended for a group of 5 people in the XX district". Alternatively, the user may input conditions regarding the recommended rank, the genre of the desired store, the distance from the current position, and the like to the terminal device 16 and display only information on stores that match the conditions.

  In the above description, the information providing apparatus 10 executes the calculation of the characteristic action amount and the leader point, but other methods may be used. For example, the terminal device 16 exchanges positional information with the nearby terminal devices 16 as needed, and forms a group using its own positional information and the positional information of other terminal devices 16. Further, the characteristic action amount and the leader point may be calculated for the terminal devices 16 included in the group. That is, the position information management unit 22 and the group information processing unit 24 may be included in the terminal device 16. Each terminal device 16 in the group may calculate only its own leader point, or a specific terminal device 16 may calculate the leader points of a plurality of terminal devices 16 collectively. The terminal device 16 that has calculated the leader point transmits the leader point to the information providing device 10. The item selection unit 26 and the transmission unit 32 of the information providing device 10 execute the same processing as before based on the received leader point, and select an appropriate item for the group.

  Also, as described above, the point calculation unit 42 calculates the leader point using the latitude and longitude information of the terminal device 16 stored in the position information management unit 22, but another method may be used. . For example, the image / video of a surveillance camera installed in a shopping mall or a shopping street is analyzed, and the group forming unit 40 executes a group forming process by using an image recognition technique, and the point calculation unit 42 performs a leader point calculation. May be executed. In this case, at least one user in the group only needs to have the terminal device 16 in order to receive the recommendation information, and it is not necessary for all members of the group to carry the terminal device 16. Further, latitude and longitude information is not necessarily required, and it is sufficient that relative positional relationships and relative movement information of users in a group can be acquired. In addition, by using surveillance cameras installed in the aisles of shopping malls and shopping streets and surveillance cameras installed at the entrance of the store and using face recognition technology, the leader point of any person entering the store is high. The clerk can know which person has a lower leader point. By knowing such information, it is possible to improve customer service and improve efficiency, for example, by giving a menu or wine list preferentially to a person with a high leader point.

  Further, the point calculation unit 42 may acquire the group action history in the store and estimate the influence on the product selection. For example, the point calculation unit 42 specifies a user who first arrives at a shelf of a product purchased by a certain user, performs processing such as adding points of the user, and performs processing such as adding a leader point (influence value ) May be calculated. That is, a leader point (influence value) related to selection (selection) of an item including not only a store but also a product or the like is calculated.

  According to the present embodiment, since a group is formed based on the time-series data of the position information, it is possible to group a plurality of subjects who act together. Further, since the grouping is automatically performed, the convenience for the user can be improved. In addition, based on the time-series data of the position information, a characteristic action amount indicating a degree of matching with a predetermined action pattern in which the influence of the subject in the group is high is calculated, and then the leader point is calculated. Leader points can be derived in consideration of the influence of the users within the group. In addition, since the item to be presented to the group is selected so that the subject having the higher leader point is prioritized, the item can be selected in consideration of the influence of the users in the group.

  In addition, at least one of the number of times the subject's movement pattern in the time-series data of the position information matches the predetermined action pattern, the matching time, the matching movement distance, and the matching probability is calculated as the characteristic action amount. Therefore, it is possible to derive a characteristic action amount that matches the degree of matching of the action pattern. In addition, since the leader point is calculated so as to have a higher value as the characteristic action amount increases, a leader point corresponding to the degree of matching of the action pattern can be derived. Further, since the predetermined action pattern is a pattern in which the subject is located at a position of a predetermined order from the head of the group with reference to the moving direction of the group, the influence of the user according to the position where the user in the group is present Can be estimated with high accuracy. In addition, since the leader point is calculated such that the higher the predetermined rank is, the higher the influence of the user existing at the head of the group can be estimated.

  In addition, the predetermined action pattern is a pattern in which, when a group is divided into a plurality of groups and then merges, the main body moves from the time of division in a direction in which the group moves after merging, so that the influence of the user is estimated with high accuracy. it can. In addition, since the leader point is calculated based on the number of subjects included in the small group when the group is divided into a plurality, the magnitude of the influence can be estimated with high accuracy. Further, the predetermined action pattern is a pattern in which, when the subject stops, the other subjects also stop following the stop, so that the influence when the subject stops can be estimated. In addition, since the leader point is calculated based on the number of other subjects that have stopped following the stop, the magnitude of the influence can be estimated with high accuracy. Also, since the leader point is calculated according to the number of subjects included in the group, the magnitude of the influence can be estimated with high accuracy.

  Also, by adding the preference level of the user corresponding to each subject to the item while adding weight to the leader point, the recommendation level for the item in the group is derived, and the item is selected according to the recommendation level. You can select items that take into account the influence of In addition, by adding the degree of preference for the item attribute of the user corresponding to each subject while weighting with the leader point, the degree of preference for the item attribute in the group is derived. Therefore, the degree of preference in consideration of the user's influence on the group. Can be derived. In addition, since an item is selected based on the degree of preference for an item attribute in a group, an item in consideration of the user's influence on the group can be selected from more recommendation candidates. In addition, a predetermined number of subjects are specified in descending order of the calculated leader points, and items are selected based on the leader points of the specified subjects. it can. In addition, by considering the influence of the users in the group, it is possible to provide information that leads to an actual action with higher accuracy than a method in which the influence of the users in the group is regarded as being equal. Further, the influence can be estimated only from the position information of the terminal device without registering the information on the influence of the users in the group in advance.

(Example 2)
Next, a second embodiment will be described. As in the first embodiment, the second embodiment relates to an information providing system including a plurality of terminal devices and an information providing device connectable thereto. The information providing apparatus according to the second embodiment also forms a group using the position information of the plurality of terminal devices, and estimates the influence of each user in the group. When estimating the influence of each user, , User attribute information is also used. The information providing system 100 and the terminal device 16 according to the second embodiment are the same types as those in FIGS. Here, the difference will be mainly described.

  FIG. 15 illustrates a configuration of the information providing apparatus 10 according to the second embodiment of the present invention. The information providing device 10 is configured in the same manner as in FIG. 2, but the group information processing unit 24 is configured to be able to refer to the database stored in the user information storage unit 28. The point calculation unit 42 calculates the leader point by estimating the influence within the group using not only the position information but also the attribute information (age, sex, etc.) of the user corresponding to the subject. The user information storage unit 28 stores user attribute information. FIG. 16 shows the data structure of the database stored in the user information storage unit 28. Terminal IDs (or user IDs) are stored in association with user attribute items such as date of birth, gender, office, post, annual income, and extra money this month. This is an example, and data of other user attribute items may be stored, or only some of the attribute items may be stored. Referring back to FIG.

  Although the point calculation unit 42 according to the first embodiment calculates the leader point based on the position information, the point calculation unit 42 according to the second embodiment also calculates the leader point based on the user attribute information. Further, the point calculation unit 42 calculates an overall leader point using the two leader points. Hereinafter, the leader point based on the position information of the terminal device i is Pg [i], the leader point based on the user attribute information is Ph [i], and the comprehensive leader point is P [i]. The point calculation unit 42 calculates the leader point Ph [i] based on the user attribute information by the following method.

  The first method is a method of calculating based on rules set for each user attribute item. For example, generally, since the influence of a user with a high age is high, Ph [i] can be calculated based on the date of birth. Specifically, the point calculation unit 42 gives a leader point of a larger value as the age is higher, based on the birth dates of the users in the group. For example, the point calculation unit 42 gives a leader point of “10” to a user with the highest age, “7” to a user with the second highest age, and “5” to a user with the third highest age. Further, the point calculation unit 42 may set the maximum value of the leader points to be given as 1, and give the leader points having a value larger than 0 and equal to or smaller than 1 according to the age. In addition, the point calculation unit 42 may include a rule such as ignoring the age difference and giving the same leader point to a person whose age rank is different by one from the other when the age difference is equal to or less than a predetermined value.

  In addition, the point calculation unit 42 may use a rule such as giving a larger leader point as the position becomes higher. For example, leader points such as "10" are given to the president, "5" to the manager, and "2" to the section manager. In addition, the point calculation unit 42 may calculate the leader point according to the annual income and the amount of the pocket money of this month. For example, since a person with a higher annual income tends to have a higher influence, a value obtained by multiplying the annual income by a predetermined coefficient (eg, 0.1%) can be set to Ph [i]. Also, the point calculation unit 42 may give a leader point such as “10” to the person with the highest annual income and “8” to the second person, and a predetermined leader point may be assigned according to the order of the annual income. You may give. The same applies to this month's pocket money, but this month's pocket money may accurately represent the amount of money each person can use at restaurants, etc., so the influence may be more accurately estimated than using the annual income . Further, the point calculation unit 42 may calculate a leader point based on gender.

The second method is to add leader points calculated for each user attribute item by using a plurality of user attribute items. That is, the point calculation unit 42 calculates the leader point based on a plurality of items included in the attribute information of the user. First, the point calculation unit 42 calculates the leader point Ph [1] [i] related to the date of birth using the rule described above in the first method, “the leader point increases as the age increases”. calculate. Next, the point calculation unit 42 calculates the leader point Ph [2] [i] related to the position using the rule “the higher the position, the higher the leader point” described in the first method. In this way, the point calculation unit 42 calculates Ph [l] [i] (l = 1 to L, i = 1 to N) for the L user attribute items, and The leader point Ph [i] based on the user attribute information is calculated. Here, w [l] (l = 1 to L) is a predetermined weight coefficient determined for each user attribute item, and w [l]> 0.

  The third method is a method of calculating based on a rule set using a combination of a plurality of user items. For example, the point calculation unit 42 calculates a relatively high leader point of a user having a birth date within a predetermined range and a predetermined gender. For example, women of a specific age group tend to have a high demand level for restaurants, so that the birth date is in a predetermined range and the gender of a female user is given a predetermined leader point. This makes it easier for such user preferences to be reflected in group decision-making. This third method may be used alone or in combination with another method. For example, the point calculation unit 42 first calculates a leader point according to the equation (4), and then gives a leader point of “10” to a user whose birth date is within a predetermined range and of a predetermined gender. May be performed.

  The fourth method is to use a rule that takes into account the organization to which the user belongs. At that time, the point calculation unit 42 refers to the information stored in the user information storage unit 28. FIG. 17 shows a data structure of another database stored in the user information storage unit 28. This indicates a business relationship between the companies. This shows a situation in which Company X orders a job from Company Y, and Company Y orders a job from Company Z. Generally, the ordering company person often selects a restaurant or the like in accordance with the taste of the ordering company person, and such a tendency is reflected in the leader point. In the examples shown in FIGS. 16 and 17, the leader point of the user with the workplace = X is the highest, the leader point of the user with the workplace = Y is the highest, and the leader point of the user with the workplace = Z is the lowest. . When there are a plurality of users who work at the same place, the leader point may be set to the same value, or a difference may be made using attribute items such as date of birth, gender, and post.

  The fifth method is a method that uses the relationship between the date of birth and the date (current date and time) at which processing is performed. For example, if the date on which the leader point is calculated is May 3, 2015, and the user whose birth date is May 3, 1979 exists in the group, a birthday party or the like of the person may be performed. Therefore, the point calculation unit 42 calculates the leader point of the person higher than usual. That is, the point calculation unit 42 calculates a relatively high leader point of a user whose processing date is on or near the birthday. As described above, the point calculation unit 42 calculates the leader point according to the relationship between the date on which the process is executed and the attribute information of the user.

  The sixth method is a method that uses the relationship between gender and the date (current date and time) at which processing is performed. For example, if the current date and time for calculating the leader point is White Day on March 14, 2015, it is highly likely that a man will select a restaurant in accordance with the preference of a woman, so the point calculation section 42 calculates the leader point of the woman. Is calculated higher than usual. Note that the point calculation unit 42 may use any one of the first to sixth methods described above alone, or may use a combination of a plurality of methods.

ここで、実施例１で説明した端末ｉの位置情報に基づくリーダーポイントをＰｇ［ｉ］（実施例１においてはＰ［ｉ］と表記）、ユーザ属性情報に基づくリーダーポイントをＰｈ［ｉ］と示す。また、α１、α２は、α１＞０、α２＞０を満たす所定の係数である。つまり、ポイント算出部４２は、位置情報に基づくリーダーポイントとユーザ属性情報に基づくリーダーポイントとの重み付き加算を実行する。なお、α１＝１、α２＝１として、位置情報に基づくリーダーポイントとユーザ属性情報に基づくリーダーポイントとの和（加算値）を総合的なリーダーポイントとしてもよい。 Next, the point calculation unit 42 calculates the overall leader point P [i] as follows.

Here, the leader point based on the position information of the terminal i described in the first embodiment is Pg [i] (in the first embodiment, described as P [i]), and the leader point based on the user attribute information is Ph [i]. Show. Α1 and α2 are predetermined coefficients that satisfy α1> 0 and α2> 0. That is, the point calculation unit 42 performs weighted addition of the leader point based on the position information and the leader point based on the user attribute information. Note that assuming α1 = 1 and α2 = 1, the sum (addition value) of the leader point based on the position information and the leader point based on the user attribute information may be used as the overall leader point.

ここで、γ１、γ２は、γ１＞０、γ２＞０を満たす所定数である。すなわち、Ｐｇ［ｉ］を基数とし、γ１を指数とする累乗値と、Ｐｈ［ｉ］を基数とし、γ２を指数とする累乗値との積（乗算値）をＰ［ｉ］とする。Ｐｇ［ｉ］をＰｈ［ｉ］に強く反映させたい場合は、γ１を大きな値にし、Ｐｈ［ｉ］をＰｈ［ｉ］に強く反映させたい場合は、γ２を大きな値にするとよい。なお、γ１＝１、γ２＝１として、位置情報に基づくリーダーポイントとユーザ属性情報に基づくリーダーポイントとの積（乗算値）を総合的なリーダーポイントとしてもよい。ポイント算出部４２が、式（５）または式（６）に基づいて、総合的なリーダーポイントをＰ［ｉ］を算出した後、実施例１と同様な処理が実行されればよい。つまり、式（５）または式（６）に基づいて算出したＰ［ｉ］が式（１）に代入される。 The point calculation unit 42 may calculate the total leader point P [i] as follows.

Here, γ1 and γ2 are predetermined numbers that satisfy γ1> 0 and γ2> 0. That is, a product (multiplied value) of a power value having Pg [i] as a radix and γ1 as an exponent and a power value having Ph [i] as a radix and γ2 as an exponent is P [i]. When it is desired to strongly reflect Pg [i] on Ph [i], γ1 is set to a large value. When it is desired to strongly reflect Ph [i] on Ph [i], γ2 is set to a large value. Note that, with γ1 = 1 and γ2 = 1, the product (multiplied value) of the leader point based on the position information and the leader point based on the user attribute information may be used as the overall leader point. After the point calculation unit 42 calculates the total leader point P [i] based on the equation (5) or the equation (6), the same processing as in the first embodiment may be executed. That is, P [i] calculated based on Expression (5) or Expression (6) is substituted into Expression (1).

  According to the present embodiment, since the leader point is calculated according to the attribute information of the user corresponding to the subject, the estimation accuracy of the influence of the users in the group can be improved. Further, since the leader point is calculated based on a plurality of items included in the user attribute information, the estimation accuracy can be improved. In addition, since the leader point is calculated according to the relationship between the date at which the process is executed and the attribute information of the user, the influence can be estimated while considering the date at which the process is executed.

(Example 3)
Next, a third embodiment will be described. Example 3 relates to an information providing system including a plurality of terminal devices and an information providing device connectable to the plurality of terminal devices, as in the past. Similarly to the second embodiment, the information providing apparatus according to the third embodiment also forms a group using the position information of a plurality of terminal devices, and estimates the influence of each user in the group. In the third embodiment, it is assumed that the group is a family. The information providing system 100, the terminal device 16, and the information providing device 10 according to the third embodiment are of the same type as those in FIGS. Here, the difference will be mainly described.

  The group forming unit 40 in FIG. 15 changes the condition for forming a group according to the attribute information of the user corresponding to the subject. The user attribute information is stored in the user information storage unit 28. FIG. 18 shows a data structure of another database stored in the user information storage unit 28 according to the third embodiment of the present invention. As shown, the terminal ID is registered for each family. Referring back to FIG. When terminal IDs are registered in family units, the group forming unit 40 relaxes the conditions for determining these terminal devices 16 as one group as compared to normal. For example, when the terminal device 16 existing within a predetermined distance for a predetermined time or longer is determined as one group, the group forming unit 40 sets the threshold value of the distance between terminals longer than usual, or sets the predetermined time to Or shorten it. The process is based on the assumption that families are likely to act together toward a common destination. For example, even if the child temporarily goes away first in the process of heading to the destination, it is possible to provide appropriate information while maintaining the determination of the group behavior.

  The point calculation unit 42 uses the date of birth and the relationship registered in the database shown in FIG. 18 to lower the leader point based on the positional information of the child of a predetermined age or less (for example, elementary school student or less) than the parent. I do. For example, even if the number of times that the child is located at the head of the group is large, the point calculation unit 42 performs a process that does not significantly reflect the leader point based on the position information. Specifically, the point calculation unit 42 multiplies the number of times the head position is observed by 0.1 for the terminal device 16 of a child under a predetermined age or sets an upper limit on the number of head positions used for leader point calculation. Control processing such as providing a control. Small children often act impulsively or suddenly, but by performing such control processing, a leader point based on position information is accurately calculated.

  In addition, since the mother generally has the strongest influence in the home, the point calculation unit 42 may perform processing to maximize the mother's leader point. Further, the point calculation unit 42 may calculate using the relevance between the month and day (current date and time) at which the processing is performed and the domestic relationship. For example, when the current date and time for calculating the leader point is June 21, 2015, which corresponds to Father's Day, the point calculation unit 42 may calculate the father's leader point higher than usual.

  According to the present embodiment, the conditions for forming a group are changed according to the attribute information of the user corresponding to the subject, so that a group suitable for circumstances such as a family can be formed. In addition, since the leader point is calculated according to the family structure, the influence within the family can be estimated with high accuracy.

(Example 4)
Next, a fourth embodiment will be described. Fourth Embodiment A fourth embodiment relates to an information providing system including a plurality of terminal devices and an information providing device connectable to them, as in the past. The information providing apparatus according to the fourth embodiment also forms a group using the positional information of a plurality of terminal devices and estimates the influence of each user in the group, as in the past. On the other hand, in the past, common information was provided to the users in the group, but in the fourth embodiment, at least one of the position information, the user preference information, and the user attribute information is used to form one group. A plurality of subgroups, and provide appropriate information to each subgroup. The information providing system 100, the terminal device 16, and the information providing device 10 according to the fourth embodiment are of the same type as in FIGS. 1, 13, and 15. Here, the difference will be mainly described.

  The group forming unit 40 in FIG. 15 forms at least two subgroups by classifying a plurality of subjects included in the group into at least two subgroups. As described above, the group forming unit 40 calculates the matrix data indicating the average distance between the terminals as shown in FIG. 5 based on the position information stored in the position information management unit 22. The group forming unit 40 forms a subgroup using the matrix data. The group forming unit 40 sets a predetermined threshold value for forming a subgroup, and performs processing for associating terminal devices whose average distance between terminals is equal to or less than the threshold value to the same subgroup. For example, the condition that the subgroup threshold value is 3 m is applied to a group including the first terminal device 16a to the fifth terminal device 16e. In that case, there are two sub-groups, a sub-group 1 composed of the first terminal 16a, the second terminal 16b, and the third terminal 16c, and a sub-group composed of the fourth terminal 16d and the fifth terminal 16e. A group is created.

ここで、Ｅ［ｉ］［ｐ］は、端末装置ｉと端末装置ｐの嗜好に関する類似度であり、数値が大きいほど、両者の嗜好が一致していることを示す。ここで、ｉ＝１〜Ｎ、ｐ＝１〜Ｎ、ｉ≠ｐである。 Separately, a method in which the group forming unit 40 forms a subgroup using the user preference information will be described. The user information storage unit 28 stores user preference information (preference degree) F [i] [j] as shown in FIG. 10 (i = 1 to N, j = 1 to M). This is a numerical value indicating the strength of preference of the terminal device i (corresponding to the user i) for the cooking genre j. Based on this data, the group forming unit 40 calculates the degree of preference matching between the two terminal devices 16 (two users) according to the following cosine scale.

Here, E [i] [p] is a similarity regarding the preference of the terminal device i and the terminal device p, and the larger the numerical value, the more the preference of the two matches. Here, i = 1 to N, p = 1 to N, and i ≠ p.

ここで、Ｆａ［ｉ］は、端末装置ｉに対応する嗜好情報Ｆ［ｉ］［ｊ］の平均値であり、Ｆａ［ｐ］は、端末装置ｐに対応する嗜好情報Ｆ［ｐ］［ｊ］の平均値である。式（８）によるＥ［ｉ］［ｐ］は、−１〜＋１の範囲の値をとり、式（７）同様に、端末装置ｉと端末装置ｐの嗜好が一致しているほど大きな値となる。また、これ以外の方法で、２つの端末装置１６の間の嗜好に関する類似度を算出してもよい。例えば、端末装置ｉの嗜好情報と、端末装置ｐの嗜好情報との距離（ユークリッド距離など）の逆数を類似度としてもよい。このようにして、グループ形成部４０は、グループ内の端末装置１６の間の類似度を算出し、類似度が所定のしきい値以上である端末装置をまとめてサブグループとする。 Alternatively, the group forming unit 40 may calculate the similarity E [i] [p] between the terminal device i and the terminal device p according to the following correlation coefficient.

Here, Fa [i] is the average value of the preference information F [i] [j] corresponding to the terminal device i, and Fa [p] is the preference information F [p] [j corresponding to the terminal device p. ] Is the average value. E [i] [p] according to equation (8) takes a value in the range of −1 to +1. Similarly to equation (7), E [i] [p] increases as the preference of the terminal device i and the terminal device p coincides with each other. Become. Alternatively, the similarity regarding the preference between the two terminal devices 16 may be calculated by other methods. For example, the reciprocal of the distance (such as the Euclidean distance) between the preference information of the terminal device i and the preference information of the terminal device p may be used as the similarity. In this way, the group forming unit 40 calculates the similarity between the terminal devices 16 in the group, and collects the terminal devices whose similarity is equal to or greater than the predetermined threshold value into a subgroup.

  Further, a method of forming a subgroup using user attribute information will be described separately. A first method of creating a subgroup from user attribute information is a method in which terminal devices 16 having the same user attribute item are set as a subgroup. For example, in the case of the user attribute information shown in FIG. 16, the group forming unit 40 forms three subgroups of X, Y, and Z for each work place. In addition, the group forming unit 40 may form a subgroup by grouping persons having the same or similar post levels. For example, subgroups such as executives, general managers, section managers, chiefs, and general members may be formed. Further, the group forming unit 40 may form a subgroup for each age group. For example, subgroups such as teens, twenties, thirties, forties, fifties, and sixties may be formed. Further, the group forming unit 40 may form a subgroup according to annual income. For example, a subgroup may be formed based on a predetermined annual income range such as 3,000,000 yen to 4,000,000 yen, 4,000,000 yen to 5,000,000 yen, 5,000,000 yen to 6,000,000 yen.

ここで、δ関数は典型的には、２つの引数が一致する場合に「１」、異なる場合に「０」を返す関数である。なお、属性項目に応じて、異なる応答特性のδ関数を用いてもよい。例えば、属性項目が、性別、勤務先などのように名義尺度である場合には、２つの引数が完全一致する場合にのみ「１」を返す関数が使用される。一方、属性項目が年収、今月の小遣い、などのように数値（間隔尺度または比例尺度）である場合には、２つの引数の相対的な差が１０％以下である場合に「１」を返し、それ以外の場合に「０」を返すような関数が使用される。 A second method of creating a subgroup from the user attribute information is a method of calculating the similarity between the terminal devices 16 related to the user attribute information. As shown in FIG. 16, the user attribute information G [i] [l] of the attribute item 1 is stored for the terminal device i (i = 1 to N, l = 1 to L). The group forming unit 40 calculates the similarity Q [i] [p] (i ≠ p) between the terminal device i and the terminal device p according to the user attribute information as follows.

Here, the δ function is a function that typically returns “1” when two arguments match and returns “0” when the two arguments differ. Note that a δ function having different response characteristics may be used depending on the attribute item. For example, when the attribute item is a nominal scale such as gender, work place, etc., a function that returns “1” only when the two arguments completely match is used. On the other hand, if the attribute item is a numerical value (interval scale or proportional scale) such as annual income, this month's pocket money, etc., “1” is returned if the relative difference between the two arguments is 10% or less. Otherwise, a function that returns "0" is used.

  Alternatively, "1" when the two arguments completely match, "0.8" if the error is within 5%, "0.5" if the error is 5 to 10%, and 10 to 20%. For example, a function having a characteristic that returns “0.2” and an error of 20% or more and returns “0” may be used. After calculating the similarity Q [i] [p] (i = 1 to N, p = 1 to N, i ≠ p) related to the user attribute information, the group forming unit 40 determines that the similarity is a predetermined threshold. The above terminal devices 16 are collectively referred to as a subgroup. Furthermore, the group forming unit 40 may create a subgroup by combining the above-described methods. For example, a subgroup may be created using all three pieces of position information, user preference information, and user attribute information.

After creating a subgroup in this way, the item selecting unit 26 creates recommendation information for each subgroup. In other words, the item selection unit 26 executes the processing related to the group so far corresponding to the sub-group. This is equivalent to selecting an item to be presented to each subgroup so that a subject having a higher leader point calculated by the point calculation unit 42 is prioritized. The recommendation information created for one subgroup may be provided only to the terminal devices 16 belonging to the subgroup, or the terminal devices 16 belonging to another subgroup (all the terminal devices 16 in the group). May be provided.

  According to this embodiment, at least two subgroups are formed by classifying a plurality of subjects included in a group into at least two subgroups. Can be In addition, since items to be presented to each subgroup are also selected so that a subject having a higher leader point is prioritized, items can be presented for each subgroup. In addition, for example, when there is no store where all the groups can enter, the group may be divided into two or more subgroups, and the stores that can enter the store may be presented for each subgroup. Each subgroup is composed of members having relatively high similarity with respect to at least one of a user's behavior, a user's preference, and a user's attribute, so that communication between users may be easily taken. It is possible to increase user convenience and satisfaction.

(Example 5)
Next, a fifth embodiment will be described. The fifth embodiment relates to an information providing system including a plurality of terminal devices and an information providing device connectable to them, as in the past. The information providing apparatus according to the fifth embodiment also forms a group using the position information of a plurality of terminal devices and estimates the influence of each user in the group, as in the past. Until now, the item was selected based on the influence of each user in the group. In the fifth embodiment, information indicating the influence of each user in the group is provided. The information providing system 100 and the terminal device 16 according to the fifth embodiment are of the same type as those in FIGS. Here, the difference will be mainly described.

  FIG. 19 illustrates a configuration of the information providing apparatus 10 according to the fifth embodiment of the present invention. The information providing device 10 includes a reception unit 20, a position information management unit 22, a group information processing unit 24, a user information storage unit 28, a transmission unit 32, and a control unit 34. Further, the group information processing unit 24 includes a group forming unit 40 and a point calculating unit 42, and the user information storing unit 28 includes a point storing unit 44. The conventional information providing apparatus 10 provides store information and the like based on the location information as recommendation information to the terminal devices 16 belonging to the group. Provide information indicating influence. That is, the point calculation unit 42 calculates a leader point for each terminal device 16, and the transmission unit 32 transmits the leader point calculated by the point calculation unit 42 or an index that aggregates the leader points to each terminal device 16. The transmission destination is not limited to the terminal device 16 in the group, and may be another terminal device 16 or another device (for example, a marketing data analysis system or the like).

  On the other hand, the transmission unit 56 of the terminal device 16 shown in FIG. The control unit 54 controls the display unit 52 to display the data received by the transmission unit 56. When the received data is display data, the control unit 54 of the terminal device 16 controls the display unit 52 accordingly. When the received data is material data, the control unit 54 of the terminal device 16 creates display data based on the data and causes the display unit 52 to display the data.

  FIG. 20 illustrates a screen displayed on the display unit 52 according to the fifth embodiment of the present invention. This corresponds to an example of the display of the leader point on the terminal device 16. In the example shown in the figure, the leader points are displayed for the five terminal devices 16 that have actually formed one group. A, B, C, D, and E on the horizontal axis indicate terminal IDs assigned to the terminal devices 16 in the group. These correspond to the first terminal device 16a, the second terminal device 16b, the third terminal device 16c, the fourth terminal device 16d, and the fifth terminal device 16e, as described above. The vertical axis indicates the leader point. As shown, the leader point of the third terminal device 16c is the highest, and the second terminal device 16b is the second highest. In this example, a value normalized so that the sum of the leader points in the group is "1" is used. However, the sum is set to "100" or the maximum value is set to a predetermined value (for example, 100). Or you may make it.

  FIG. 21 shows another screen displayed on the display unit 52 according to the fifth embodiment of the present invention. This corresponds to another example of the leader point display. Here, the leader point Pg based on the position information is indicated by a dot pattern, and the leader point Ph based on the user attribute information is indicated by a hatched pattern. With such a display, the leader point of each user can be represented in more detail. For example, after the third terminal device 16c, the leader point of the second terminal device 16b is larger, but the configurations of both are greatly different. It can be seen that the third terminal device 16c has a large leader point based on the user attribute information, while the second terminal device 16b has a large leader point based on the location information. In this example, the leader point is the sum (addition value) of Pg and Ph, and is normalized such that the sum of Pg in the group is “100” and the sum of Ph is “100”.

  FIG. 22 shows another screen displayed on the display unit 52 according to the fifth embodiment of the present invention. This corresponds to an example of a leader point display when a subgroup is formed. In this example, the first terminal device 16a, the second terminal device 16b, and the third terminal device 16c correspond to the subgroup 1, and the fourth terminal device 16d and the fifth terminal device 16e correspond to the subgroup 2. Pg and Ph are normalized such that the sum in each subgroup becomes “100”. It returns to FIG.

  The leader point-related data calculated by the point calculation unit 42 may be based on an action history when a store or the like is selected once or may be based on a plurality of action histories. The leader point in one group action can be said to be a short-term leader point. Also, a value (index) obtained by summing (aggregating) leader points in a plurality of group actions can be said to be a long-term leader point. In order to provide long-term leader points, the point storage unit 44 stores the leader points calculated in the past by the point calculation unit 42 in association with each subject. FIG. 23 shows a data structure of a database (point history table) stored in the point calculation unit 42 according to the fifth embodiment of the present invention. The point history table stores point history data in which a group ID, a terminal ID, a calculation date and time, a leader point, a store ID, and a purpose classification are associated. As described above, the group ID is an ID capable of uniquely identifying a group, and a new group ID is assigned every time a group is generated. The purpose classification is information indicating a destination such as a store or the purpose of an action selected by the group.

  As the information indicating the rough purpose, for example, classification such as “meal”, “shopping”, “drinking”, and “theme park” can be used. That is, it is a classification corresponding to the superordinate concept (more abstract concept) of the store genre. Further, as more detailed destinations, for example, "Japanese restaurant", "French restaurant", "Chinese restaurant", "Izakaya", "bar", "clothes store", "grocery store", "bookstore", " A classification (store genre) such as "record shop" or "florist" can be used. It should be noted that the arrival of the group at the destination can be determined on condition that the position information of each terminal device 16 of the group does not substantially change for a predetermined time (for example, 5 minutes) or more as described above. Further, by searching a store database or the like using the position information, it is possible to acquire store information (genre and the like), and the information may be used for the purpose classification. At an appropriate time, such as when the group arrives at the destination, a questionnaire survey asking the purpose of this action is sent to the group terminal device, and the information obtained by collecting it is used as the purpose classification. Good. Further, by using such a point history table, it is also possible to compare the leader points with a combination of arbitrary terminal devices 16 that do not actually act as the same group.

  The point calculator 42 refers to the point storage unit 44 and handles data whose calculation date and time falls within a predetermined period (for example, 01/01/2015 to 01/31/2015) to correspond to one entity. Calculate an index that aggregates multiple leader points. More specifically, the point calculation unit 42 calculates a total value, a representative value (average value, median value, mode value), a maximum value, a minimum value, and a value (standard deviation, Each index, such as variance and (maximum value-minimum value), is calculated. Further, the point calculation unit 42 may calculate an index in which the leader points for each terminal ID are aggregated for each purpose classification (category). That is, the point calculation unit 42 may calculate at least one of the frequency for each category and the relative frequency for each category as an index. Since the number of times the leader point is calculated within the predetermined period differs for each terminal device 16, when comparing the terminal devices 16, the average value or the median value is more appropriate than the total value.

  Further, the point calculation unit 42 may correct the leader point according to the number of members of the group. In a case where the sum of the leader points is normalized to be "1" in one group, the average value (expected value) of the leader points per person becomes smaller as the number of members in the group increases. Further, even when the group is not normalized, if the number of people in the group increases, the frequency of each user being located at the head of the group decreases, so that the leader point per person tends to decrease. Therefore, when the number of groups is large, the point calculation unit 42 performs a correction process such as increasing the value of the leader point. For example, the number of persons is calculated for each group ID, and when the number of persons of a certain group ID is equal to or more than a third predetermined value and less than a fourth predetermined value, the leader point corresponding to the group ID is multiplied by “1.5 times”. Further, when the number of persons of a certain group ID is equal to or more than the fourth predetermined value, processing such as "double" the leader point corresponding to the group ID is performed. In addition, using the corrected leader points, an index that aggregates the leader points is calculated.

  FIG. 24 shows another screen displayed on the display unit 52 according to the fifth embodiment of the present invention. This corresponds to an example of displaying the leader points for each purpose classification. Three purpose classifications of “meal”, “drinking”, and “shopping” are used. The leader points shown here are calculated by the point calculation unit 42 by calculating the average value of the leader points acquired by each terminal device for each predetermined period (average value for each combination of the terminal device and the target class) obtained by the terminal device in the predetermined period. This is a value normalized so that the sum of the average values for each device becomes “1”. In other words, it can be said that the point calculation unit 42 calculates the index for each purpose of the action of the group or for each category or genre of the item selected in the group. In the example shown here, for example, it can be seen that the first terminal device 16a exerts a small influence in the scenes of “meal” and “drinking”, but exerts a strong influence in the scene of “shopping”. In addition, it can be seen that the third terminal device 16c exerts a comparatively strong influence in any scene. The first terminal device 16a to the fifth terminal device 16e shown here may be the terminal devices 16 that have actually acted as one group, or the terminal devices 16 that have not actually acted as one group. May be a comparison that collects That is, the user may specify a combination of arbitrary terminal devices 16 (users) and display the leader point. For example, one's family and friends may be designated and their leader points may be displayed so that they can be compared. In addition, each user specifies in advance whether or not his / her leader point can be disclosed, information to be disclosed, conditions of a partner to be disclosed, and the like, and stores the information in the information providing apparatus 10. May be disclosed to the user. Further, the user may arbitrarily select a group formed in the past and display the leader points of the terminal devices 16 constituting the group so as to be comparable. The terminal device 16 of the user who performs the operation may or may not be included in the group.

  FIG. 25 shows another screen displayed on the display unit 52 according to the fifth embodiment of the present invention. This corresponds to an example of displaying the leader points for each group member. The number of persons in the group is displayed in three stages of "2", "3 to 5", and "6 or more". It can also be said that leader points are displayed in categories determined according to the number of groups. The leader points shown here are calculated by the point calculation unit 42 for each of the terminal devices 16 to be displayed, by averaging the leader points acquired by the terminal devices in the predetermined period by the number of groups (for each of the three categories). This is a value normalized so that the total value of the average value for each group is “1”. For example, normalization is performed so that the total of the leader points of the first terminal device 16a to the fifth terminal device 16e corresponding to the group number “2” becomes “1”. In the example shown here, it can be seen that the first terminal device 16a has a large influence when the number of groups is two (small), but the influence is small as the number of groups increases. On the other hand, in contrast to the first terminal device 16a, the fifth terminal device 16e has a large influence when the number of groups is 6 or more (multiple people), but the influence is small when the number of groups is reduced. It turns out to be small. Further, it can be seen that the third terminal device 16c exerts a relatively strong influence in any scene regardless of the number of groups.

  FIG. 26 shows another screen displayed on the display unit 52 according to the fifth embodiment of the present invention. That is, the leader points acquired by each terminal device 16 during a predetermined period are classified into a plurality of categories (also referred to as leader point categories and Rp categories) based on a predetermined rule, and the frequency (frequency) for each category is displayed. This corresponds to one example. For example, when the value of the leader point (Rp) is “0.6 or more”, the point calculation unit 42 associates the value with the category “Rp large”, and when the value is “0.4 or more and less than 0.6”, the category “ Rp medium "and" less than 0.4 "correspond to category" Rp small ". Here, it is assumed that the leader points are normalized. On the other hand, when the leader point is not normalized, for example, when the value of Rp is “120 or more”, the point calculation unit 42 corresponds to the category “Lp large”. The category is set to "Rp medium", and if the category is "less than 50", the category is set to "Rp small".

  Further, the point calculation unit 42 counts the number of data corresponding to each of the three categories in the point storage unit 44 for each terminal ID. Of course, any other number of categories may be used. Such a leader point category can also be said to be an index that aggregates leader points. When the number of groups is small, the condition of “large Rp” is strictly set (the threshold value corresponding to large Rp is increased), and when the number of groups is large, the condition of “large Rp” is set. May be reduced (the threshold value corresponding to the large Rp is reduced).

  The above-mentioned predetermined rule is shown, for example, in FIG. FIG. 27 illustrates a data structure of a database stored in the group information processing unit 24 according to the fifth embodiment of the present invention. In this example, when the number of groups is "2", if the leader point (Rp) is "0.8 or more", the leader point (Rp) corresponds to "Rp large". If the value is less than 0.4, it corresponds to "low Rp".

  In FIG. 26, it can be seen that the first terminal device 16a has a high ratio of “Rp large” and exerts a strong influence in many group actions. On the other hand, it can be understood that the second terminal device 16b does not exert any influence in almost all group actions in which the ratio of “Rp large” is low. In addition, the third terminal device 16c has a large number of “Rp large”, but also has a similar number of “Rp small”. For this reason, it is understood that whether or not the third terminal device 16c exerts a strong influence is not fixed but is case-by-case. In FIG. 26, the frequency (absolute frequency) for each category is displayed. Is also good.

  In the above description, the information providing apparatus 10 executes the calculation of the characteristic action amount and the leader point, but other methods may be used. For example, the terminal device 16 exchanges positional information with the nearby terminal devices 16 as needed, and forms a group using its own positional information and the positional information of other terminal devices 16. Further, the characteristic action amount and the leader point may be calculated for the terminal devices 16 included in the group. That is, the position information management unit 22 and the group information processing unit 24 may be included in the terminal device 16. Each terminal device 16 in the group may calculate only its own leader point, or a specific terminal device 16 may calculate the leader points of a plurality of terminal devices 16 collectively. In addition, the terminal device 16 that has calculated the leader point transmits the leader point to another terminal device 16. Each of the terminal devices 16 in the group calculates a plurality of terminal devices by itself, or receives from another terminal device 16, or combines the one calculated by itself with the one received from another terminal device, and The leader point corresponding to No. 16 is acquired. The terminal device 16 uses the leader points to cause the display unit 52 to display the information shown so far.

  According to the present embodiment, since the information indicating the estimated influence is provided to the user, the user can objectively and quantitatively grasp the human relationship (implicit role, etc.) in the group. In addition, since the human relationships of the users in the group are objectively and quantitatively grasped, the users can easily improve their own actions and review the interpersonal relationships. In addition, since an index that aggregates a plurality of leader points is calculated, the estimation accuracy of the influence can be improved. In addition, at least one of a total value, a representative value, a maximum value, a minimum value, a value indicating variation, a frequency for each category determined according to the size of the leader point, and a relative frequency for each category is used as an index. , The influence can be grasped quantitatively by the user. In addition, since the index is calculated for each purpose of the action of the group or for each genre of the item selected in the group, it is possible to make the user grasp the influence in more detail. In addition, since the user is allowed to freely select a combination of the terminal devices 16 (users) and the leader points are displayed so as to be comparable, the user can easily grasp the influence.

  In addition, by providing information indicating the influence of the user in the group action, the influence of each other that the user does not necessarily realize can be known, and the user can enjoy the information as some kind of entertainment information. For example, "I thought that this group was controlled by Mr. A, but it turned out that Mr. B actually had a strong influence." Can be given. It will also help you review your position and role in the group. For example, "I'm not showing leadership at the mercy of other users in the group. I'll try to make suggestions a little more actively." Let's dwell on someone else's taste a little. " In other words, it is possible to provide information that leads to improvement of own behavior and improvement of human relations of the group. Further, since the group formation and the derivation of the leader point are performed by the terminal device, the influence can be estimated only by the terminal device.

(Example 6)
Next, a sixth embodiment will be described. The sixth embodiment relates to an information providing system including a plurality of terminal devices and an information providing device connectable to them, as in the past. The information providing apparatus according to the sixth embodiment also forms a group using the position information of a plurality of terminal devices and estimates the influence of each user in the group, as in the past. On the other hand, until now, the leader point has been calculated based on the characteristic behavior of the user in one group behavior. In the sixth embodiment, the leader point calculated in the past group behavior is used as the current leader point. Reflect on leader points in group actions. Specifically, the past leader points of other users are reflected on the leader points of a certain user (one user) in the group. The information providing system 100, the terminal device 16, and the information providing device 10 according to the sixth embodiment are of the same type as in FIG. 1, FIG. 13, FIG. 15, or FIG. Here, the difference will be mainly described.

  The point calculation unit 42 in FIG. 15 or FIG. 19 calculates the leader point as in the past. The information is stored in the group information processing unit 24. FIG. 28 illustrates a data structure of a database (group information table) stored in the group information processing unit 24 according to the fifth embodiment of the present invention. In the following description, the group ID of the processing target group (current group) is “G21”, and in the group information table, “G21” has three “C”, “D”, and “E”. It is assumed that the terminal device 16 is compatible. The three terminal devices 16 correspond to a third terminal device 16c, a fourth terminal device 16d, and a fifth terminal device 16e. Further, the leader points (temporary points) at a certain point in time are “C = 80”, “D = 100”, and “E = 60”.

It is assumed that a point history table shown in FIG. 23 is stored in the point storage unit 44 in the user information storage unit 28 of FIG. 15 or FIG. The point history table is a table in which past leader points for each terminal ID are shown, for example, a table in which calculation dates and times and leader points are associated with each other.
In the present embodiment, after executing step S110, the information providing apparatus 10 executes step S115 (not shown) before executing step S120. The detailed processing in step S115 will be described with reference to the flowchart in FIG. First, in step S300, the point calculation unit 42 ranks the terminal devices 16 in the target processing group in descending order of the leader points (temporary points). In the example described above, “D = 1st”, “C = 2nd”, and “E = 3rd”.

  Next, in step S310, the point calculation unit 42 acquires a past leader point for the terminal device 16 in the predetermined order. Here, the predetermined order is each order from the second place to a certain place (X place) below the second place. More specifically, the point calculation unit 42 first obtains a past leader point with reference to the point history table stored in the point storage unit 44 for the second-place terminal device. In the above example, the past leader point of the third terminal device 16c is obtained. In the example of FIG. 23, the point calculation unit 42 determines three (“G11, C) = 0.4”, “(G12, C) = 0.6”, and “(G13, C) = 0.8”. Get the data. X may be the second place, and the predetermined order may be only the second place. If X = third and the predetermined ranking is “2nd and 3rd”, the point calculation unit 42 further acquires the past leader points of the fifth terminal device 16e.

  In the example of FIG. 23, there are two of “(G11, E) = 0.05” and “(G12, E) = 0.15”. In addition, the point calculation unit 42 may acquire the past leader points limited to the group ID that does not include the terminal device 16 at the top. For example, when acquiring the past leader points for the second terminal device 16, the point calculation unit 42 acquires the terminal ID excluding the group ID including the fourth terminal device 16 d. In the above example, since the fourth terminal device 16d is included in G11, the point calculation unit 42 excludes G11, and sets “(G12, C) = 0.6” and “(G13, C) = 0.8. May be obtained. The point calculation unit 42 also obtains the past leader points of the third terminal device 16 that do not include the first terminal device 16. As described above, by acquiring past leader points limited to data that does not include the terminal device 16 at the top, the correction accuracy of leader points, which will be described later, may be improved.

  Next, in step S320, the point calculation unit 42 calculates a representative value of a past leader point for each terminal device 16 using the acquired leader point. Any of an average value, a median value, and a mode value can be used as the representative value, but in the following description, the average value is used. In the example described above, the average value of the third terminal device 16c is “(0.4 + 0.6 + 0.8) ÷ 3 = 0.6”. The average value of the fifth terminal device 16e is “(0.05 + 0.15) ÷ 2 = 0.1”.

  Next, in step S330, the point calculation unit 42 determines whether the calculated representative value satisfies a predetermined condition, and corrects the first leader point according to the determination result. Specifically, the point calculation unit 42 sets the representative value to be equal to or greater than the predetermined value θ1 as a predetermined condition, and increases the number of first leader points when the condition is satisfied. For example, when “θ1 = 0.3”, in the above-described example, the third terminal device 16c satisfies this condition, and the first-place leader point “D = 100” is multiplied by 1.5 to “D = 100”. 150 "or" 30 "is added to make" D = 130 ". That is, the point calculation unit 42 corrects the leader point by multiplying the first leader point by the predetermined coefficient or adding a predetermined number. In the case of the terminal devices 16 (users) in the second and lower ranks, if there are terminal devices 16 (users) with a higher leader point in the past, the influence of the terminal device 16 (user) with the higher leader point is higher than those. Can be estimated to be very strong. In other words, a person having a higher leader point than a person having a high leader point in the past can be estimated as a person having a very strong influence. For this reason, the point calculation unit 42 performs a process of setting the leader point higher than usual.

  In addition, the point calculation unit 42 may set a predetermined condition that the representative value is less than the predetermined value θ2, and may perform a process of reducing the number of first leader points when the condition is satisfied. For example, when “θ2 = 0.2”, in the above example, the fifth terminal device 16e satisfies this condition, and the first leader point “D = 100” is multiplied by 0.8 to obtain “D = 80, or "10" is subtracted to obtain "D = 90". In the case of the terminal device 16 (user) of the second or lower rank, if there is a terminal device 16 (user) having a low past leader point, the influence of the terminal device 16 (user) of the first rank in the current group is determined by other users. Is estimated to be stronger than the entity because of the weak influence. Therefore, the point calculation unit 42 performs a process of lowering the leader point than usual.

  Further, the point calculation unit 42 may change the degree of increasing or decreasing the leader points according to the number of the terminal devices 16 whose representative values are equal to or larger than the predetermined value θ1. For example, when the number of the terminal devices 16 whose representative values are equal to or more than θ1 is one, the point calculation unit 42 “1.5 times”. When the number of the terminal devices 16 is two, “two times”. In such a case, a process such as "3 times" may be performed. In addition, the point calculation unit 42 sets “double” when the representative value of the second-rank terminal device 16 is equal to or more than θ1, and sets “1.5” when the representative value of the third-rank terminal device 16 is equal to or more than θ1. “2 × 1.5 = 3 times” when both the second and third place terminal devices 16 are equal to or larger than θ1. That is, the point calculation unit 42 may change the degree of the leader point correction according to the order of the terminal devices 16 whose representative values satisfy the predetermined condition.

  In the above description, the leader point of the terminal device 16 in the first place is corrected. However, when the number of terminal devices 16 included in the group is large, the leader point of the terminal device 16 in the second place or less is similarly considered. The points may be corrected. For example, the leader points of the second terminal device 16 may be increased or decreased using the past leader points of the third terminal device 16. That is, the point calculation unit 42 calculates a leader point for one subject included in the group, based on a previously calculated leader point corresponding to another subject included in the group.

  According to the present embodiment, for one subject included in the group, the leader point is calculated based on the previously calculated leader points corresponding to the other subjects included in the group, so that the calculation accuracy is improved. Can be improved. In addition, since the leader points of the processing target group are corrected using the past leader points accumulated in the point history table, the influence of the user can be more accurately reflected on the leader points.

  The present invention has been described based on the embodiments. This embodiment is an exemplification, and it is understood by those skilled in the art that various modifications can be made to the combination of each component and each processing process, and that such modifications are also within the scope of the present invention. .

Embodiments 1 to 6 may be defined by the following items.
[Item 1]
A group forming unit that acquires position information of a subject that is a user or a terminal device, and forms a group that includes a plurality of subjects based on time-series data of the position information related to the plurality of subjects.
Based on the time-series data of the position information, for each of the plurality of subjects included in the group formed by the group forming unit, a predetermined behavior pattern is determined that the influence of the subject in the group is high. A point calculation unit that calculates a characteristic action amount indicating a degree of performing, and calculates a point for each subject based on the characteristic action amount;
An item selection unit that selects an item to be presented to the group, so that the point calculated by the point calculation unit has a higher priority.
An information providing device comprising:

[Item 2]
The point calculation unit determines at least one of the number of times, the matching time, the matching moving distance, and the matching probability that the movement pattern of the subject in the time-series data of the position information matches the predetermined action pattern. 2. The information providing apparatus according to item 1, wherein the point is calculated as the characteristic action amount, and the point is calculated such that the point becomes higher as the characteristic action amount increases.

[Item 3]
The information according to item 1 or 2, wherein the predetermined action pattern in the point calculation unit is a pattern in which a subject is located at a position of a predetermined order from the head of the group with reference to a moving direction of the group. Providing device.

[Item 4]
The information providing device according to item 3, wherein the point calculation unit calculates a point that is higher as the predetermined order is higher.

[Item 5]
The predetermined behavior pattern in the point calculation unit is a pattern in which, when a group is divided into a plurality of groups and then merges, the main body moves from the time of division in a direction in which the group moves after merging. 5. The information providing device according to any one of items 1 to 4.

[Item 6]
The information providing apparatus according to item 5, wherein the point calculation unit calculates points based on the number of subjects included in the small group when the group is divided into a plurality of groups.

[Item 7]
7. The information according to any one of items 1 to 6, wherein the predetermined action pattern in the point calculation unit is a pattern in which, when the subject stops, the other subjects also stop following the stop. Providing device.

[Item 8]
8. The information providing apparatus according to item 7, wherein the point calculation unit calculates points based on the number of other subjects stopped following the stop.

[Item 9]
9. The information providing apparatus according to any one of items 1 to 8, wherein the point calculation unit calculates points according to attribute information of a user corresponding to the subject.

[Item 10]
The information providing apparatus according to item 9, wherein the point calculation unit calculates points based on a plurality of items included in the attribute information of the user.

[Item 11]
10. The information providing apparatus according to item 9, wherein the point calculation unit calculates points according to a relationship between a date at which a process is executed and attribute information of the user.

[Item 12]
The information providing apparatus according to any one of items 1 to 11, wherein the group forming unit changes a condition for forming a group according to attribute information of a user corresponding to the subject.

[Item 13]
The group forming unit forms at least two subgroups by classifying the plurality of subjects included in the group into at least two subgroups,
The item selection unit may select an item to be presented for each subgroup so that a subject having a high point calculated by the point calculation unit is prioritized. Information providing device as described.

[Item 14]
The information providing device according to any one of items 1 to 13, wherein the point calculation unit calculates the points according to the number of subjects included in a group.

[Item 15]
The point calculation unit calculates the points for one subject included in the group based on previously calculated points corresponding to other subjects included in the group. 15. The information providing device according to any one of 1 to 14.

[Item 16]
The item selection unit (1) adds a preference degree for an item or a preference for an item attribute of a user corresponding to each subject while weighting by the points of each subject calculated in the point calculation unit, and The information providing apparatus according to any one of items 1 to 15, wherein a recommendation degree for each item is derived, and (2) an item is selected based on the recommendation degree for each item in the derived group.

[Item 17]
The item selection unit according to any one of items 1 to 16, wherein the item selection unit specifies a predetermined number of subjects in descending order of the points calculated by the point calculation unit, and selects an item based on the points of the specified subjects. Information providing device described in Crab.

[Item 18]
An acquisition unit that acquires position information of a subject that is a user or a terminal device;
A transmitting unit that transmits time-series data of the position information acquired by the acquiring unit to the information providing device,
A receiving unit that receives, from the information providing apparatus, information about an item selected in the information providing apparatus that has transmitted the time-series data of the position information in the transmitting unit;
A display unit for displaying information about the item received in the receiving unit,
The information on the item received by the receiving unit is, in the information providing device, (1) a group including a plurality of subjects is formed based on time-series data of position information related to the plurality of subjects; Based on the time-series data of the location information, a characteristic action indicating the degree to which each of a plurality of subjects included in the formed group matches a predetermined behavior pattern in which the influence of the subject in the group is high. Calculate the amount, calculate the points for each subject based on the characteristic action amount, and (3) generate items by selecting items to be presented to the group so that the subject with the calculated points has higher priority. A terminal device characterized in that the information is obtained information.

[Item 19]
Acquiring position information of a subject that is a user or a terminal device, and forming a group including a plurality of subjects based on time-series data of the position information related to the plurality of subjects,
Based on the time-series data of the position information, a characteristic indicating a degree of matching with a predetermined action pattern in which the influence of the subject in the group is high for each of the plurality of subjects included in the formed group. Calculating an activity amount, and calculating points for each subject based on the characteristic activity amount;
Selecting items to be presented to the group so that the subject with the calculated points has a higher priority;
An information providing method, comprising:

[Item 20]
Acquiring position information of a subject that is a user or a terminal device, and forming a group including a plurality of subjects based on time-series data of the position information related to the plurality of subjects,
Based on the time-series data of the position information, a characteristic indicating a degree of matching with a predetermined action pattern in which the influence of the subject in the group is high for each of the plurality of subjects included in the formed group. Calculating an activity amount, and calculating points for each subject based on the characteristic activity amount;
Selecting an item to be presented to the group so that the subject having the calculated points has a higher priority.

  Reference Signs List 10 information providing device, 12 network, 14 base station device, 16 terminal device, 20 receiving unit, 22 position information managing unit, 24 group information processing unit, 26 item selecting unit, 28 user information storing unit, 30 provided information storing unit, 32 transmission part, 34 control part, 40 group formation part, 42 point calculation part, 50 acquisition part, 52 display part, 54 control part, 56 transmission part, 58 reception part, 100 information providing system.

Claims (9)

A group forming unit that acquires position information of a subject that is a user or a terminal device, and forms a group that includes a plurality of subjects based on time-series data of the position information related to the plurality of subjects.
Based on the time-series data of the position information, for each of the plurality of subjects included in the group formed by the group forming unit, a predetermined behavior pattern is determined that the influence of the subject in the group is high. A point calculation unit that calculates a characteristic action amount indicating a degree of performing, and calculates a point for each subject based on the characteristic action amount;
An item selection unit that selects an item to be presented to the group, so that the point calculated by the point calculation unit has a higher priority.
With
The information providing, wherein the predetermined behavior pattern in the point calculation unit is a pattern in which, when a group is divided into a plurality of groups and then merged, the main body moves from the time of division in a direction in which the group moves after merging. apparatus. A group forming unit that acquires position information of a subject that is a user or a terminal device, and forms a group that includes a plurality of subjects based on time-series data of the position information related to the plurality of subjects.
Based on the time-series data of the position information, for each of the plurality of subjects included in the group formed by the group forming unit, a predetermined behavior pattern is determined that the influence of the subject in the group is high. A point calculation unit that calculates a characteristic action amount indicating a degree of performing, and calculates a point for each subject based on the characteristic action amount;
An item selection unit that selects an item to be presented to the group, so that the point calculated by the point calculation unit has a higher priority.
With
The information providing apparatus, wherein the predetermined action pattern in the point calculation unit is a pattern in which, when the subject stops, other subjects also stop following the stop. A group forming unit that acquires position information of a subject that is a user or a terminal device, and forms a group that includes a plurality of subjects based on time-series data of the position information related to the plurality of subjects.
Based on the time-series data of the position information, for each of the plurality of subjects included in the group formed by the group forming unit, a predetermined behavior pattern is determined that the influence of the subject in the group is high. A point calculation unit that calculates a characteristic action amount indicating a degree of performing, and calculates a point for each subject based on the characteristic action amount;
An item selection unit that selects an item to be presented to the group, so that the point calculated by the point calculation unit has a higher priority.
With
The information, wherein the point calculation unit calculates the points based on previously calculated points corresponding to other subjects included in the group, for one subject included in the group. Providing device. A group forming unit that acquires position information of a subject that is a user or a terminal device, and forms a group that includes a plurality of subjects based on time-series data of the position information related to the plurality of subjects.
The initial value of the point of each subject at the time when the group is formed by the group forming unit is set to zero, and for a predetermined period after the time, one subject included in the group based on the time-series data of the location information It is determined whether or not the position information of the group and the position information of the other subjects included in the group match any one of a plurality of types of patterns regarding the position information of two or more subjects. A point calculation unit that calculates a point for each subject included in the group by adding a characteristic action amount according to the type of the matching pattern to the point of the one subject;
An item that selects an item to be presented to a group based on a value obtained by adding the preference level for the item of each subject or the preference level for the item attribute while weighting the points of each subject calculated by the point calculation unit. A selection unit,
An information providing device comprising: One of the plurality of types of patterns is a pattern in which the one subject is located within a predetermined order from the head of the group, and the predetermined order becomes larger as the number of subjects included in the group increases. Is the value of
The point calculation unit uses a characteristic action amount that has a larger value as the ranking from the head of the one subject is higher,
The information providing apparatus according to claim 4, wherein: On the computer,
A group forming step of acquiring position information of a subject which is a user or a terminal device, and forming a group including a plurality of subjects based on time-series data of position information relating to the plurality of subjects,
Based on the time-series data of the position information, a characteristic indicating a degree of matching with a predetermined action pattern in which the influence of the subject in the group is high for each of the plurality of subjects included in the formed group. A point calculation step of calculating an activity amount, and calculating a point for each subject based on a value obtained by adding a plurality of the characteristic activity amounts in a predetermined period after a group is formed in the group formation step,
An item selecting step of selecting an item to be presented to the group so that the calculated points have higher priority.
The program is characterized in that, when the group is divided into a plurality of groups and merges, the predetermined behavior pattern is a pattern in which a subject moves from the time of division in a direction in which the group moves after merging. On the computer,
A group forming step of acquiring position information of a subject which is a user or a terminal device, and forming a group including a plurality of subjects based on time-series data of position information relating to the plurality of subjects,
Based on the time-series data of the position information, a characteristic indicating a degree of matching with a predetermined action pattern in which the influence of the subject in the group is high for each of the plurality of subjects included in the formed group. A point calculation step of calculating an activity amount, and calculating a point for each subject based on a value obtained by adding a plurality of the characteristic activity amounts in a predetermined period after a group is formed in the group formation step,
An item selecting step of selecting an item to be presented to the group so that the calculated points have higher priority.
The program is characterized in that the predetermined behavior pattern is a pattern in which, when the subject stops, other subjects also stop following the stop. On the computer,
A group forming step of acquiring position information of a subject which is a user or a terminal device, and forming a group including a plurality of subjects based on time-series data of position information relating to the plurality of subjects,
Based on the time-series data of the position information, a characteristic indicating a degree of matching with a predetermined action pattern in which the influence of the subject in the group is high for each of the plurality of subjects included in the formed group. A point calculation step of calculating an activity amount, and calculating a point for each subject based on a value obtained by adding a plurality of the characteristic activity amounts in a predetermined period after a group is formed in the group formation step,
An item selecting step of selecting an item to be presented to the group so that the calculated points have higher priority.
The program, wherein the point calculation step calculates, for one subject included in a group, the points based on previously calculated points corresponding to other subjects included in the group. . On the computer,
A group forming step of acquiring position information of a subject which is a user or a terminal device, and forming a group including a plurality of subjects based on time-series data of position information relating to the plurality of subjects,
The initial value of the point of each subject at the time when the group is formed in the group forming step is set to zero, and for a predetermined period after the time, one subject included in the group based on the time-series data of the position information It is determined whether or not the position information of the group and the position information of the other subjects included in the group match any one of a plurality of types of patterns regarding the position information of two or more subjects. A point calculation step of calculating a point for each subject included in the group by adding a characteristic action amount according to the type of the matching pattern to the point of the one subject;
An item for selecting an item to be presented to the group based on a value obtained by adding the preference level for the item of each subject or the preference level for the item attribute while weighting the points of each subject calculated in the point calculation step. A selection step;
A program that executes

Images