JP2013058176A - Server device and program for determining value of geographical region in communication service - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a server device for evaluating a value of each region in a communication service in correlation with an actual usage status of the communication service.SOLUTION: A server device obtains and stores plural pieces of log information from plural radio devices, applies a clustering analysis to a line interconnecting two positions indicated by the log information to divide a service provision area into plural regions, represents each of the plural regions as a node, causes the log information to be related to a link between nodes, assigns a value that becomes higher as the number of pieces of log information corresponding to the links is greater to links between individual regions, generates an adjacency matrix used in a network analysis, calculates a centricity of each region from the adjacent matrix, and calculates a price of a region so as to increase with the centricity of the region. Further, the log information at least includes a first log category including information of a position of a radio device and information indicating an interested position of a user determined by an operation of the user on the radio device.

Description

  The present invention relates to a technique for evaluating the value of a geographical area for information presentation in a communication service.

  Currently, wireless devices that communicate via a wireless network, such as mobile phones, are equipped with camera functions, positioning functions using the global positioning system (GPS), and acceleration and geomagnetic sensors. Augmented reality (AR) services that are used are provided. In the augmented reality service, for example, when a user photographs a signboard of a newly released music album (CD) using the camera function of the wireless device, a preview button or a purchase button of the music album is displayed on the display. Is displayed, and the user can view each song by tapping the preview button, and actually purchase the music album by tapping the purchase button.

  As described above, in the augmented reality service, an advertisement related to a certain location or geographical area is stored in advance in the device for augmented reality service on the network, and the direction in which the user is shooting from the wireless device is specified. When the data for receiving is received, the device for the augmented reality server transmits information for displaying the advertisement included in the shooting area to the wireless device. Here, it is necessary to evaluate the value of each position or area in the advertisement display (hereinafter, the value of the geographical area in the communication service is expressed as a land price) by some method.

  Here, Patent Document 1 discloses a configuration in which the position of each position is ranked by calculating the movement vector of each wireless device using the positioning function of the wireless device.

JP 2007-156737 A

  However, in the method described in Patent Document 1, it is only possible to obtain information about a place where people gather, and simply using a place where people gather as a basis for land price calculation in augmented reality service It is insufficient as a standard. This is because in augmented reality services, people who actually use wireless devices as augmented reality service terminals, that is, launch applications that support augmented reality services, shoot with camera functions, and display advertisements on the display. If the percentage of people who do it is small, even if many people gather, it has no meaning.

  That is, it is desired to determine the value of each geographical area in correspondence with the actual use situation of augmented reality service. In addition to augmented reality systems, in various communication services such as so-called location games and position-linked advertisements, it is desired to evaluate the value for a certain geographical area in accordance with the usage status of the communication service. .

  Therefore, the present invention provides a server device and a program for evaluating the value of each geographical area in the communication service in correspondence with the actual usage situation of the communication service.

  The server device according to the present invention includes a storage unit that acquires and stores a plurality of log information having information indicating two positions in a service providing area from a plurality of wireless devices, and logs each of the plurality of log information. An area dividing means for dividing a service providing area into a plurality of areas by performing a clustering analysis on a plurality of lines corresponding to the plurality of log information in association with a line connecting two positions indicated by the information; Each of the plurality of areas is a node, and the log information is a link between the first area including one position included in the log information and the second area including the other position included in the log information. And a matrix generating means for generating an adjacency matrix in network analysis by assigning a value that increases as the number of log information corresponding to the link increases to the link between the areas, A centrality calculating means for calculating the centrality of each area from the adjacency matrix; and a price calculating means for calculating the price of the area so as to be higher as the centrality of the area is higher. And at least a first log type including information indicating a position of the wireless device and a position of interest of the user determined from an operation by the user of the wireless device.

  The area is divided by log information including the position of interest of the user when the wireless device is actually receiving the communication service, and the centrality is obtained. An area with a large centrality at this time can be regarded as an area that many users pay attention to in the communication service, and therefore, the value of the geographical area should be evaluated according to the actual usage status of the communication service. Can do.

The schematic block diagram of the land price determination system by one Embodiment. 1 is a schematic configuration diagram of a wireless device according to an embodiment. FIG. Explanatory drawing of an area | region. Explanatory drawing of area division. Explanatory drawing of a link matrix.

  In the following, the present embodiment will be described assuming that the communication service is an augmented reality service. FIG. 1 is a schematic configuration diagram of a land price determination system according to this embodiment. The system includes a server device 100 and a plurality of wireless devices 200, and each wireless device 200 and server device 100 communicate via a wireless network 300. It is possible. The wireless network 300 and each wireless device 200 communicate wirelessly, but the communication between the server device 100 and the wireless network 300 may be wired or wireless. In the wireless network 300, when the wireless device 200 starts up an application corresponding to the augmented reality system and operates as a terminal of the augmented reality system, the wireless device 200 displays an advertisement or the like on the wireless device 200. A device for transmitting the data is also connected, but is omitted from FIG. 1 because it is not necessary for the description of the present invention.

  FIG. 2 is a schematic configuration diagram of the wireless device 200. The wireless device 200 includes a position determination unit 22 that determines the position of the wireless device 200 using GPS or the position of a base station or access point of the wireless network 300 that is communicating wirelessly, an acceleration sensor 23, and a geomagnetism. A sensor 24 is provided. As is well known, the acceleration sensor 23 and the geomagnetic sensor 24 are used to determine the current direction of the wireless device 200, for example, the direction in which the camera lens mounted on the wireless device 200 is facing. Can do. Further, the operation log acquisition unit 25 acquires the operation content of the user's wireless device 200.

  The log transmission unit 21 repeatedly transmits log information to the server device 100 based on information from the position determination unit 22, the acceleration sensor 23, the geomagnetic sensor 24, and the operation log acquisition unit 25. In addition, the repetition interval etc. are arbitrary. Here, the log information to be transmitted is classified into two types: a movement log and an interest log.

  The movement log is log information including two positions, the first position and the last position of the wireless device 200 in a predetermined period. That is, the position determination unit 22 is information including the position of the wireless device 200 measured last time and the position of the wireless device 200 measured this time. For example, if the previous latitude and longitude of the wireless device 200 are (X1, Y1) and the current latitude and longitude (X2, Y2) of the wireless device 200, the movement log includes the position (X1, Y1) and the position The information includes (X2, Y2). The wireless device 200 repeatedly transmits this movement log to the server device 100. Note that the measurement cycle of the position determination unit 22 is arbitrary. In addition, when the log transmission unit 21 continuously acquires position information from the position determination unit, the log transmission unit 21 can select the position information at a predetermined timing and use it as log information. In the following description, the position is displayed in (latitude, longitude) unless otherwise specified.

  The interest log refers to the position of the wireless device 200 and the focus position on which the camera is focused when the user uses the wireless device 200 as a terminal of the augmented reality system and actually points the camera in a certain direction. It is information to include. The focus position can be regarded as a position (interest position) in which the user is interested, which is determined from a predetermined operation on the wireless device 200 of the user. That is, for example, when the user focuses the camera on the object at the position (X4, Y4) at the position (X3, Y3), the interest log information includes the position (X3, Y3) and the position (X4, Y4). ) Is included. The position (X4, Y4) can be determined from the orientation of the camera lens, the camera operation information acquired by the operation log acquisition unit 25, and the current position (X3, Y3). In the case of an interest log, information about a period in which the user focuses the camera on the target at the position (X4, Y4) can also be included in the log information. In addition, a user who uses the wireless device 200 as a terminal of an augmented reality system performs some operation on information such as an advertisement actually displayed on the display by the augmented reality system, for example, a purchase operation of a product. In such a case, information indicating the operation content can also be included in the log information.

  Next, returning to FIG. 1, each function of the server apparatus 100 will be described. The log collection unit 15 acquires log information transmitted by the plurality of wireless devices 200 and stores the log information in the log storage unit 16. The area dividing unit 13 first divides a service providing area, for example, the whole of Japan, into a plurality of areas based on the log information stored in the log storage unit 16 at the time of activation. At this time, the area dividing unit 13 uses log information for a predetermined period, for example, the past week, one month, or one year. The area division will be described in detail below.

  FIG. 3 is an explanatory diagram of each region generated by the region dividing unit 13. First, in this embodiment, the area can take a hierarchical structure. In FIG. 3, reference numeral 4 represents the entire service providing area such as Japan, which is first divided into four areas A, B, C, and D. These areas A, B, C, and D are referred to as the uppermost area (primary area). Note that each area in FIG. 3 does not represent an actual size, and each area does not have to be the same size. Further, in FIG. 3, the region D is further divided into 16 regions, which is referred to as a secondary region. Further, the secondary area D16 is divided into 16 tertiary areas.

  Next, how the area dividing unit 13 divides the entire service providing area into areas as shown in FIG. 3 will be described with reference to FIG. As described above, the log information from the wireless device 200 includes information indicating two positions, and a line connecting the two positions is considered. Each log information is given a related weight value. For example, the movement log has a weight value of 1, the interest log has a weight value of 2, and if the interest log includes information indicating that the user has performed an operation such as purchase, the weight value is higher. For example, the weight value is 3. Note that these weight values are examples, and for example, the interest log can be set to a higher value as the period during which the user of the wireless device 200 focuses the camera on the object is longer. However, the weight value of the interest log is set higher than that of the movement log, and in the interest log, when information indicating that the user has performed an operation such as purchase is added, the weight value is set higher than when the information is not added. It is desirable to do. Each line in FIG. 4 corresponds to log information, and the value of each line represents a weight value.

  The region dividing unit 13 determines the highest region by performing a clustering analysis on the line shown in FIG. 4 by a so-called X-MEANS method. Specifically, first, two positions (points) are appropriately determined, and each line is divided into two groups according to which of the two points is closer. Subsequently, a point at which an average distance from a line belonging to the same group is closest is obtained for each group. The lines are grouped again based on the distance between the two points thus obtained. This process is repeated until the positions of the two points converge to divide each line into two groups. FIG. 4 shows a state in which the beam finally converges on points 51 and 52 and is divided into a region 41 including five lines and a region 42 including three lines. It should be noted that a line straddling the region 41 and the region 42 is included in the region 41.

  The distance between the point and each line can be, for example, the distance between the point and the center of the line. Moreover, it can be set as the shorter one of the two distances from the point to the two end points of the line. Alternatively, a perpendicular line can be drawn from the point to the line, and the length of the perpendicular line can be the distance. In addition, after convergence to two points in the above clustering, it is necessary to determine the boundary between the regions 41 and 42. Each region has one of the points corresponding to the region and all the lines belonging to the region. As long as the end point (that is, one position of the log information) is included, the boundary of the region can be arbitrarily determined.

  Subsequently, the region dividing unit 13 determines whether or not each region obtained by the division meets the division stop condition. In this embodiment, the division stop condition is a case where the sum of the weight values of the lines belonging to each region is equal to or less than a threshold value. For example, if the threshold value is 4, the total of the weight values of the lines belonging to the area 42 is 3, so that the division stop condition is met. Therefore, the area 42 is not further divided. On the other hand, since the total of the weight values of the lines belonging to the area 41 is 8, the division stop condition is not met, and the area 42 is again divided into two by the method described above. When all the areas meet the division stop condition, the division process is stopped there, and the highest area is determined. Even if all the areas do not meet the division stop condition, the division processing may be stopped at that time when the area is divided into a predetermined number of areas (for example, 256).

  Subsequently, the area dividing unit 13 determines whether or not each uppermost area is further divided into secondary areas. At this time, a link matrix generated by the matrix generation unit 14 is used. FIG. 5 is a diagram illustrating a link matrix corresponding to the highest-order matrix. The link matrix corresponds to a weighted adjacency matrix in network analysis. In other words, each uppermost area is set as a node, and two positions indicated by log information are associated with links between the areas. Specifically, when the first position of the log information is included in the first area and the second position of the log information is included in the second area, the log information is stored in the first area and the second area. Corresponding to the link connecting the areas. Then, the sum of the weight values of all the log information corresponding to the link connecting the first area and the second area is the value of the link connecting the first area and the second area, that is, the first The element value of the column corresponding to the second area of the row corresponding to this area and the value of the element of the column corresponding to the first area of the row corresponding to the second area. For example, when there is no log information corresponding to the link between the third area and the fourth area, the value of the link is set to 0.

For example, assuming that the uppermost area is four areas A, B, C, and D as shown in FIG. 3, the link matrix is a 4 × 4 matrix, and the element n _ij of i rows and j columns is , One position is the area i, and the other position is the sum of the weight values of the log information included in the area j. That is, it is the sum of the weight values of the log information corresponding to the links extending over the areas i and j. Note that the element n _ii in the i row and i column is the sum of the weight values of the log information corresponding to the links closed in the same area. Specifically, in FIG. 5, the sum of the weight values of log information in which two positions are both in the area A is 5, one position is in the area A, and the other position is in the area B. It is indicated that the total weight value of the log information is 2. Here, the diagonal component in FIG. 5 is log information in which the two positions are both in the same region, and a large value means that a large amount of log information exists in the same region. Therefore, the area where the value of the diagonal component is equal to or greater than the predetermined value is further divided in order to average the number of logs in the area. For example, when the division threshold is 10, only the region D is divided into secondary regions in FIG.

  The division into the secondary regions is performed, for example, by dividing the original region into a predetermined number (for example, 256) of regions at equal intervals, that is, the same size. For the secondary region thus formed, the link matrix generating unit 14 obtains a link matrix corresponding to the secondary region, and the region dividing unit 13 determines whether or not each region is divided into tertiary regions. The determination is made in the same manner as the division into the secondary area. For example, when the area A in FIG. 3 is also divided into secondary areas in the same manner as the area D, the link matrix generation unit 14 generates the link matrix of the secondary area of the area A and the secondary area of the area D. Each link matrix is generated. As described above, the region dividing unit 13 performs region division by repeating the division into lower regions until the diagonal component of the link matrix becomes less than a predetermined value.

  In this way, the matrix generated by the link matrix generation unit 14 indicates the degree of connection between the areas in the augmented reality system in addition to the movement of people between the areas, and has high centrality in network analysis. The region has a higher value as a position or a region for advertisement presentation. The centrality includes the degree centrality, etc. In this embodiment, for example, the eigenvector centrality proposed by Bonachic is used. Thus, the centrality calculation unit 12 obtains an eigenvector corresponding to the maximum eigenvalue of each order link matrix generated by the matrix generation unit 14, and sets the value of each element of the eigenvector as the centrality value of the corresponding region. For example, if the uppermost region (primary region) in FIG. 3 is a 4 × 4 matrix and its eigenvector is (M1, M2, M3, M4), the center of regions A, B, C, and D The sex values are M1, M2, M3, and M4, respectively. The secondary region of the region D is a matrix of 16 rows and 16 columns, and if the eigenvector is (L1, L2,..., L16), the centrality value of the region D1 is L1, and the center of the region D16 The sex value is L16. Further, the tertiary region of the region 16 is a matrix of 16 rows and 16 columns, and when the eigenvector is (L17, L18,..., L32), the centrality value of the region D17 is L17.

And the price calculation part 11 calculates the price basic value for determining the price of each area | region,
1 / (| k1 * logC | * | k2 * logC '| * | k3 * logC''| ...)
Ask for. Here, C is the centrality value of the uppermost region, C ′ is the centrality value of the secondary region, and C ″ is the centrality value of the tertiary region. K1, k2, and k3 are coefficients indicating weights. For example, in the above example, the price base value of the region D17 is
1 / (| k1 * logM4 | * | k2 * logL16 | * | k3 * logL17 |)
It becomes. Note that the calculation is performed assuming that the centrality value of the tertiary region is 1/16, which is the number of divisions, for the case where the division is stopped in the secondary region. That is, the price base value of the region D1 is
1 / (| k1 * logM4 | * | k2 * logL1 | * | k3 * log (1/16) |)
It becomes. Similarly, for the case where the division is stopped in the primary region, the calculation is performed assuming that the centrality values of the secondary region and the tertiary region are each 1/16 that is the number of divisions. In other words, the price base value of area A is
1 / (| k1 * log M1 | * | k2 * log (1/16) | * | k3 * log (1/16) |)
It becomes.

  The price basic value is an index indicating the value in presenting the advertisement in the area, and the higher the value, the higher the value. And the price calculation part 11 calculates the price for advertisement presentation of each area | region by multiplying the price basic value of each area | region, for example with a predetermined coefficient.

  In the present embodiment, the region dividing unit 13 repeats the above processing every predetermined period, for example, every day, and re-divides the region, whereby the centrality calculating unit 12 and the price calculating unit 11 Update. Note that the log information used at this time is the log information for a predetermined period, for example, the past one week, one month, or one year, as described above. With this configuration, it is possible to set prices according to changes in user behavior. Further, in the present embodiment, it is determined whether or not further division is performed based on the weight value of the diagonal component of the link matrix. As a result, the area where log information is small can be divided unnecessarily, thereby preventing an unnecessary increase in load such as matrix processing, so that pricing can be performed dynamically. It becomes possible.

  The server device according to the present invention can be realized by a program that causes a computer to operate as the server device described above. These computer programs can be stored in a computer-readable storage medium or distributed via a network. Furthermore, the present invention can be realized by a combination of hardware and software. In the above-described embodiment, the log information is an undirected link in order to correspond to the link. However, the log information may be a directed link. Furthermore, although the price has been determined by eigenvector centrality, other centralities can be used. In addition, the area dividing unit 13 divides the service providing area using the X-MEANS method, but other clustering analysis such as the K-MEANS method can also be used.

  Although the description has been given based on the augmented reality service, the present invention can be applied to a communication service such as a position game and a position-linked advertisement, in which a user performs an operation on information regarding a certain position with respect to the wireless device 200. For example, in the case of a position-linked advertisement, the position associated with the advertisement when the user performs an operation of browsing the displayed advertisement at a certain position can be set as the position of interest in the interest log information. Further, in the case of a position game, when the user performs an operation on a certain position displayed on the wireless device 200 regarding the game, the position may be set as the position of interest. That is, the present invention can be provided to a communication service in which a user performs an operation related to a position with respect to the wireless device 200, and the position of the operation target is the position of interest.

Claims (12)

Storage means for acquiring and storing a plurality of log information having information indicating two positions in a service providing area from a plurality of wireless devices;
By associating each of the plurality of log information with a line connecting two positions indicated by the log information, and performing clustering analysis on the plurality of lines corresponding to the plurality of log information, Area dividing means for dividing the area;
Each of the plurality of areas is a node, and log information is between a first area including one position included in the log information and a second area including the other position included in the log information. A matrix generating means for generating an adjacency matrix in network analysis by allocating a value that increases as the number of log information corresponding to the link increases to the link between each region,
Centrality calculating means for calculating the centrality of each region from the adjacency matrix;
A price calculation means for calculating the price of the area so that the price becomes higher as the centrality of the area is higher;
With
The log information includes at least a first log type including information indicating a position of the wireless device and a position of interest determined from an operation of the wireless device by a user of the wireless device.
Server device. Each of the plurality of log information is assigned a weight value,
The value assigned to the link is the sum of the weight values of log information corresponding to the link.
The server device according to claim 1. The log information includes a second log type having information indicating two positions of a first position and a last position of the wireless device in a predetermined period.
The server device according to claim 1 or 2. The region dividing unit further divides a region corresponding to a diagonal component larger than the threshold among the diagonal components of the adjacent matrix generated by the matrix generating unit into a predetermined number of secondary regions.
The server device according to any one of claims 1 to 3. The matrix generation means generates an adjacency matrix corresponding to the secondary region,
The region dividing means further divides a region corresponding to a diagonal component larger than the threshold among diagonal components of an adjacency matrix corresponding to the secondary region into the predetermined number of tertiary regions.
The server device according to claim 4. The area dividing means divides the service providing area into a plurality of areas by using the X-MEANS method.
The server device according to any one of claims 1 to 5. The area dividing unit determines an area to which each of the plurality of lines belongs by using the X-MEANS method, assigns a value that increases as the number of lines belonging to the area increases, to the area, When the assigned value is equal to or less than a predetermined value, the division of the area is stopped.
The server device according to any one of claims 1 to 6. The centrality calculated by the centrality calculating means is eigenvector centrality.
The server device according to any one of claims 1 to 7. When the log information is the first log type, the log information includes information on whether or not the user has performed an operation on the information on the position of interest of the user displayed on the display of the wireless device.
The weight value of the log information of the first log type when the user performs an operation is higher than the weight value of the log information of the first log type when the user does not perform an operation.
The server device according to any one of claims 1 to 8. The position of interest is a focus position on which a camera of a wireless device using an augmented reality service is focused.
The server apparatus of any one of Claim 1 to 9. When the log information is the first log type, the log information includes information indicating a period during which the focus position is focused;
The weight value of the log information of the first log type is higher as the period is longer.
The server device according to claim 10.   A program that causes a computer to function as the server device according to any one of claims 1 to 11.

Images