JP2015011107A - Population change decision device, and population change decision method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a population change decision device in which the degree of change of population is accurately decided, even when location information is not acquired sufficiently.SOLUTION: A location information acquisition unit 11 acquires location information showing the location of people associated with time. The number of people calculation unit 13 calculates the degree of the number of people every time zone and area, based on the distribution of people according to the distance from the location shown by the location information acquired by the location information acquisition part 11. A target period number of people acquisition unit 15 acquires the degree of the number of people in a target period of an area of interest based on the degree of the number of people calculated by the number of people calculation unit 13. A past period number of people acquisition unit 16 acquires the degree of the number of people in a past period from the target period of the area of interest based on the number of people calculated by the number of people calculation unit 13, and decides the change degree of the number of people in the target period by using the degree of the number of people acquired by the number of people acquisition unit 15 during the target period and the degree of the number of people acquired by the number of people acquisition 18 during the past period, and outputs information based on the results decided.

Description

  The present invention relates to a population change determination device and a population change determination method.

  In recent years, services using location information have attracted attention as services for mobile phones. In particular, with the spread of smartphones equipped with GPS (Global Position System) receivers, service providers use services that utilize location information based on the positioning results of smartphone GPS receivers (locations around smartphone users and their surroundings). Location information).

  Specifically, the service provider can provide information on popular spots and popular places by using not only general spot guidance but also location information of each mobile phone.

  Generally, if there are events, etc., it is thought that the number of people will increase rapidly, so the service provider identifies the mesh where the population (congestion level) of the mesh unit estimated based on the location information of the mobile phone has increased rapidly, It is conceivable that the mesh is the above-mentioned popular spot or popular place.

  For example, the information providing apparatus described in Patent Document 1 manages current number information at each counting position, determines a rapid increase in population based on the number information, and displays a notice of a place where the population has increased rapidly.

JP 2009-98446 A

  Since the information providing apparatus described above determines a rapid increase in the population based on the positioning result corresponding to the total position, it is assumed that the number of positioning results that can estimate the population at the total position are acquired. Therefore, there is a problem that when the position information indicating the position of a person is not sufficiently obtained, the degree of change in the population (for example, a rapid increase) may not be accurately determined.

  For example, in an area where you want to determine the degree of population change, if location information (for example, GPS positioning results) for only some people is not available (if the positioning results that can be acquired are sparse), Cannot be judged.

  The present invention has been made to solve such a problem, and provides a population change determination device that can accurately determine the degree of population change even when position information cannot be obtained sufficiently. For the purpose.

  The population change determination device according to the present invention includes a position information acquisition unit that acquires position information indicating a person's position associated with time, and a distance from the position indicated by the position information acquired by the position information acquisition unit. Number of people calculation means for calculating the degree of the number of people for each time zone and area based on the distribution of the corresponding person, and the target period for acquiring the degree of the number of people in the target period of the target area based on the degree of the number of people calculated by the number of people calculation means The number of people acquisition means, the number of people in the past period that acquires the degree of the number of people in the past period from the target period of the area of interest based on the number of people calculated by the number of people calculation means, The degree of change judgment means for judging the degree of change of the number of people in the target period using the degree of number of people acquired by the period number of people acquisition means, and the result of judgment by the degree of change judgment means Includes a change degree outputting means for outputting the basis information, the.

  Further, the population change determination method according to the present invention is a population change determination method by a population change determination device, and includes a position information acquisition step for acquiring position information indicating the position of a person associated with time, and position information. Based on the distribution of people according to the distance from the position indicated by the location information acquired in the acquisition step, the number of people calculation step to calculate the degree of the number of people in each time zone and area, and the degree of the number of people calculated in the number of people calculation step Based on the number of people calculated in the target number calculation step, the number of people in the target period in the target period based on the target period, and in the past period number of people acquisition step in the past period from the target period in the target area based on the number of people calculated in the number of people calculation step And the degree of the number of people acquired in the target period number of people acquisition step and the degree of the number of people acquired in the previous period number of people acquisition step, Including a change degree determination step of determining the degree of change in number of elephants period, the change degree output step of outputting information based on the result of the determination by the change degree decision step.

  According to such a population change determination apparatus and population change determination method, the degree of the number of people in the area is calculated using the distribution of users, so the degree of the number of people in the target area is measured only from the position information. Even if it is difficult to do so, the degree of the number of people in the target area can be estimated, and the degree of change in the number of people in the target period can be accurately determined. In other words, according to the population change determination device and the population change determination method, it is possible to accurately determine the degree of population change even when position information cannot be acquired sufficiently.

  In the population change determination device according to the present invention, the number calculation means calculates a contribution rate to each area for each position corresponding to the position information based on the distribution, and uses the contribution rate to calculate the time zone and the area. It is good also as calculating the degree of the number of people for every. According to this configuration, since the degree of the number of persons is calculated in consideration of the position spread of the position information, it is possible to accurately determine the degree of change of the number of persons.

  In the population change determination device according to the present invention, the number-of-people calculating means is based on the distribution of the number of position information included in a predetermined area and the number of position information included in an area around the predetermined area, respectively. It is good also as calculating the degree of the number of people for every time zone and area by weighting. According to this configuration, since the number calculation means calculates the degree of the number of persons in consideration of the position spread of the position information, the number of persons can be accurately determined.

  In the population change determination device according to the present invention, the distribution may be determined based on the accuracy of each piece of position information. According to this configuration, in the population change determination device, the degree of the number of persons is calculated in consideration of the displacement of the position information due to the positioning error of the position information. Therefore, the degree of the number of persons can be accurately calculated.

  In the population change determination device according to the present invention, the distribution may be determined based on the geographical shape of each position. According to this configuration, since the population change determination device calculates the degree of the number of people using the distribution based on the geographical shape, for example, the position information included in the area considered that there is no person is The degree can be excluded from the calculation target. Therefore, the population change determination device can accurately calculate the degree of the number of people.

  The population change determination device according to the present invention further includes a determination criterion setting unit that sets a determination criterion based on the degree of the reference number of persons in the area of interest, and the change degree determination unit is based on the determination criterion. The degree of change in the number of persons in the target period may be determined using the degree of the number of persons acquired by the period number of persons acquisition means and the degree of the number of persons acquired by the number of persons in the past period. According to this configuration, since the population change determination device determines the degree of change in the population using the determination criteria for the position of the determination target, it is simply determined whether the population has changed based on the degree of change. Compared with, it is possible to accurately determine the degree of change in the population.

  According to the present invention, it is possible to accurately determine the degree of population change even when position information cannot be acquired sufficiently.

It is a figure which shows the function structure of the population change judgment apparatus which concerns on embodiment of this invention. It is a figure which shows the example which calculates a congestion degree based on a positioning position. It is a figure which shows the example which considered that a predetermined point contributes to a some mesh. It is a figure which shows the example which considered that a predetermined point contributes to a some mesh. It is a figure which shows the example which calculates a congestion degree with the 2nd congestion degree calculation method. It is a figure which shows the example which calculates a congestion degree with the 2nd congestion degree calculation method. It is a figure which shows the data stored in a mesh congestion degree memory | storage part. It is a figure which shows the relationship between the object period which concerns on embodiment of this invention, and a normal period. It is a figure which shows the hardware constitutions of the population change judgment apparatus which concerns on embodiment of this invention. It is a flowchart which shows the process of the method of calculating the degree of the 1st person based on embodiment of this invention. It is a flowchart which shows the process of the method of calculating the degree of the 2nd number of persons based on embodiment of this invention. It is a flowchart which shows the process which calculates the population change degree which concerns on embodiment of this invention.

  Embodiments according to the present invention will be described below with reference to the drawings. Where possible, the same parts are denoted by the same reference numerals, and redundant description is omitted.

(Configuration of Population Change Determination Device 10)
FIG. 1 shows a population change determination device 10 according to this embodiment. The population change determination device 10 is an information processing device such as a server. The population change determination device 10 is a device that determines the degree of change in population at a predetermined position (for example, an area). More specifically, the population change determination device 10 calculates the degree of the number of people in a predetermined time zone based on the position information acquired from each user terminal 50, and the degree of change of the population (based on the degree of the number of people) For example, it is determined whether or not the population is rapidly increasing. The user terminal 50 is a terminal device that can be carried by a user such as a mobile phone. The position information is information including at least information indicating a position (for example, latitude and longitude), time at which the position is specified, and accuracy information (for example, information on positioning accuracy) of information indicating the position. Then, the population change determination device 10 outputs, for example, a determination result of the degree of population change to an information service providing device (not shown). Based on the determination result acquired from the population change determination device 10, the information service providing device transmits information regarding a place where the population change degree is high to the service provision target device.

  As shown in FIG. 1, the population change determination device 10 is connected to a user terminal 50 via a network such as a mobile communication network. The population change determination device 10 can transmit and receive information to and from the user terminal 50 using a network. The user terminal 50 has a device (for example, a GPS receiver) that can specify the position of the user terminal (own terminal) 50. The user terminal 50 acquires the position information of the user terminal 50 (for example, information based on the positioning result by the GPS receiver) using the device that can identify the position, and sends the position information to the population change determination device 10. Send.

  The population change determination device 10 includes a position information acquisition unit 11 (position information acquisition unit), a map information storage unit 12, a number of people calculation unit 13 (number of people calculation unit), a mesh congestion degree storage unit 14, and the acquisition of the number of people for a target period. Unit 15 (target period number acquisition means), past period number acquisition unit 16 (past period number acquisition means), reference number acquisition unit 17 (reference number acquisition means), and determination reference generation unit 18 (determination reference setting means) And a change degree determination unit 19 (change degree determination unit) and a change degree output unit 20 (change degree output unit).

  The position information acquisition unit 11 acquires position information indicating the position of the person (the position of the user terminal 50) associated with the time. Specifically, the position information acquisition unit 11 acquires position information of the user terminal 50 carried by a person (user) by receiving the position information of the user from the user terminal 50 via the network. The position information acquisition unit 11 may acquire the position information from the server device that has collected the position information of the user terminal 50.

  The map information storage unit 12 stores map information. Here, the map information is information in which position information and terrain information are associated with each other, for example, identification information (ID, name, etc.) of a map object (for example, a mountain, a lake, a road, etc.) This is information in which coordinates at which the map object is arranged are associated with information capable of determining whether the map object is indoor or outdoor.

  Further, the map information storage unit 12 also holds information regarding meshes as map information. Here, the mesh refers to a range obtained by dividing an area into a mesh shape (rectangular shape) based on latitude and longitude. Information on the mesh includes information about the mesh range (for example, information on the lower left latitude and longitude of the mesh and information on the upper right latitude and longitude), and coefficient information based on whether or not there is a person in the mesh (for example, mountains and lakes) The coefficient 0 is set for the mesh occupied by the majority of the map objects that are considered to have no people, and the coefficient 1 is set for the other meshes). Furthermore, the map information storage unit 12 also holds weighting information between a predetermined mesh and a mesh around the mesh (for example, a mesh in contact with the predetermined mesh) as information about the mesh.

  The number calculation unit 13 is based on the distribution of users (users, people) of the user terminal 50 according to the distance from the position indicated by the position information acquired by the position information acquisition unit 11, and the number of persons for each time zone and mesh. The degree of is calculated. The number calculation unit 13 calculates the number of position information (positioning results) belonging to the mesh in a predetermined time unit (for example, 1 hour) as the degree of the number of people for each time zone and each mesh. Hereinafter, the degree of the number of people for each time zone and each mesh is referred to as the degree of congestion. Then, the number calculation unit 13 registers the calculated congestion degree in the mesh congestion degree storage unit 14 described later.

  As a method of calculating the degree of congestion, the contribution rate to each mesh is calculated for each position corresponding to the position information based on the distribution (probability distribution) of the user's position, and the degree of congestion is calculated using the contribution rate. A distribution method (hereinafter referred to as a first congestion degree calculation method), the number of position information included in a predetermined mesh, and the number of position information included in a mesh around the predetermined mesh are distributed. There is a method of calculating the degree of congestion by weighting based on this (hereinafter referred to as a second congestion degree calculating method). Each method will be described below. The above-described distribution of the user's position refers to a distribution indicating the possibility that the user exists in another place when the user exists at the reference position. Further, the contribution rate indicates the degree of the number of persons existing in the mesh based on the position information when position information corresponding to a certain position exists. This is because the population has a spatially smooth value. That is, when a person is present at a certain location, a smooth distribution of the degree of the number of people is formed around the location based on the distance from the center.

(First congestion degree calculation method)
Before explaining the first congestion degree calculation method, an example of calculating information (number of persons) indicating whether or not there are a large number of persons based on the positioning positions included in the mesh will be described with reference to FIG. As shown in FIG. 2, there are meshes M <b> 1 to M <b> 3, and positioning positions P <b> 1 to P <b> 5 indicating the positions of positioning results by the user terminal 50 belong to the mesh.

  Specifically, the mesh M1 includes positioning positions P1 to P3, and the mesh M3 includes a positioning position P4 and a positioning position P5. In this case, the mesh M1 indicates that there are three users holding the user terminal 50, and the mesh M3 indicates that there are two users holding the user terminal 50.

  In this case, it may be determined that there are few people in the mesh M2 because there is no user who holds the user terminal 50 in the mesh M2. That is, even if some position information of the user terminal 50 is collected and estimated, in reality, only sparse information may be obtained. In this case, when extracting a rapid change in the population, it is usually determined that there is no accidental person by obtaining only sparse information in a mesh with people, and then determining that there are people in that mesh, The mesh will misjudge that the population has soared.

  In particular, when trying to estimate the population with respect to a fine mesh, there may be a mesh where there was a user accidentally as described above and a mesh that is not, and there may be a deficiency in the population data for each mesh.

  Further, from another viewpoint, since positioning by GPS generally includes a positioning error, in the above method, although the user of the user terminal 50 is positioned on the mesh M2, the mesh M1 or mesh It may be determined that it is located at M3.

  As shown in FIG. 3, based on the probability distribution of each position, the positioning position P1 contributes 80% to the mesh M1 and contributes 20% to the mesh M2. That is, the positioning position P1 belongs to the mesh M1 with an 80% probability and belongs to the mesh M2 with a 20% probability. Therefore, the number calculating unit 13 calculates the degree of the number of persons by interpolating and correcting data with respect to sparse position information (for example, positioning information by GPS). Thereby, the number calculation part 13 can interpolate the defect | deletion of the acquired positional information with the probability in space. Therefore, the population change determination device 10 can appropriately determine a mesh having a high degree of change.

  As described above, according to the first congestion degree calculation method, the position information based on certain position information is compared with the case where the degree of the number of persons of the mesh is calculated based only on the number of positioning results located in the predetermined mesh. Since the degree of the number of people in each mesh is calculated in consideration of the contribution ratio to the mesh other than the mesh including, the number of user terminals 50 existing in each mesh can be specified accurately. Therefore, the number calculation unit 13 calculates a contribution rate to each mesh for each position corresponding to the position information as a first congestion degree calculation method, and calculates the congestion degree using the contribution ratio.

Specifically, as shown in the following formula held in the number calculation unit 13, the number calculation unit 13 uses the latitude and longitude of the acquired position information of the same date and the same time zone as a contribution rate to each mesh. And is diffused in consideration of the human distribution based on the acquired position, and integrated for each mesh whose congestion degree is to be calculated. And the number calculation part 13 makes the value calculated by the result integrated by the following formula | equation the number (degree of a number of persons) of the positional information which belongs to each mesh.

Here, M (i, j) indicates the degree of the number of persons belonging to the mesh (i, j) (i is an index in the x direction, j is an index in the y direction), and (x, y) is Indicates the coordinates of the position. f (x, y) is a function that diffuses around the acquired position, and is defined in the number-of-people calculating unit 13. As a specific example, f (x, y) is a Gaussian function as shown in the following formula (2). (X ₀ , y ₀ ) indicates position information acquired from the user terminal 50. According to the Gaussian function, the contribution rate decreases with the position as the center, according to the distance from the position. Note that f (x, y) is not limited to a Gaussian function, and may be a function whose contribution rate decreases according to the distance from the position centered on the position acquired from the user terminal 40.

Σ (x, y) represents a parameter of the degree of diffusion of the contribution rate depending on the position. The number-of-people calculating unit 13 holds in advance the value of σ (x ₀ , y ₀ ) indicated by σ in the above equation (2). σ (x ₀ , y ₀ ) corresponds to, for example, the standard deviation of the contribution rate, and the greater the value, the greater the degree of diffusion of the contribution rate. Further, for example, the number-of-people calculating unit 13 determines σ (x ₀ , y ₀ ) according to the positioning accuracy of the position information of the position (x ₀ , y ₀ ). Specifically, the number of persons calculating unit 13 sets a higher value as the positioning accuracy is lower. This is because if the positioning accuracy is low (if the positioning error is large), the degree of diffusion of the contribution rate is considered to increase accordingly. For example, based on the map information, the number calculation unit 13 corresponds to a large value (for example, 100) and a position (x ₀ , y ₀ ) if the map object corresponding to the position (x ₀ , y ₀ ) is indoors. If the map object to be performed is outdoors, a small value (for example, 1) is held. This is because if the position (x ₀ , y ₀ ) is indoors, there is generally a high possibility of including a positioning error. The number calculation unit 13 may determine σ (x ₀ , y ₀ ) based on positioning accuracy information in the position information of the position (x ₀ , y ₀ ). It should be noted that the sum of the contribution rates of all positions is 1.

  Since the integration as in the above equation is equivalent to obtaining the volume, FIG. 4 shows only one dimension of x in the above equation. As shown in FIG. 4, at the center of the mesh M2 in the x-axis direction (left-right direction in the figure), that is, the point on the dotted line in FIG. 4 (the position indicated by the position information acquired by the position information acquisition unit 11). When the contribution ratio to each mesh is calculated based on the result of integration with respect to x, it is 0.7 for the mesh M2 and 0.15 for the mesh M1 and the mesh M3.

  The number-of-people calculating unit 13 calculates the contribution rate of each position to the mesh for each mesh M (i, j) based on the above formula (1), and the total value of the calculated contribution rates is used as the congestion degree. .

  The number-of-people calculation unit 13 obtains the congestion degree calculated based on the above formula (1), the date, the time zone, and mesh information corresponding to the congestion degree (for example, information on latitude and longitude constituting the mesh). The registered information is registered in the mesh congestion degree storage unit 14.

(Second congestion calculation method)
Next, a second congestion degree calculation method will be described. In the second congestion degree calculation method, on the same date and the same time zone, the number of position information belonging to the mesh of interest (mesh of congestion degree calculation target) determined by the population change determination device 10 and the surrounding meshes, and the map Multiplying by the weight determined for each mesh held in the information storage unit 12, and the sum of the results of the multiplication is used as the congestion degree of the target mesh.

  Examples of the mesh of interest include meshes corresponding to areas provided by the information service providing apparatus. The weight information determined for each target mesh is information in which the target mesh and its surrounding meshes are defined, and the respective mesh weights of the target mesh and the surrounding meshes are defined. The value of the weight information is such that the value decreases as the distance from the target mesh increases around the target mesh. For example, when the mesh (i, j) is used, the weight is calculated with respect to the center coordinates of the mesh (i, j) by the value of each mesh calculated using the above equation (1).

  An example of the specific calculation is shown in FIG. In FIG. 5, “weighting based on position” and “number of positioning results at corresponding positions in the mesh” are assigned to each mesh in order from the left side of the figure.

  The number calculation unit 13 sets the value of “weight based on position” to 8/16 of the maximum value for the mesh M5 that is the target mesh, and 1/16 to the surrounding mesh. As described above, this is a value based on the probability distribution of the position. When a person is present at a certain location, the smooth distribution is formed based on the distance from the center with the location as the center.

Note that the surrounding mesh having the weight is not limited to the mesh in contact with the target mesh. For example, the surrounding mesh having a weight is determined by the value of the diffusion degree parameter (σ (x ₀ , y ₀ ) in Expression (1)).

  Note that the weights of the target mesh and surrounding meshes may be set based on the positioning accuracy of the position information in the target mesh. Specifically, the number calculation unit 13 sets the value of σ described in the equations (1) and (2) based on the positioning accuracy of the position information in the target mesh. For example, the number of persons calculating unit 13 divides the positioning accuracy into a plurality of stages (levels), and totals the positioning results included in the position in the target mesh for each level of positioning accuracy.

For example, when there are three levels of positioning accuracy (assuming that 3 is the highest accuracy), σ shown in Equation (2) is determined as follows.
(A) (number of positioning results at level 3 of positioning accuracy) / (number of positioning results in the mesh)> 0.5, σ = 1
(B) When (the number of positioning results of positioning accuracy level 2) / (the number of positioning results in the mesh)> 0.5, σ = 10
(C) When (number of positioning results of positioning accuracy level 1) / (number of positioning results in mesh)> 0.5, σ = 100

  As described above, the value of σ is increased as the number of positioning results with low positioning accuracy increases. This is because when the positioning accuracy is low, the possibility of including a positioning error is generally high.

  Then, the number calculation unit 13 multiplies “the weight based on the position” and “the number of positioning results corresponding to the position in the mesh” for each mesh, and adds the multiplied value to the mesh of interest. It is calculated as the degree of congestion.

  In addition, the number calculation unit 13 may further use “information on coefficients based on whether or not a person can exist in the mesh” stored in the map information storage unit 12.

  As described above, an example in the case of calculation is shown in FIG. In FIG. 6, in order from the left side of the figure, “coefficient information based on whether or not a person can exist in the mesh”, “weighting based on position”, and “number of positioning results corresponding to the mesh” Assigned to each mesh. As shown in FIG. 6, since the meshes M3, M6, M8, and M9 are assigned 0 as "information on the coefficient based on whether or not a person can exist in the mesh" Indicates a location that is unlikely to exist. As a result, for each mesh, multiplying "coefficient information based on whether a person can exist in the mesh", "weighting based on position", and "number of positioning results corresponding to the mesh" The “number of positioning results corresponding to the mesh” corresponding to the meshes M3, M6, M8, and M9 becomes invalid. In addition, the distribution in which the “information on the coefficient based on whether or not a person can exist in the mesh” is assigned to each mesh is a distribution determined based on the geographical shape of each position.

  In this way, the number calculation unit 13 can calculate the degree of congestion by eliminating the positioning result included in the position where the person is not supposed to be present due to the positioning error or the like.

  The number-of-people calculation unit 13 obtains the congestion degree calculated by the second congestion degree calculation method, the date, the time zone, and the mesh information corresponding to the congestion degree (for example, information on the latitude and longitude constituting the mesh). The registered information is registered in the mesh congestion degree storage unit 14.

  According to the second congestion degree calculation method, it is not necessary to perform calculation for each point unlike the first congestion degree calculation method, and the amount of calculation can be reduced as compared with the first congestion degree calculation method.

  The mesh congestion degree storage unit 14 stores information on the mesh congestion degree. FIG. 7 shows an example of information stored in the mesh congestion degree storage unit 14.

  As illustrated in FIG. 7, the mesh congestion degree storage unit 14 includes a “date” indicating a date when the congestion degree is measured, a “time” indicating a measurement start time of the congestion degree, and a “lower left latitude” indicating the lower left latitude of the mesh. ”,“ Lower left longitude ”indicating the lower left longitude of the mesh,“ upper right latitude ”indicating the upper right latitude of the mesh,“ upper right longitude ”indicating the upper right longitude of the mesh, and“ congestion degree ”indicating the degree of congestion. Store information.

  The information stored in the mesh congestion degree storage unit 14 may be associated with the mesh ID instead of the latitude / longitude information on the lower left and upper right of the mesh as long as the time zone, the mesh, and the congestion degree are associated with each other.

The target period number acquisition unit 15 refers to the mesh congestion degree storage unit 14 and determines the degree of the number of persons in the target period for determining whether or not the population has changed in the target mesh based on the number of persons calculated by the number of persons calculation unit 13. Get the degree of the number of people in the target period. It is assumed that the target period number acquisition unit 15 holds the target period. Specifically, the target period number acquisition unit 15 uses the time T on the predetermined date of the target period (for example, the date acquired by the target period number acquisition unit 15) and the time width of the target period as the congestion degree of the target period. The sum of the degree of congestion of the time width W _tgt shown is acquired based on the information stored in the mesh degree of congestion storage unit 14. It is assumed that the target period number acquisition unit 15 holds the width of the time width W _tgt in advance. Here, the sum of the degree of congestion in the mesh m, time T, and time width W is expressed as freq (m, T, W). That is, the target period number acquisition unit 15 calculates freq (m, T, W _tgt ). When T = 21, and the time width W _tgt = 3, freq (m, 21, 3) = freq (m, 19, 1) + freq (m, 20, 1) + freq (m, 21, 1) expressed.

The past period number of people acquisition unit 16 refers to the mesh congestion degree storage unit 14, and based on the number of people calculated by the number of people calculation unit 13, the degree of the number of people in a period (normal period) including at least a past period from the target period of the target area. Get the degree of the number of people in the past period. Specifically, the past period number acquisition unit 16 acquires the sum of the congestion degrees of the time T and the time width W _all from the date of the target period as the congestion degree of the normal period. That is, the past period number acquisition unit 16 calculates freq (m, T, W _all ). W _all means the whole time zone (24 hours). M is a period that is somewhat long, such as one week, for example, and the population change determination device 10 holds the value of M. FIG. 8A shows the relationship between the time width W _tgt and the time width W _all . As shown in FIG. 8A, the time width W _all indicates the entire period of the most recent M days. In this case, the past period number acquisition unit 16 can eliminate the periodic influence of the day of the week. The time width W _all is not limited to the entire period of the most recent M days, but may be the same time zone as the target period on the most recent M days, as shown in FIG. 8B. In this case, the past period number acquisition unit 16 can detect an increase in the number of people due to an event at a spot where there is always a person such as an amusement park (comparison with the same time period in the past). Furthermore, when the device holds the time at which the event occurs, the population change determination device 10 may determine the target period and the normal period based on the time at which the event occurs.

  The degree-of-change determination unit 19 determines the degree of change in the number of people in the target period using the degree of the number of people acquired by the number-of-targets acquisition unit 15 and the number of people acquired by the number-of-people acquisition unit 16 in the past period. Information (judgment criteria) based on the reference number of people in the mesh (the number of people usually present in the target mesh) is further used.

  The following reference number acquisition unit 17 acquires the reference number of people, and the determination reference generation unit 16 generates the determination reference.

The reference number acquiring unit 17 acquires the degree of reference number of the target mesh (target area). Specifically, the reference number acquisition unit 17, for each date of the most recent N days to obtain the degree of congestion of the time width W _tgt minute from the time T, for each of the most recent N days time T, the time width W _tgt The sum of the congestion levels is calculated, and the result of each sum is acquired as reference number information. When acquiring the degree of congestion for the time width W _tgt from the time T, the reference number acquisition unit 17 goes back to the past by the time width W _tgt with respect to the time T. N is preferably a long period, for example, 30 days. The reference number acquisition unit 17 holds N in advance. Note that M <N. The time width W _tgt has a time width of 3 hours, for example. For example, when time T = 21 and time width W _tgt = 3, it is from 18:00 to 21:00.

  The determination criterion generation unit 18 sets a determination criterion based on the reference number information acquired by the reference number acquisition unit 17. Specifically, the determination criterion generation unit 18 calculates a median from the reference number information (a set of sums of congestion degrees) acquired by the reference number acquisition unit 17, and uses the calculated result as a determination criterion. The judgment criterion is a smoothing parameter.

  A statistical value other than the median (for example, an average value) may be used as the smoothing parameter. Further, the smoothing parameter may be calculated based on the statistical value. Specifically, the result of inputting the statistical value to the function is calculated as a smoothing parameter, or the mapping table (information on the correspondence between the degree of the standard number (statistical value) and the smoothing parameter) is searched using the statistical value as a key. The result is used as a smoothing parameter. The mapping table may be acquired in advance from an external device and stored in the determination criterion generation unit 18.

  The criterion generation unit 18 calculates the smoothing parameter in units of meshes. The reason for determining the smoothing parameter for each location is to change the value of the smoothing parameter depending on the regional characteristics of each location. For example, the value of the smoothing parameter in the urban mesh (mesh with a high degree of reference number) is set large, and the value of the smoothing parameter in the suburban mesh (mesh with a low degree of reference number) is set small.

As an example of calculating the smoothing parameter k by a function, a function represented by the following equation is applied.

In the above equation, u represents a constant, N represents the number of data of congestion degree in the past N days of a certain mesh m, time T, and time width W _tgt , and m_a represents a certain mesh m, time T, time width W The average value of the congestion degree of the past N days of _tgt is shown, m_s shows the standard deviation of the congestion degree of the past N days of a certain mesh m, time T, and time width W _tgt .

The degree-of-change determination unit 19 uses the degree of the number of people acquired by the target period number of people acquisition unit 15, the degree of the number of people acquired by the number of people acquisition unit 16 in the past period, and the determination criterion generated by the determination criterion generation unit 18. Judge the degree of change in the number of people in the period. In other words, smoothing the change degree determination unit 19, freq indicating the degree of period number (m, _{T, W all)} and freq indicating the degree of the past period number (the number of normal _{period) (m, T, W tgt} ) and Using the parameter k, a change score (specifically, a sudden rise score indicating the degree of rapid rise) is calculated as the degree of change in the number of people in the target period, and whether or not the number of people is highly changed based on the change score. to decide. Specifically, the degree-of-change determining unit 19 calculates a change score based on the following mathematical formula.

  Here, the case where the change score is calculated without using the smoothing parameter k is compared with the case where the change score is calculated using the smoothing parameter k. Since the change score is calculated by dividing the congestion degree between the two counting periods, when the change score is calculated without using the smoothing parameter k, a result in which the normal number of people in the target mesh is considered is not obtained. However, it is also important to consider the normal number of people in the target mesh when detecting a location that is actually attracting attention.

For example, consider a case where there is an increase from 10 to 100 in one mesh A and an increase from 100 to 1000 in another mesh B (the normal number of people in mesh A and B is 100) And). If you do not have a smoothing parameter:
The change score of mesh A is -log (100/110) = 0.041392685
The change score of mesh B is -log (1000/1100) = 0.041392685
It becomes.

  Considering the normal number of people in the target mesh, the mesh B changes more greatly, but the change score is the same.

Next, a smoothing parameter k (k = 100) is set. The change score of mesh A is -log (100 / (110 + 100)) = 0.322219295
The change score of mesh B is -log (1000 / (1100 + 100)) = 0.079181246
It becomes.

  Since the change score indicates that the value increases rapidly as the value decreases, it is determined that the mesh B is changing. By using the smoothing parameter k, a normal number of people can be considered.

  The change degree output unit 20 outputs the result determined by the change degree determination unit 19. As a specific example, when there is a mesh with a high degree of change, the change degree output unit 20 notifies the service information providing apparatus of a mesh with a high degree of change.

  FIG. 9 shows a hardware configuration of the population change determination device 10. As shown in FIG. 9, the population change determination device 10 includes one or a plurality of CPUs 101, a RAM 102 and a ROM 103 which are main storage devices, an input device 104 such as a keyboard and a mouse which are input devices, an output device 105 such as a display, and a network card. The computer system includes a communication module 106 that is a data transmission / reception device such as an auxiliary storage device 107 such as a semiconductor memory. Each function in FIG. 1 operates the input device 104, the output device 105, and the communication module 106 under the control of the CPU 101 by loading predetermined computer software on the hardware such as the CPU 101 and the RAM 102 shown in FIG. In addition, it is realized by reading and writing data in the RAM 102 and the auxiliary storage device 107.

  Next, processing of the first congestion degree calculation method will be described using the flowchart shown in FIG. The position information acquisition unit 11 acquires position information (step S1, position information acquisition step), and the number of persons calculation unit 13 calculates a contribution rate to each mesh in the position information (step S2, number of persons calculation step). If the calculation has not been completed for the position (step S3; No), the process moves to step S1, and if the calculation process has been completed for all positions (step S3; Yes), the number calculation unit 13 The degree of congestion in the mesh is calculated by adding the contribution ratios of the position information to the mesh (step S4, number of people calculation step).

  Next, processing of the second congestion degree calculation method will be described using the flowchart shown in FIG. The number-of-persons calculation unit 13 calculates the number of position information corresponding to the target mesh and surrounding meshes (step S11, number of persons calculation step), and based on the number of position information of each mesh and the weight of each mesh Is calculated (step S12, number of people calculation step).

  Next, processing for determining population change will be described with reference to the flowchart shown in FIG. As a premise, it is assumed that congestion degree information is stored in the mesh congestion degree storage unit 14. The reference number acquisition unit 17 calculates the sum of the congestion levels of the time T and the time width W in the most recent N days (step S21), and the determination criterion generation unit 18 sets the smoothing parameter k based on the sum of the congestion levels. Calculate (step S22), the past period number of people acquisition unit 16 calculates the congestion degree of the normal period (step S23, past period number of people acquisition step), and the target period number of people acquisition unit 15 calculates the degree of congestion of the target period. (Step S24, target period number acquisition step), the degree of change determination unit 19 calculates the degree of change using the smoothing parameter k, the normal period congestion degree, and the target period congestion degree (step S25, change degree determination step). ). Then, the change degree output unit 20 outputs information based on the determination result of the change degree (step S26, change degree output step).

(Function and effect)
Then, an effect is demonstrated. In the population change determination device 10, the position information acquisition unit 11 acquires position information indicating the position of the person associated with the time, and the number of persons calculation unit 13 is indicated by the position information acquired by the position information acquisition unit 11. The degree of congestion is calculated based on the distribution of people according to the distance from the position to be detected, and the target period number acquisition unit 15 calculates the number of persons in the target period of the target mesh (target area) based on the degree of congestion calculated by the number of persons calculation unit 13. The past period number acquisition unit 16 acquires the degree of the number of people in the past period from the target period of the target mesh based on the number of people calculated by the number calculation unit 13, and the change degree determination unit 19 The degree of change of the number of people in the target period is determined using the degree of the number of people acquired by the number of people acquisition unit 18 and the degree of the number of people acquired by the number of people acquisition unit 16 in the past period. And outputs the information based on the result of focus determination unit 19 has determined.

  According to the above-described population change determination device 10, the degree of mesh congestion is calculated using the distribution of people. Therefore, even if it is difficult to measure the degree of the number of target meshes from only position information, The degree of the number of meshes (congestion degree) can be estimated, and the congestion degree in the target period can be accurately determined.

  The number-of-people calculating unit 13 calculates the contribution rate to each mesh for each position corresponding to the position information based on the distribution, and calculates the congestion degree for each time zone and each mesh using the contribution rate. Good. In this case, since the population change determination device 10 calculates the degree of congestion in consideration of the spread of the position of the position information, it can accurately determine the degree of change in the number of people.

  The number calculation unit 13 weights the position information number of positions included in the predetermined mesh and the position information numbers of positions included in the meshes around the predetermined mesh based on the distribution, and the time zone and the mesh Calculate the degree of congestion for each. In this case, since the population change determination device 10 calculates the degree of congestion in consideration of the spread of the position of the position information, it can accurately determine the degree of change in the number of people.

  The distribution may be determined based on the accuracy of each piece of position information. In this case, since the population change determination device 10 calculates the degree of the number of people in consideration of the positional information shift due to the positioning information positioning error, the number of people can be accurately calculated.

  The distribution may be determined based on the geographical shape of each position. In this case, the population change determination device 10 uses the distribution based on the geographical shape to calculate the degree of congestion, and therefore excludes position information included in an area considered to be free of people from the target for calculating the degree of congestion. can do. Therefore, the population change determination device 10 can accurately calculate the degree of the number of people.

  The reference number acquisition unit 17 acquires the degree of the reference number of the target mesh, and the determination reference generation unit 18 generates the smoothing parameter k based on the reference number of degree acquired by the reference number acquisition unit 17 and determines the degree of change. The unit 19 determines the degree of change in the number of persons in the target period using the target period number information, the past period number information, and the smoothing parameter k. In this case, since the population change determination device 10 determines the degree of population change using the determination criteria for the position of the determination target, it is compared with a case where it is determined whether or not the population has simply changed based on the degree of change. Thus, the degree of change in the population can be accurately determined.

(Other embodiments)
In the above-described embodiment, the case of calculating the degree of congestion for the mesh has been described. However, the present invention is not limited to this, and the degree of congestion for a predetermined area may be calculated.

  In the above-described embodiment, the case where the degree-of-change determining unit 19 determines the degree of change in the number of persons in the target period using the target period number of persons information, the past period number of persons information, and the smoothing parameter k has been described. The degree of change in the number of people may be determined based on the past period number information. In this case, the degree-of-change determining unit 19 determines the degree of change using a value calculated by an expression excluding the smoothing parameter k in Expression (4).

The degree-of-change determination unit 19 calculates the smoothing parameter k for each location and determines the degree of change, using the past period number information in which the time width W _all is the same time zone as the target period on the most recent M days. The degree of change may be determined by a plurality of determination methods such as a method of determining the degree of change, and the degree of change may be finally determined based on the result. For example, the degree-of-change determination unit 19 may determine that it is rapidly increasing if there is any method that has determined that it has rapidly increased, or may determine by a majority vote.

  The degree-of-change determination unit 19 calculates the change score using Equation (4), but uses the test used in statistics to test the absolute value of the degree of congestion for each mesh, It may be determined whether the change in the degree of congestion is within a normal range or is rapidly increasing (or rapidly decreasing). In this case, the degree-of-change determination unit 19 can determine how much the level of change has changed by examining the change rate of the degree of congestion. Thus, the degree-of-change determination unit 19 may determine the degree of change from the average value and variance of the degree of congestion.

  In the above-described embodiment, the smoothing parameter k is calculated based on the congestion degree stored in the mesh congestion degree storage unit 14, but predetermined statistical information (statistically calculated congestion degree information). ). Note that the population change determination device 10 may hold the statistical information, or the population change determination device 10 may acquire the statistical information held in the external device. As described above, data for setting a criterion may be acquired from an external device.

  DESCRIPTION OF SYMBOLS 10 ... Population change judgment apparatus, 11 ... Position information acquisition part, 12 ... Map information storage part, 13 ... Number of persons calculation part, 14 ... Mesh congestion degree calculation part, 15 ... Reference | standard number of persons acquisition part, 16 ... Judgment reference generation part, 17 ... Target period number acquisition unit, 18... Past period number acquisition unit, 20.

Claims (7)

Position information acquisition means for acquiring position information indicating the position of the person associated with the time;
Number of people calculating means for calculating the degree of the number of people for each time zone and area based on the distribution of people according to the distance from the position indicated by the position information acquired by the position information acquiring means;
A target period number acquisition means for acquiring the degree of the number of persons in the target period of the area of interest based on the number of persons calculated by the number of persons calculation means;
Past number of persons acquisition means for acquiring the degree of the number of persons in the past period from the target period of the area of interest based on the number of persons calculated by the number of persons calculation means;
Using the degree of the number of people acquired by the target period number of people acquisition means and the degree of the number of people acquired by the past period number of people acquisition means, a change degree judgment means for judging the degree of change of the number of people in the target period;
A population change determination device comprising: a change degree output unit that outputs information based on a result of determination by the change degree determination unit.   The number calculation means calculates a contribution rate to each area for each position corresponding to the position information based on the distribution, and calculates the degree of the number of people for each time zone and area using the contribution rate. The population change determination device according to claim 1.   The number of persons calculating means weights the position information number of positions included in a predetermined area and the position information number of positions included in an area around the predetermined area based on the distribution, respectively, The population change determination device according to claim 1, wherein the degree of the number of people for each area is calculated.   The population change determination device according to any one of claims 1 to 3, wherein the distribution is determined based on accuracy of each position information.   The population change determination device according to any one of claims 1 to 4, wherein the distribution is determined based on a geographical shape of each position. A determination criterion setting means for setting a determination criterion based on the degree of the reference number of persons in the area of interest;
The change degree determination means uses the degree of the number of persons acquired by the target period number of persons acquisition means and the degree of the number of persons acquired by the past period number of persons acquisition means based on the determination criteria, and changes in the number of persons in the target period. The population change determination device according to any one of claims 1 to 5, which determines a degree. A population change determination method by a population change determination device, a location information acquisition step for acquiring location information indicating a location of a person associated with a time,
Based on the distribution of people according to the distance from the position indicated by the position information acquired in the position information acquisition step, the number of people calculation step to calculate the degree of the number of people for each time zone and area,
A target period number of people acquisition step for acquiring the degree of the number of people in the target period of the area of interest based on the number of people calculated in the number of people calculation step;
A past period number of people acquisition step for acquiring the degree of the number of people in the past period from the target period of the area of interest based on the number of people calculated in the number of people calculation step;
Using the degree of the number of people acquired in the target period number of people acquisition step and the degree of the number of people acquired in the previous period number of people acquisition step, a change degree determination step of determining the degree of change of the number of people in the target period;
A change degree output step for outputting information based on the result determined in the change degree determination step;
How to determine population change including