Information processing device, information processing method, and information processing program

Abstract

This method estimates user interests by suppressing the influence of fluctuations in activity volume due to local traffic conditions. [Solution] The information processing device acquires activity data indicating the amount of activity the user engages in, acquires boarding and alighting point information indicating the boarding and alighting points of the means of transportation used by the user, identifies an actual boarding and alighting point area where boarding and alighting points are considered to substantially exist within a predetermined area where the user engages in activity based on the positional relationship of the boarding and alighting points, calculates the boarding and alighting point density, which is the effective density of boarding and alighting points in the predetermined area, corrects the amount of activity indicated by the activity data according to the boarding and alighting point density, performs profiling to estimate the user's interests based on the corrected amount of activity, and outputs the profiling results.

Description

Technical Field

[0001] The present disclosure relates to user profiling technology.

Background Art

[0002] Technology development for estimating user interests using behavior data generated from user behaviors is progressing. For example, Patent Document 1 discloses a technology that, upon receiving a content distribution request, estimates a context corresponding to the user's location and time zone, determines a model that corresponds to the estimated context and is for estimating the user's interest in content, and determines the content to be distributed to the user using the determined model.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the above conventional technology does not take into account that the user's behavior volume fluctuates according to the traffic environment of the area where the user acts, so further improvement is required to accurately estimate the user's interests.

[0005] An object of the present disclosure is to provide a technology for estimating user interests by suppressing the influence of the fluctuation component of behavior volume that corresponds to the local traffic environment.

Means for Solving the Problems

[0006] An information processing device in one aspect of the present disclosure is an information processing device including a processor, the processor acquires activity data indicating the amount of activity of a user, acquires boarding and alighting point information indicating boarding and alighting points of the means of transport used by the user, identifies an actual boarding and alighting point area from the positional relationship of the boarding and alighting points in a predetermined area where the user acts, calculates a boarding and alighting point density which is the actual density of the boarding and alighting points in the predetermined area based on the predetermined area and the actual boarding and alighting point area, corrects the amount of activity indicated by the activity data according to the boarding and alighting point density, performs profiling to estimate the user's interests based on the corrected amount of activity, and outputs the results of the profiling. [Effects of the Invention]

[0007] According to this disclosure, it is possible to estimate user interests while suppressing the influence of fluctuations in activity volume depending on the local traffic environment. [Brief explanation of the drawing]

[0008] [Figure 1] This is a block diagram showing an example of the configuration of an information processing system in an embodiment of the present disclosure. [Figure 2] This is a flowchart showing an example of processing in the information processing device according to the embodiments of this disclosure. [Figure 3] This diagram shows an overview of the process for identifying the actual boarding and alighting point area. [Figure 4] This diagram shows an overview of the process for identifying the actual boarding and alighting point area. [Figure 5] This flowchart shows the details of the process for identifying the actual boarding and alighting points. [Figure 6] This graph shows the correlation coefficient between the number of steps taken per day and the mode of transportation. [Figure 7] This is a flowchart of Modification 1 in the embodiment. [Figure 8] This is an explanatory diagram of the process in the modified example 2. [Figure 9] This is an explanatory diagram of the process in the modified example 2. [Modes for carrying out the invention]

[0009] (Knowledge forming the basis of this disclosure) Whether a user prefers indoor or outdoor activities is fundamental information that reflects their characteristics. Unlike strenuous sports, "movement," such as walking or running, is a common activity for all users in their daily lives. Therefore, information generated by a user's movement, such as the number of steps, speed, and acceleration during walking or running, can be used as useful input when profiling to analyze user characteristics.

[0010] In particular, the "number of steps" resulting from users' movement behaviors such as "walking" and "running" is useful for comparing the behavioral trends of multiple users and is helpful in estimating user interests. Furthermore, the "number of steps" is easily influenced by human thought. For example, when viewed over time, a high number of steps can be estimated to indicate that a particular user has become health-conscious and is interested in health. Moreover, a high number of steps, when combined with the user's location information, can be used to determine if the user is prone to going out. Furthermore, by considering the "number of steps" along with location information or event information of the place of visit, it is possible to estimate the level of interest a user has in others.

[0011] Generally, in areas with extensive bus networks, users tend to use buses more frequently, resulting in a lower number of steps taken. This is because bus stops are densely located, allowing users to reach them without having to walk very far from their homes.

[0012] On the other hand, in areas with well-developed railway networks, users tend to walk more frequently. This is because railway stations are more sparsely located than bus stops, increasing the number of steps taken from home to the railway station.

[0013] Therefore, if the user's "step count" is treated uniformly without considering the regional traffic environment and the user's interest is estimated from the user's "step count", the user's interest cannot be estimated accurately. For example, it is not appropriate to conclude that a user in an area with well-developed bus stops where the step count tends to decrease has low interest in health just because the user's step count is small. On the other hand, it is not appropriate to conclude that a user in an area with a well-developed railway network where the step count tends to increase has high interest in health just because the user's step count is large.

[0014] Here, in order to eliminate the influence of changes in the step count according to the regional traffic environment, a method of increasing or decreasing the user's step count according to the density of boarding and alighting points such as bus stops or railway stations is also conceivable.

[0015] However, it has been found that there are regions where boarding and alighting points in an area can be regarded as having substantial boarding and alighting points due to their positional arrangement. Therefore, when calculating the density of boarding and alighting points, such regions need to be considered.

[0016] The present disclosure has been made based on the above findings.

[0017] (1) An information processing apparatus according to an aspect of the present disclosure is an information processing apparatus including a processor, wherein the processor acquires activity amount data indicating a user's activity amount, acquires boarding and alighting point information indicating boarding and alighting points of a transportation means used by the user, identifies, from the positional relationship of the boarding and alighting points, an actual boarding and alighting point region where the boarding and alighting points can be regarded as substantially existing in a predetermined region where the user acts, calculates a boarding and alighting point density that is a substantial density of the boarding and alighting points in the predetermined region based on the predetermined region and the actual boarding and alighting point region, corrects the activity amount indicated by the activity amount data according to the boarding and alighting point density, executes profiling for estimating the user's interest based on a corrected activity amount that is the corrected activity amount, and outputs a result of the profiling.

[0018] According to this configuration, the user's activity amount is corrected according to the density of boarding and alighting points of transportation means, and the user's interest is estimated based on the corrected activity amount which is the corrected activity amount. Thereby, the influence of the fluctuation component of the user's activity amount corresponding to the local traffic environment can be suppressed, and the user's interest can be estimated. Furthermore, in this configuration, an actual boarding and alighting point area that can be regarded as substantially existing is identified from the positional relationship of the boarding and alighting points, and the boarding and alighting point density which is the substantial density of the boarding and alighting points is calculated from the area and the actual boarding and alighting point area. Therefore, the user's interest can be accurately estimated using the substantial boarding and alighting point density.

[0019] (2) In the information processing apparatus according to (1) described above, the activity amount may be calculated by statistically processing activity data indicating the activity of the user.

[0020] According to this configuration, the user's activity amount can be accurately calculated based on the activity data.

[0021] (3) In the information processing apparatus according to (1) or (2) described above, the boarding and alighting points may include bus stops.

[0022] According to this configuration, the corrected activity amount is calculated such that the fluctuation component of the activity amount corresponding to the bus stop is removed from the activity amount, so that the user's interest in the area with bus stops can be accurately estimated.

[0023] (4) In the information processing apparatus according to any one of (1) to (3) described above, correcting the activity amount may include acquiring terrain information indicating a distribution of slopes in a predetermined area, calculating the slope degree of the predetermined area based on the terrain information, and correcting the boarding and alighting point density such that the smaller the slope degree is, the smaller the boarding and alighting point density becomes.

[0024] The variable component of activity volume depends not only on the local transportation environment but also on the slope of the local terrain. For example, users in areas with a gentle slope tend to use local transportation less frequently, resulting in an increase in the variable component of activity volume that depends on the transportation environment. In this configuration, the density of boarding and alighting points is corrected so that the density decreases as the slope decreases. Therefore, it is possible to estimate user interests while taking into account the variable component of activity volume that depends on the terrain.

[0025] (5) In the information processing device described in any of (1) to (4) above, identifying the actual boarding / alighting point area may include, under the constraint that the boarding / alighting point area including the boarding / alighting point cannot be entered, moving a probe having a shape corresponding to the boarding / alighting point area in a search area including at least the predetermined area to identify an area that the probe cannot reach, and identifying the identified area and the boarding / alighting point area as the actual boarding / alighting point area.

[0026] This configuration allows for the precise identification of actual boarding / alighting point areas where boarding / alighting points are considered to exist, using probes.

[0027] (6) In the information processing device described in (5) above, the boarding / alighting point area is rectangular, and the probe may be a rectangle larger in size than the boarding / alighting point area.

[0028] With this configuration, a rectangular probe larger than the boarding / alighting area is moved across the search area, making it possible to accurately find areas that the probe cannot reach.

[0029] (7) In the information processing device described in (5) or (6) above, the boarding / alighting point area may be circular, and the probe may be circular.

[0030] This configuration allows for the precise detection of areas that cannot be reached by the circular probe.

[0031] (8) In the information processing apparatus described in any of (5) to (7) above, the search area may include the predetermined area and a padding area that is set to surround the predetermined area and has a shape that allows the probe to move.

[0032] With this configuration, a padding area is set around the predetermined area, so the probe can reliably explore the entire area. For example, even if the predetermined area is divided in the middle by boarding and alighting points, the probe can bypass the boarding and alighting points through the padding area and explore the entire area.

[0033] (9) In the information processing device described in any of (1) to (8) above, correcting the amount of activity may include increasing the amount of activity as the density of boarding and alighting points increases, in the case of a user using a means of transport in which the amount of activity tends to decrease as the number of boarding and alighting points increases.

[0034] According to this configuration, for users of transportation methods that tend to decrease their activity level as the number of boarding and alighting points increases, their activity level increases as the density of boarding and alighting points increases. This allows for an inflated activity level for users of such transportation methods, enabling an accurate estimation of their interests.

[0035] (10) In the information processing device described in any of (1) to (9) above, the correction of the amount of activity may be made by reducing the amount of activity as the density of the boarding and alighting points increases, in the case of a user using a means of transport in which the amount of activity tends to increase as the number of boarding and alighting points increases.

[0036] According to this configuration, for users of transportation methods that tend to increase their activity level as the number of boarding and alighting points increases, their activity level decreases as the density of boarding and alighting points increases. This allows for a reduction in the activity level of users of such transportation methods, enabling a more accurate estimation of their interests.

[0037] (11) In the information processing device described in any of (1) to (10) above, the amount of activity may be the amount of walking.

[0038] This configuration allows for the calculation of the user's actual walking volume, after removing the component of step count fluctuations due to traffic conditions, and enables an accurate estimation of the user's interests.

[0039] This disclosure can also be implemented as an information processing program that causes a computer to execute each of the characteristic configurations included in such an information update method, or as an information processing system that operates using this information processing program. Furthermore, it goes without saying that such a computer program can be distributed via a computer-readable, non-temporary recording medium such as a CD-ROM, or via a communication network such as the Internet.

[0040] (12) An information processing method in another aspect of the present disclosure is an information processing method performed by a computer, which includes: acquiring activity data indicating the amount of activity of a user; acquiring boarding / alighting point information indicating boarding / alighting points of the means of transport used by the user; identifying an actual boarding / alighting point area from the positional relationship of the boarding / alighting points in a predetermined area in which the boarding / alighting points are considered to substantially exist; calculating a boarding / alighting point density, which is the substantial density of the boarding / alighting points in the predetermined area, based on the predetermined area and the actual boarding / alighting point area; correcting the amount of activity indicated by the activity data according to the boarding / alighting point density; performing profiling to estimate the user's interests based on the corrected amount of activity; and outputting the results of the profiling.

[0041] (13) An information processing program in yet another aspect of the present disclosure causes a computer to perform the following actions: acquire activity data indicating the amount of activity of a user; acquire boarding and alighting point information indicating the boarding and alighting points of the means of transport used by the user; identify an actual boarding and alighting point area in which the boarding and alighting points can be considered to substantially exist in a predetermined area in which the user is active, based on the positional relationship of the boarding and alighting points; calculate the boarding and alighting point density, which is the substantial density of the boarding and alighting points in the predetermined area, based on the predetermined area and the actual boarding and alighting point area; correct the amount of activity indicated by the activity data according to the boarding and alighting point density; perform profiling to estimate the user's interests based on the corrected amount of activity, which is the corrected amount of activity; and output the results of the profiling.

[0042] The embodiments described below are all specific examples of this disclosure. The numerical values, shapes, components, steps, and order of steps shown in the following embodiments are examples only and are not intended to limit this disclosure. Furthermore, among the components in the following embodiments, those not described in the independent claim representing the highest-level concept will be described as optional components. In addition, the contents of each embodiment can be combined.

[0043] (Embodiment) Figure 1 is a block diagram showing an example of the configuration of an information processing system 100 in an embodiment of the present disclosure. The information processing system 100 includes an information processing device 1 and a terminal 4. The information processing device 1 is composed of computers such as a cloud server and an edge server. The information processing device 1 and the terminal 4 are connected via a network. The network may be a wide-area communication network such as the Internet, or a local area network. The information processing device 1 may be implemented in the terminal 4.

[0044] Terminal 4 is a device that acquires user information that is the subject of information processing. Terminal 4 may be, for example, a smartphone, a personal computer, a wearable device, or a physical measurement device. Wearable devices may be, for example, smart glasses or a smartwatch. A physical measurement device may be, for example, a pedometer. However, the form of Terminal 4 is not limited to these examples.

[0045] Terminal 4 may be one or more terminals. Terminal 4 inputs the acquired user log (hereinafter referred to as user log) to the information processing device 1 at a predetermined timing. In this embodiment, the user log is behavioral data that indicates the user's actions. Behavioral data includes, for example, the user's step count, the number of times the user has operated an application software (hereinafter referred to as an app), the number of times the user has operated a home appliance, and the user's location information. The user's location information is acquired, for example, by a GPS sensor.

[0046] The app, for example, encourages users to take walks to promote their health. Specifically, the app accepts user submissions of recommended places to visit and posts them on a bulletin board accessible to all users. The app also accepts comments from other users on these places and posts those comments on the bulletin board, associating them with the places. This allows the app to motivate users to visit the places and take walks. Furthermore, users can get to know other users through comments and build a personal network.

[0047] User location information is used to determine the user's range of activity. Examples of household appliances include washing machines, televisions, vacuum cleaners, and refrigerators.

[0048] Behavioral data may include behavioral data recorded at regular intervals (for example, every hour or every 10 minutes) and behavioral data recorded irregularly. Behavioral data recorded at regular intervals may include, for example, the user's location information obtained by a GPS sensor. On the other hand, irregularly recorded behavioral data may include app operations. App operations may include, for example, switching screens within the app, posting, commenting, or using stamps to indicate a "like". The information processing device 1 uses at least one of the behavioral data recorded at regular intervals and the behavioral data recorded irregularly.

[0049] In this embodiment, behavioral data can include data representing the amount of exercise, such as the number of steps, the number of times the user goes out, the distance traveled, and the distance walked. Terminal 4 calculates the number of steps by, for example, dividing the distance traveled, calculated from location information, by the average stride length (e.g., 70 cm). The walking distance is not simply the distance walked, but the distance the user walked. Terminal 4 may calculate the walking distance separately for each type of activity, such as walking and moving to a destination. The walking distance for each type of activity can be determined from operations on the app, etc. For example, Terminal 4 calculates the distance traveled during the time when the user operates the app to display information about a spot such as a park on Terminal 4 and views that information as the walking distance. Behavioral data may also include information generated by the user's movement, such as walking or running. Information generated by the user's movement includes not only the number of steps, but also speed and acceleration.

[0050] Terminal 4 can, for example, count the number of times the user leaves the house by comparing location information with the location information of the user's home, and calculate the counted number as the number of times the user leaves the house. Terminal 4 can, for example, calculate the distance the user travels outside the home area as the distance traveled. The home area is a circular area with a predetermined radius centered on the user's location. Terminal 4 may calculate the walking distance as the user's travel distance and exclude the distance traveled by means of transportation such as trains and automobiles. Terminal 4 can, for example, calculate the user's speed from location information, and if the speed is below a threshold, it can determine that the user is walking. In the following explanation, the behavioral data will represent the number of steps.

[0051] The information processing device 1 includes a processor 2 and a memory 3. The processor 2 is, for example, a central processing unit (CPU). The processor 2 includes an acquisition unit 21, a preprocessing unit 22, a density calculation unit 23, a correction unit 24, a profiling unit 25, and an output unit 26. The acquisition unit 21 to the output unit 26 are realized by the CPU executing an information processing program stored in the memory 3. However, this is just an example, and the acquisition unit 21 to the output unit 26 may be composed of dedicated hardware circuits. The acquisition unit 21 to the output unit 26 may be distributed across multiple computers.

[0052] Memory 3 consists of a non-volatile storage device such as a solid-state drive and includes a log storage unit 31.

[0053] The acquisition unit 21 acquires behavioral data indicating the amount of user activity. For example, the acquisition unit 21 can acquire behavioral data from the terminal 4 via a communication interface (not shown in the diagram). The acquisition unit 21 stores the acquired behavioral data in the log storage unit 31. The behavioral data includes a timestamp indicating the date and time the user took action. In this disclosure, the behavioral data is time-series data in which values ​​indicating the user's actions are arranged sequentially over time. The time unit in the behavioral data is not specified. For example, it may be in units of one day or one hour.

[0054] The acquisition unit 21 acquires pick-up / drop-off point information indicating the pick-up / drop-off locations for the means of transportation used by the user. The means of transportation include buses, trains, cars, motorcycles, bicycles, kick scooters, taxis, vertical take-off and landing aircraft, etc. The means of transportation may also be cars, motorcycles, bicycles, and kick scooters in a sharing service. The pick-up / drop-off locations are, for example, bus stops, train stations, taxi stands, and vertical take-off and landing aircraft. The pick-up / drop-off locations may also be the locations where cars, bicycles, and kick scooters are installed in a sharing service. If the means of transportation is a car, motorcycle, bicycle, or kick scooter owned by the user, the pick-up / drop-off locations may also be the user's home and storage locations such as parking lots.

[0055] The preprocessing unit 22 reads behavioral data from the log storage unit 31 and performs statistical processing on the read behavioral data so that it can be used to estimate the user's interests, thereby calculating behavioral data that indicates the user's activity level. The activity level is the total number of steps taken by the user each day. Hereafter, any one user among multiple users will be referred to as user i.

[0056] For example, the log storage unit 31 stores activity data showing the number of steps taken each hour. In this case, the preprocessing unit 22 calculates the number of steps taken by user i each day by adding up the number of steps taken each hour over a predetermined period. The preprocessing unit 22 then calculates the total number of steps taken by user i each day as the activity amount, and calculates the activity amount data by arranging the calculated activity amounts in chronological order. The activity amount data is chronological data such as, for example, the activity amount (number of steps) on February 5th was 5000 steps, the activity amount (number of steps) on February 6th was 6200 steps, and the activity amount (number of steps) on February 7th was 4100 steps.

[0057] Activity level is not limited to the total number of steps taken by user i each day. For example, activity level may be the average number of steps taken by user i each day, the median number of steps taken by user i each day, or the mode number of steps taken by user i each day. Activity level may also be the number of times one goes out each day, the distance traveled each day, and the distance walked each day. Steps and walking distance are examples of walking volume.

[0058] The specified period is the period for which the activity volume data is calculated. For example, the specified period is two years, from January 1, 2022 to January 1, 2024. A longer specified period is preferable from the standpoint of noise reduction. However, this is just an example, and the activity volume data may consist of data from a single activity.

[0059] The density calculation unit 23 identifies an actual boarding / alighting point area where boarding / alighting points can be considered to substantially exist within a predetermined area where the user is active, based on the positional relationship of boarding / alighting points indicated by the boarding / alighting point information.

[0060] The designated area is the area of ​​focus in the user behavior analysis, and is the area where user i is expected to normally behave. For example, the designated area could be the city including the user's place of residence, or the area inside a square with sides of 1 km centered on the place of residence.

[0061] For example, the density calculation unit 23, under the constraint that the boarding / alighting point area including the boarding / alighting point cannot be entered, moves a probe having a shape corresponding to the boarding / alighting point area in a search area that includes at least a predetermined area where the user acts, identifies areas that the probe cannot reach, and identifies the identified areas and the boarding / alighting point area as the actual boarding / alighting point area. The shape corresponding to the boarding / alighting point area is, for example, a similar shape.

[0062] The predetermined area is demarcated by multiple grids arranged in a matrix. The grids are rectangular. The probe has a size equal to the matrix arrangement of the multiple grids. The area containing the boarding / alighting points is the grid where the boarding / alighting points are located. Therefore, the probe is composed of a rectangle larger than the boarding / alighting point area. As a result, when the probe is moved through the search area under the aforementioned constraints, areas that the probe cannot reach will occur. The search area is set to surround the predetermined area and includes a padding area that has a shape that the probe can move through, and the predetermined area itself.

[0063] The density calculation unit 23 calculates the boarding / alighting point density, which is the effective density of boarding / alighting points in the predetermined region, based on the predetermined region and the actual boarding / alighting point region. The boarding / alighting point density is, for example, the area ratio of the actual boarding / alighting point region within the predetermined region.

[0064] The correction unit 24 corrects the amount of activity included in the activity data according to the density of boarding and alighting points. For example, in the case of a user using a mode of transportation (hereinafter referred to as the first mode of transportation) in which the amount of activity tends to decrease as the number of boarding and alighting points increases, the correction unit 24 increases the amount of activity as the density of boarding and alighting points increases.

[0065] The correction unit 24 reduces the amount of activity as the density of boarding and alighting points increases for users of a mode of transportation (hereinafter referred to as the second mode of transportation) that tends to increase in activity as the number of boarding and alighting points increases.

[0066] The correction unit 24 may acquire topographic information showing the distribution of ground slope in a predetermined area. In this case, the correction unit 24 calculates the degree of slope in the predetermined area from the topographic information and corrects the density of boarding and alighting points so that the smaller the degree of slope, the smaller the density of boarding and alighting points. Details of this process will be explained in Modification 1 below.

[0067] The profiling unit 25 performs profiling to estimate the user's interests based on the corrected behavioral amount, which is an adjusted amount of behavior. Based on the trend of the corrected behavioral amount, the profiling unit 25 evaluates the user's interest in health. For example, if the corrected behavioral amount is on an increasing trend, the profiling unit 25 evaluates that the user's interest in health is increasing. For example, if the corrected behavioral amount is on a decreasing trend, the profiling unit 25 evaluates that the user's interest in health is decreasing.

[0068] The output unit 26 outputs the profiling results from the profiling unit 25. For example, the output unit 26 may output the profiling results to terminal 4. The output unit 26 may also output the profiling results to the terminal of the administrator providing this service.

[0069] Figure 2 is a flowchart showing an example of processing by the information processing device 1 in an embodiment of this disclosure. The following processing is performed for each user.

[0070] (Step S1) The acquisition unit 21 acquires behavioral data from the terminal 4 and stores the acquired behavioral data in the log storage unit 31.

[0071] (Step S2) The preprocessing unit 22 reads behavioral data from the log storage unit 31 and calculates activity volume data by statistically processing the read behavioral data. Activity volume is, for example, the average number of steps taken each day. Activity volume data is time-series data of activity volume.

[0072] (Step S3) The acquisition unit 21 acquires the category of the mode of transport used by user i. As mentioned above, the categories of modes of transport include bus, train, etc. The acquisition unit 21 only needs to acquire the category of the mode of transport entered by user i from terminal 4. If user i's mode of transport category is registered in memory 3 in advance, the acquisition unit 21 only needs to acquire user i's mode of transport category from memory 3. A user can input multiple categories of modes of transport. In this case, the flowchart in Figure 2 is executed for each of the multiple categories of modes of transport.

[0073] (Step S4) The acquisition unit 21 acquires area information indicating a predetermined area in which user i will be acting. Furthermore, the acquisition unit 21 acquires boarding / alighting point information indicating the boarding and alighting points of the means of transportation used by the user. The area information is a two-dimensional coordinate space on which the predetermined area is mapped. The boarding / alighting point information is two-dimensional coordinate data indicating the position of the boarding / alighting point within the predetermined area.

[0074] (Step S5) The density calculation unit 23 performs a process to identify the actual boarding / alighting point area using a probe to identify the actual boarding / alighting points from a predetermined area. For the sake of simplicity, the transportation category is assumed to be buses, and the boarding / alighting points are assumed to be bus stops. The predetermined area is assumed to be a plane that ignores the curvature of the ground resulting from the Earth being a sphere. That is, the predetermined area is assumed to be an xy-plane with two axes, the x-coordinate and the y-coordinate. The way the origin of the predetermined area and the x-axis and y-axis are chosen does not affect the processing results described later, so the x-axis is assumed to be latitude and the y-axis is assumed to be longitude. In the following explanation, the x-axis and y-axis each have a length of 100m per unit. The grid is, for example, a square with a side length of 1.

[0075] Figures 3 and 4 show an overview of the process for identifying the actual boarding and alighting point area. Multiple grids 503 are set within the predetermined area 500. Bus stop grid 502 is a grid 503 that contains at least one bus stop. Grids 503 other than bus stop grid 502 in the predetermined area 500 are grids 503 that do not contain bus stops.

[0076] Probe 401 is similar to the probe used in the SASA (solvent accessible surface area) approximation calculation method. SASA is the surface area of ​​molecules that can be accessed by a solvent. The SASA approximation calculation method uses solvent molecules (e.g., water molecules) as probes with a certain radius, moves these probes across the molecular surface, and calculates the area of ​​the contact portion. This embodiment uses the SASA approximation calculation method approach to identify the actual boarding and alighting point area.

[0077] Probe 401 is a square with sides of length 2. The search using probe 401 is performed under the constraint that probe 401 cannot enter the bus stop grid 502, and probe 401 is moved thoroughly within a predetermined area 500. After probe 401 has moved thoroughly and the search is complete, each grid 503 is divided into "grids that the probe can reach" and "grids that the probe cannot reach". As a result, grids 503 that the probe cannot reach (hereinafter referred to as unreachable grids 504) are identified, as shown in Figure 4.

[0078] Figure 5 is a flowchart detailing the process for identifying the actual boarding and alighting point area. This flowchart will be explained below with reference to Figures 3 and 4.

[0079] (Step S41) The density calculation unit 23 performs padding, setting a padding region 501 to surround the predetermined region 500. The padding region 501 has a shape that ensures the probe 401 can always complete one circuit around the predetermined region 500. State F1 in Figure 3 shows the predetermined region 500 after padding has been performed. In this example, the padding region 501 has a width of two grids in the x-axis direction and a width of two grids in the y-axis direction. That is, the padding region 501 has a shape in which the width in the x-axis direction is greater than the width of the probe 401 in the x-axis direction and the width in the y-axis direction is greater than the width of the probe 401 in the y-axis direction. The region composed of the predetermined region 500 and the padding region 501 is the search region 600.

[0080] (Step S42) The density calculation unit 23 places the probe 401 at the initial position 505. The initial position 505 is any position in the padding region 501. State F2 in Figure 3 shows the probe 401 placed at the initial position 505. The initial position 505 is, for example, the position where both the x-axis and y-axis values ​​are minimized, that is, the position where the leftmost vertex of the probe 401 is located at the leftmost vertex of the padding region 501.

[0081] (Step S43) The density calculation unit 23 sets the search step Si to 1. State F3 in Figure 3 is when the search step Si is 1. State F4 in Figure 4 is when the search step Si is 6. State F5 in Figure 4 is when the search step Si is 9.

[0082] (Step S44) The density calculation unit 23 identifies possible directions in which the probe 401 can be moved one grid unit from inside to outside the boundary 506 of the explored range in the exploration step Si-1, and moves the probe 401 in the identified possible directions. In state F3, the boundary 506 is the top and right edges of the probe 401. In states F4 and F5, the boundary 506 is shown by a thick line.

[0083] For example, in states F4 and F5, the probe 401 is slid one grid at a time along the boundary 506 inside the boundary 506, and at each sliding position, the possible directions in which the probe 401 can move from the inside to the outside of the boundary 506 are explored. Then, the probe 401 is moved in the direction of movement identified at each sliding position.

[0084] In state F3, the probe 401 is located at its initial position 505, so it can move upward, to the right, and diagonally upward to the right. In state F4, for example, if the probe 401 is located at slide position 507, it can move upward and diagonally upward to the right. For example, in state F4, if the probe 401 is located at slide position 508, due to the influence of the surrounding bus stop grid 502, it can only move upward.

[0085] (Step S45) The density calculation unit 23 identifies reachable grids. In state F3, there are no bus stop grids 502 around the initial position 505, so the five grids 503 adjacent to the initial position 505 become reachable grids. Therefore, at the stage where the search step Si is 1, the nine grids 503 on the far left, which are colored in dark gray, become reachable grids.

[0086] In states F4 and F5, the grids 503 located outside the boundary 506 and adjacent to the boundary 506 that are in the direction of movement become reachable grids. In this way, as the search step Si increases, the reachable grids are gradually filled in by the probe 401 so that the number of reachable grids increases towards the upper right.

[0087] (Step S46) The density calculation unit 23 stores the reachable grids in memory 3. Each grid 503 is assigned a grid index to distinguish it from other grids 503. Therefore, the density calculation unit 23 only needs to store the grid index of the reachable grids in memory 3.

[0088] (Step S47) The density calculation unit 23 determines whether the number of possible directions of movement is 0 or not. State F6 in Figure 4 shows the search area 600 after the search is completed. In state F6, the boundary 506 (not shown) is located on the top and right edges of the search area 600, and the probe 401 cannot be moved any further. Therefore, the number of possible directions of movement is 0. If the number of possible directions of movement is 0 (YES in step S47), the process proceeds to step S49. On the other hand, if the number of possible directions of movement is not 0 (NO in step S47), the process proceeds to step S48.

[0089] (Step S48) The density calculation unit 23 adds 1 to the search step Si. When step S48 is completed, the process returns to step S44. As a result, the boundary 506 is shifted outward by one grid unit.

[0090] (Step S49) The density calculation unit 23 excludes the padding area 501 and the reachable area from the search area 600, and identifies the remaining grid as an unreachable grid 504 that the probe 401 cannot reach.

[0091] (Step S50) The density calculation unit 23 identifies the area consisting of the unreachable grid 504 and the bus stop grid 502 as the actual boarding and alighting point area.

[0092] The process for identifying boarding and alighting points described above performs padding, allowing the search area to be thoroughly explored with a finite number of iterations. Furthermore, because a padding area 501 is set, even if bus stops are located in such a way that the predetermined area 500 is divided into two, the probe 401 can move through the area where the initial position 505 does not exist among the two divided areas.

[0093] (Step S6) The density calculation unit 23 calculates the boarding / alighting point density using equation (1). The boarding / alighting point density is the area ratio of the actual boarding / alighting point area within a predetermined area 500.

[0094] boarding / alighting point density = actual boarding / alighting point area / predetermined area (1) Furthermore, if there are grids in the predetermined area 500 that correspond to lakes or forests that user i cannot access, the area containing these grids may be excluded from the predetermined area 500.

[0095] (Step S7) The correction unit 24 corrects the activity volume included in the activity volume data according to the density of boarding and alighting points and calculates the corrected activity volume. There are two types of corrections: correction performed on a per-user basis and correction performed on different time periods for the same user. The former is called inter-user correction, and the latter is called time period correction. Inter-user correction and time period correction will be explained in order below.

[0096] In user-to-user correction, the correction unit 24 corrects user i's activity data Fi (time-series data of the number of steps per day) using the mode of transport category C and the boarding / alighting point density Di.

[0097] Figure 6 is a graph showing the correlation coefficient between the number of steps per day and the mode of transportation. As shown in Figure 6, some modes of transportation have a positive correlation, increasing the number of steps, while others have a negative correlation, decreasing the number of steps. A symbol K(C) is defined to determine whether the number of steps increases or decreases depending on the mode of transportation. In detail, railways, which have sparse boarding and alighting points, and bicycles, which are powered by human effort, have a positive correlation with the number of steps. On the other hand, taxis, automobiles, and buses have a negative correlation with the number of steps. Shared services have many boarding and alighting points, so they have a negative correlation with the number of steps.

[0098] Based on this, we will now explain the relationship between the transportation category C and code K entered in step S2.

[0099] For modes of transportation that have a positive correlation with the number of steps, the number of steps tends to increase, so the correction unit 24 corrects the number of steps in the direction of decrease. On the other hand, for modes of transportation that have a negative correlation with the number of steps, the number of steps tends to decrease, so the correction unit 24 corrects the number of steps in the direction of increase. Therefore, the sign K(C) has the opposite sign to the positive or negative correlation shown in Figure 6. Category C for which the sign K(C) is positive includes taxis, automobiles, motorcycles, buses, vertical take-off and landing aircraft, shared service automobiles, shared service motorcycles, shared service bicycles, and shared service kick scooters. Category C for which the sign K(C) is negative includes railways and bicycles. The mode of transportation for which the sign K(C) is positive is the first mode of transportation described above. The mode of transportation for which the sign K(C) is negative is the second mode of transportation described above.

[0100] If the corrected activity data Fi for user i is the corrected activity data F'i, and the correction amount is α, then the corrected activity data F'i is expressed by equation (2). The corrected activity data F'i is the time series data of the corrected activity.

[0101] F'i = Fi(1 + α) (2) The corrected activity data F'i may be expressed as a power of (1+α). The correction amount α is expressed by equation (3) using the ratio of the boarding / alighting point density Di at the maximum value Dmax and the sign K(C).

[0102] α = β·K(C)·Di / Dmax (3) β is an adjustment factor, for example, 0.5. Di is the boarding / alighting point density for user i. Dmax is the maximum boarding / alighting point density for all users.

[0103] For example, consider user i, who lives near many bus stops. Assume the boarding / alighting point density Di is 80% of the maximum value Dmax, and that the average number of steps (activity) per day is 8000 steps. In this case, the corrected activity data F'i would be F'i = 8000(1 + 0.5 · 0.8) = 11200 steps.

[0104] In this way, the step count of users residing in areas where the number of steps tends to decrease due to the traffic environment is inflated. This removes the influence of the variable component depending on the mode of transportation from the step count, allowing for a fair comparison of step counts among users. The correction unit 24 only needs to perform this correction on all activity amounts (number of steps per day) included in the activity amount data Fi.

[0105] (Step S8) The profiling unit 25 performs profiling using the corrected behavior data F'i. A specific example of profiling is described above.

[0106] (Step S9) The output unit 26 outputs the profiling results to the terminal 4 or the like.

[0107] Thus, according to this embodiment, the amount of user activity is corrected according to the density of boarding and alighting points for means of transport, and the user's interests are estimated based on the corrected amount of activity. This makes it possible to estimate the user's interests while suppressing the influence of fluctuations in the amount of user activity according to the local traffic environment. Furthermore, in this embodiment, an actual boarding and alighting point area that can be considered to exist substantially based on the positional relationship of boarding and alighting points is identified, and the boarding and alighting point density, which is the actual density of boarding and alighting points, is calculated from the region and the actual boarding and alighting point area. Therefore, the user's interests can be accurately estimated using the actual boarding and alighting point density.

[0108] (Variation 1) Next, we will describe a modified example 1 of the flowchart in Figure 2. Figure 7 is a flowchart of modified example 1 in the embodiment. In Figure 7, the same steps as in Figure 2 are given the same step numbers, and their explanations are omitted.

[0109] The flowchart in Modification 1 has an additional step S30 compared to Figure 2. Step S30 is a process that corrects the boarding / alighting point density Di using terrain information.

[0110] (Step S31) The correction unit 24 acquires terrain information. The terrain information includes the slope angle of each grid 503 in a predetermined area 500. The terrain information is stored in memory 3 in advance. Therefore, the correction unit 24 only needs to acquire the terrain information from memory 3.

[0111] The correction unit 24 calculates the slope of a predetermined area from the terrain information. The slope is, for example, the average slope or the median of the slope angles in the predetermined area 500. The correction unit 24 can calculate the average slope by averaging the slope angles in each grid 503. The terrain information may also be the elevation difference between adjacent grids 503 in each grid 503. In this case, the correction unit 24 may calculate the slope as the average or median of the elevation differences in each grid 503.

[0112] (Step S32) The correction unit 24 corrects the boarding / alighting point density Di using the gradient.

[0113] In flat areas, bicycle use tends to increase. Consequently, bus usage decreases despite the presence of many bus stops. As a result, the variable component related to the mode of transport in flat areas tends to increase compared to areas with higher gradients. To correct this increasing trend, the correction unit 24 corrects the boarding / alighting point density Di using the gradient. The corrected boarding / alighting point density D'i is expressed by equation (4).

[0114] D'i = Di(1-γ) (4) The density correction factor γ is expressed by equation (5).

[0115] γ = δ(1 - Gi / Gmax) (5) Gi is the gradient. Gmax is the maximum gradient for all users. δ is the correction adjustment coefficient, for example, 0.5.

[0116] As the gradient Gi increases, the density correction amount γ decreases, and consequently, the boarding / alighting point density D'i decreases. In the modified example 1, the correction unit 24 calculates the correction amount α by substituting the boarding / alighting point density D'i into Di in equation (3). As the boarding / alighting point density D'i decreases, the correction amount α decreases. Therefore, the correction amount α in flat areas is smaller than the correction amount α in areas with high gradients. Thus, it is possible to take into account the decrease in bus usage frequency in flat areas and reduce the increase in the number of steps.

[0117] For example, consider user i who lives in a flat area. Assume the gradient Gi is 20% of the maximum value Gmax. Assume the pre-correction pick-up / drop-off point density Di is 0.4. In this case, the pick-up / drop-off point density D'i = 0.4·(1-0.5·(1-0.2)) = 0.24. Thus, users living in flat areas will use bicycles more often and buses less often. Modification 1 takes this into account, lowering the pick-up / drop-off point density D'i and reducing the increase in step count. This allows users to compare each other using step counts that are free from topographical factors. Note that if step S30 is not performed, the density correction amount γ will be 0.

[0118] Next, we will explain period adjustment. Period adjustment adjusts the activity volume data Fi for two or more periods within the same user in order to compare them. Period adjustment can be achieved by replacing the index of user i in the inter-user adjustment with the index of multiple periods for the same user.

[0119] In other words, processor 2 extracts two or more time periods from the activity data Fi calculated in step S2. Then, processor 2 applies the processing shown in Figure 2 or Figure 7 to each of the extracted time periods. As a result, corrected activity amounts are obtained for each time period, with the variable components corresponding to the mode of transport removed.

[0120] For example, when comparing the past and present, processor 2 divides the activity volume data Fi for a predetermined period into an older period and a newer period at the center of the time series. Then, processor 2 obtains corrected activity volume data F'i from each of the older and newer periods, with the variable component corresponding to the mode of transport removed. Information processing device 1 then presents the corrected activity volume data F'i for both periods to the user. This allows the user's actual activity volume to be obtained, for example, even if the user has moved. The period from which the activity volume data Fi is extracted may be, for example, monthly, semi-annual, or annual. The length of the extraction period can be specified by the administrator or the user.

[0121] (Modification 2) Modification 2 uses a circular probe instead of a rectangular probe. Figures 8 and 9 are explanatory diagrams of the process in Modification 2. Bus stops indicated by stars are placed in the predetermined area 700. The bus stop area 701 is a circle centered on the bus stop. The radius t of the bus stop area 701 is, for example, 0.5. The density calculation unit 23 uses a circular probe 601 to identify the actual boarding and alighting point area. The radius s of the probe 601 is, for example, 1.

[0122] In the modified example 2, the flowchart in Figure 5 can also be applied to the process of identifying the actual boarding and alighting points. However, as shown in Figure 9, if the distance d between the probe 601 and the bus stop area 701 is less than the sum of the radius s of the probe 601 and the radius t of the bus stop area 701 (= s + t), the density calculation unit 23 determines that the probe 601 overlaps with the bus stop area 701. Therefore, the density calculation unit 23 moves the probe 601 in the predetermined area 700 under the condition that the sum is greater than or equal to the distance d. The density calculation unit 23 allows the probe 601 to extend beyond the predetermined area 700 if the center of the probe 601 is within the predetermined area 700.

[0123] The density calculation unit 23 identifies the areas in the predetermined region 700 that the probe 601 could not reach as unreachable regions 702. In Figure 8, the gray areas are the areas that the probe 601 could reach, so in the predetermined region 700, the areas other than the bus stop region 701 and the gray areas are identified as unreachable regions 702. The density calculation unit 23 identifies the area consisting of the bus stop region 701 and the unreachable regions 702 as the actual boarding / alighting point region. Subsequently, the density calculation unit 23 uses this actual boarding / alighting point region to calculate the boarding / alighting point density Di or boarding / alighting point density D'i.

[0124] The subsequent processing is the same as in the embodiment, so the explanation will be omitted. When using probe 601, the actual boarding and alighting point area can be identified more accurately, and the user's interest can be accurately estimated.

[0125] (Examples of application) The following describes examples of how this disclosure applies.

[0126] The app is a health promotion app for a local community. Users can refer to this app to check their step count. The processing of this disclosure applies to such apps. Users within the app belong to a local community (e.g., residents of a town). Users use the app over a long period of time. User A, who is interested in their health, is a resident of area X, which has a well-developed bus network and many bus stops. User A frequently uses the bus in their daily life. For example, they can take the bus from a bus stop near their home to their final destination, the city hall. Therefore, User A hardly walks at all. On the other hand, User B is a resident of area Y, which has a well-developed railway network. User B's residence is far from the nearest station. Therefore, User B walks the distance from their home to the station and the distance from the station closer to their destination to their final destination. In this case, if User A's step count is lower than User B's, it is difficult to distinguish whether User A simply dislikes walking or if it is due to the influence of their mode of transportation. Therefore, it is not possible to accurately estimate User A's interest in health from the increase or decrease in step count. Furthermore, when verifying the effectiveness of a campaign encouraging increased step counts, the step counts of user A and user B include variable components depending on their mode of transportation, making it impossible to accurately compare their step counts. As a result, it is impossible to accurately estimate their health-related interests.

[0127] On the other hand, in the user profiling method described in this disclosure, the component of variation due to the mode of transportation is removed from the step count, allowing for a highly accurate estimation of the user's interests. This enables the display of step counts with the variation due to the mode of transportation removed to the user. As a result, the user can understand their own walking effort. Therefore, even if a user moves to a city with significant differences in transportation infrastructure, they will still be motivated to use this app.

[0128] A specific example of the processing in this example will be explained step by step, referring to the configuration of this embodiment. A resident in a certain local community uses a smartphone as terminal 4. The smartphone has an application installed that is intended to promote the health of the local community. The acquisition unit 21 acquires the user's step count using this application and saves it to the log storage unit 31. The log storage unit 31 may be a database built on the cloud.

[0129] The information processing device 1 performs a process to estimate the user's interests at 11 PM every day. This process is called as a batch process at 11 PM. The preprocessing unit 22 calculates the activity data Fi by calculating the total number of steps per day from the activity data obtained from the log storage unit 31. The correction unit 24 corrects the number of steps based on the activity data Fi calculated by the preprocessing unit 22 so that the variable component corresponding to the mode of transportation is removed. The profiling unit 25 estimates the user's interests based on the corrected activity data F'i. The output unit 26 notifies the administrator of the estimation results. This allows the administrator to check whether the user's level of interest in health over the past month has increased compared to before. If the level of interest in health has decreased, the administrator notifies the user of a notification or event to increase their interest in health. [Industrial applicability]

[0130] This disclosure is useful for accurately estimating user interests. [Explanation of Symbols]

[0131] 1: Information Processing Device 21: Acquisition section 22: Pre-processing section 23: Density calculation section 24: Correction section 25: Profiling Department 26: Output section 31: Log storage section 100: Information Processing Systems 401: Probe 500: Predetermined area 501: Padding area

Claims

1. An information processing device including a processor, The aforementioned processor, We obtain activity data that shows the amount of user activity, The user obtains boarding and alighting point information indicating the boarding and alighting points of the means of transportation used by the user. Based on the positional relationship of the boarding and alighting points, the actual boarding and alighting point area is identified in which the boarding and alighting points can be considered to substantially exist within the predetermined area where the user acts. Based on the predetermined area and the actual boarding / alighting point area, the boarding / alighting point density, which is the actual density of boarding / alighting points in the predetermined area, is calculated. The activity volume shown in the activity volume data is corrected according to the density of boarding and alighting points. Based on the corrected behavioral amount, which is the corrected behavioral amount, profiling is performed to estimate the user's interests. Output the results of the aforementioned profiling. Information processing device.

2. The aforementioned activity volume is calculated by statistically processing the activity data that represents the user's actions. The information processing apparatus according to claim 1.

3. The aforementioned boarding and alighting points include bus stops, The information processing apparatus according to claim 1 or 2.

4. Correcting the aforementioned amount of activity means To obtain topographic information showing the distribution of slopes in the predetermined region, This includes calculating the gradient of the predetermined area based on the topographic information, and correcting the density of boarding / alighting points so that the density of boarding / alighting points decreases as the gradient decreases. The information processing apparatus according to claim 1 or 2.

5. Identifying the actual boarding / alighting point area includes, under the constraint that the boarding / alighting point area including the boarding / alighting point cannot be entered, moving a probe having a shape corresponding to the boarding / alighting point area in a search area including at least the predetermined area to identify areas that the probe cannot reach, and identifying the identified areas and the boarding / alighting point area as the actual boarding / alighting point area. The information processing apparatus according to claim 1 or 2.

6. The aforementioned boarding / alighting area is rectangular. The probe is a rectangle that is larger in size than the boarding / alighting area. The information processing apparatus according to claim 5.

7. The boarding / alighting area is circular. The aforementioned probe is circular. The information processing apparatus according to claim 5.

8. The search area includes a padding area that is set to surround the predetermined area and has a shape that allows the probe to move, and the predetermined area. The information processing apparatus according to claim 5.

9. Correcting the aforementioned amount of activity means In the case of a user using a mode of transportation in which the amount of activity tends to decrease as the number of boarding and alighting points increases, the method includes increasing the amount of activity as the density of boarding and alighting points increases. The information processing apparatus according to claim 1 or 2.

10. Correcting the aforementioned amount of activity means In the case of a user of a mode of transportation in which the amount of activity tends to increase as the number of boarding and alighting points increases, the amount of activity decreases as the density of boarding and alighting points increases. The information processing apparatus according to claim 1 or 2.

11. The aforementioned activity volume is the amount of walking. The information processing apparatus according to claim 1 or 2.

12. A method of information processing performed by a computer, To obtain activity data that shows the amount of user activity, The process involves obtaining boarding and alighting point information that indicates the boarding and alighting points of the means of transportation used by the user, Based on the positional relationship of the aforementioned boarding and alighting points, the actual boarding and alighting point area can be identified in which the aforementioned boarding and alighting points can be considered to substantially exist within the predetermined area in which the user acts. Based on the predetermined region and the actual boarding / alighting point region, the boarding / alighting point density, which is the actual density of boarding / alighting points in the predetermined region, Correcting the amount of activity shown in the activity data according to the density of boarding and alighting points, Performing profiling to estimate the user's interests based on the corrected behavior quantity, which is the corrected behavior quantity, This includes outputting the results of the profiling, Information processing methods.

13. On the computer, To obtain activity data that shows the amount of user activity, The process involves obtaining boarding and alighting point information that indicates the boarding and alighting points of the means of transportation used by the user, Based on the positional relationship of the aforementioned boarding and alighting points, the actual boarding and alighting point area can be identified in which the aforementioned boarding and alighting points can be considered to substantially exist within the predetermined area in which the user acts. Based on the predetermined region and the actual boarding / alighting point region, the boarding / alighting point density, which is the actual density of boarding / alighting points in the predetermined region, Correcting the amount of activity shown in the activity data according to the density of boarding and alighting points, Performing profiling to estimate the user's interests based on the corrected behavior quantity, which is the corrected behavior quantity, To output the results of the aforementioned profiling and to execute the following: Information processing program.

Images