BEHAVIOR PREDICTION METHOD, PROGRAM, AND BEHAVIOR PREDICTION SYSTEM

Abstract

The purpose of the present disclosure is to improve the accuracy of predicting whether a user visits a facility. A query unit (11) transmits question information to a portable terminal (3) carried by a user who is a prediction subject, the question information for querying an intent to visit a facility during a first period which is after a prediction time point. A response reception unit (12) receives response information from the portable terminal (3), the response information indicating an intent to visit the facility. An assessment unit (13) assesses the reliability of the response information during a second period, which is prior to the prediction time point, on the basis of a plurality of response information received from the portable terminal (3), and a track record for visits during the second period obtained from a positioning result from a positioning system (2). The positioning system (2) measures the position of the portable terminal (3) within the facility. A prediction unit (14) predicts whether a user subject to prediction visits the facility during the first period on the basis of: the response information received from the portable terminal (3) at the prediction time point; and the results of comparing high values and low values of the reliability and a threshold.

Description

[Technical field]

[0001] The present disclosure relates to a behavior prediction method, a program, and a behavior prediction system. More particularly, the present disclosure relates to a behavior prediction method, a program, and a behavior prediction system that predict a user's behavior. [Background technology]

[0002] Patent Document 1 discloses an information providing device that provides information to a terminal device used by a user. The information providing device includes an estimation unit and a determination unit. The estimation unit estimates a state related to the movement of the user based on location information indicating the position of the terminal device detected by a detection device. The determination unit determines the content of the processing to be provided to the user based on the state estimated by the estimation unit. For example, when the user starts moving, the estimation unit estimates the intention of the user to move, and determines the content of the information to be provided to the user and the timing of providing the information.

[0003] Since the above-mentioned information providing device cannot predict the movement of a user before the user starts moving, there has been a demand for the device to be able to predict the movement (attendance) of a user in advance with high accuracy. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2019-053435 A Summary of the Invention

[0005] An object of the present disclosure is to provide a behavior prediction method, program, and behavior prediction system that can improve the accuracy of predicting whether or not a user will visit a facility.

[0006] A behavior prediction method according to one embodiment of the present disclosure includes: A method for predicting behavior executed by one or more processors, comprising:The method predicts whether or not a user to be predicted among a plurality of users will visit a facility where the plurality of users visit and perform a predetermined action at a prediction time that arrives regularly or irregularly. The method includes a questioning step, a receiving step, an evaluation step, and a prediction step. In the questioning step, question information inquiring about the intention to visit the facility in a first period after the prediction time is transmitted to a mobile device carried by the user to be predicted. In the receiving step, answer information indicating the intention to visit the facility is received from the mobile device. In the evaluation step, the accuracy of the answer information is evaluated by comparing the plurality of pieces of answer information received from the mobile device with the visit record to the facility obtained from the positioning result of a positioning system in a second period before the prediction time. The positioning system measures the position of the mobile device within the facility. In the prediction step, the accuracy of the answer information of the user to be predicted in the first period is predicted based on the answer information received from the mobile device at the prediction time and a result of comparing whether the accuracy is high or low with a threshold value.

[0007] A program according to one embodiment of the present disclosure is a program for causing one or more processors to execute the behavior prediction method.

[0008] A behavior prediction system according to an embodiment of the present disclosure is a behavior prediction system that predicts whether or not a user to be predicted among a plurality of users will visit a facility where the plurality of users visit and perform a predetermined action at a prediction time that arrives regularly or irregularly. The behavior prediction system includes a questioning unit, a response receiving unit, an evaluation unit, and a prediction unit. The questioning unit transmits question information inquiring about an intention to visit the facility in a first period after the prediction time to a mobile device carried by the user to be predicted. The response receiving unit receives response information indicating an intention to visit the facility from the mobile device. The evaluation unit evaluates the accuracy of the response information based on a plurality of pieces of response information received from the mobile device in a second period before the prediction time and a record of visits to the facility in the second period obtained from a positioning result of a positioning system. The positioning system measures the position of the mobile device within the facility. The prediction unit predicts whether or not the user to be predicted will attend during the first period based on the response information received from the mobile terminal at the time of prediction and the result of comparing the accuracy with a threshold value. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a schematic block diagram of a behavior prediction system according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is a sequence diagram illustrating the operation of the behavior prediction system. [Diagram 3] FIG. 3 is a sequence diagram illustrating the operation of the behavior prediction system. [Figure 4] FIG. 4 is a diagram showing an example of question information that the behavior prediction system transmits to a mobile terminal. [Diagram 5] FIG. 5 is a diagram showing an example of a display screen that the behavior prediction system displays on a mobile terminal. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] (Embodiment) (1) Overview The behavior prediction system 1 (see FIG. 1) of this embodiment is applied to a facility F1 where users visit and perform a predetermined action, and predicts whether or not the user will visit the facility F1.

[0011] The behavior prediction system 1 of this embodiment predicts whether or not a user to be predicted will visit a facility F1 at a prediction time that arrives regularly or irregularly. The facility F1 is a facility where multiple users visit and perform a predetermined action.

[0012] The behavior prediction system 1 includes a questioning unit 11, an answer receiving unit 12, an evaluation unit 13, and a prediction unit 14.

[0013] The question unit 11 transmits question information inquiring about the intention to visit the facility F1 in a first period after the prediction time point to the mobile terminal 3 carried by the user who is the prediction target.

[0014] The response receiving unit 12 receives, from the mobile terminal 3, response information indicating an intention to visit the facility F1.

[0015] The evaluation unit 13 evaluates the accuracy of the answer information by comparing the multiple pieces of answer information received from the mobile terminal 3 during a second period before the prediction time with the attendance record to the facility F1 obtained from the positioning result of the positioning system 2. The positioning system 2 measures the position of the mobile terminal 3 within the facility F1.

[0016] The prediction unit 14 predicts whether or not the user to be predicted will attend the event during the first period based on the answer information received from the mobile terminal 3 at the time of prediction and the result of comparing the accuracy with a threshold value. In the following description, the user to be predicted may be simply referred to as the user.

[0017] Here, if the facility F1 is an office building used by a company or a government office, the user is an employee working at the company or an official working at a government office, and the "prescribed action" is to work in the facility F1. The prescribed action is not limited to working, but may be an action performed at the facility F1 used for a prescribed purpose, and may be an action that can be performed online or offline at home. For example, if the facility F1 is a membership-based fitness club, the prescribed action performed at the facility F1 is training. Also, if the facility F1 is a cram school or a culture center, the prescribed action performed at the facility F1 is a learning action. In the following, an example will be described in which the facility F1 is an office building used by a company, and the user who uses the facility F1 is an employee working at the company. Therefore, in the following, the user's visit to the facility F1 may be referred to as "going to work," and the user's not visit to the facility F1 may be referred to as "staying at home." Also, the user's visit to the facility F1 and working may be referred to as "working at the office," and the user's work at home may be referred to as "working at home."

[0018] Conventionally, the working style of company employees was mainly office work, where employees come to the facility F1 to work, but due to the influence of the global epidemic of infectious diseases such as the COVID-19 pandemic, telework, where users work at their homes, is increasing. The management department of the company has a desire to reduce the number of office work employees who come to the facility F1 to work in order to suppress the spread of infection within the facility F1. For example, if the ratio (attendance rate) of the number of people scheduled to work in the office tomorrow to all users exceeds a predetermined target value, the management department has a desire to suppress the attendance rate to below the target value by sending a notification to users who are scheduled to work in the office to encourage them to change to teleworking. In addition, users also have a desire to check the attendance rate of the facility F1 tomorrow in advance to decide whether to work in the office or telework tomorrow. The behavior prediction system 1 of this embodiment predicts whether users will visit the facility F1 in a first period (for example, the next day) after the prediction time point, and when the facility F1 is used by multiple users, predicts the overall attendance rate based on the prediction results of the multiple users.

[0019] The questioning unit 11 of the behavior prediction system 1 inquires as to "intention to visit" in the first period whether the user will visit the facility F1, will not visit the facility F1, or is undecided. In the following, an example will be described in which the facility F1 is a company, so the questioning unit 11 inquires as to "intention to visit" in the first period whether the user will come to work, stay at home, or is undecided. "Answer information" is information that the user answers to the question information using the mobile terminal 3, and is information that answers whether the user intends to come to work, stay at home, or is undecided in the first period.

[0020] The response receiving unit 12 receives the response information from the mobile terminal 3, but the user does not necessarily behave in the same way as the content of the response information. A user who answered that he / she works in the office may suddenly change to telecommuting, or conversely, a user who answered that he / she works at home may change to work in the office. In this embodiment, the evaluation unit 13 evaluates the accuracy indicating the accuracy of the response information based on a plurality of pieces of response information answered by the user to be predicted in the second period in the past and the attendance record of the user to be predicted in the second period obtained from the positioning result of the positioning system 2. Then, the prediction unit 14 predicts whether or not the user to be predicted will visit based on the result of comparing the accuracy obtained by the evaluation unit 13 with the threshold value and the response information, so that it is possible to provide a behavior prediction system 1 that can improve the prediction accuracy of the attendance compared to a case where prediction is based only on the response information.

[0021] (2)Details The behavior prediction system 1 of this embodiment will be described in detail below with reference to FIGS.

[0022] (2.1) Structure As described above, the behavior prediction system 1 predicts whether or not a prediction target user will visit the facility F1. In this embodiment, the facility F1 is, for example, an office building used by a company, the users of the facility F1 are employees of the company, and a case where multiple users (employees) use the facility F1 will be described as an example. Therefore, in the following, visiting the facility F1 may be referred to as coming to work, and not visiting the facility F1 may be referred to as being at home. The behavior prediction system 1 predicts whether or not the prediction target user will come to work during a first period, with each of the multiple employees who use the facility F1 as a prediction target user. Then, the behavior prediction system 1 obtains the attendance rate of a group consisting of multiple users (employees) based on the prediction results of the multiple users who use the facility F1. The user carries a mobile terminal 3 such as a smartphone or a tablet computer. The mobile terminal 3 has a function of transmitting a beacon signal, and a positioning system 2 that measures the position of the mobile terminal 3 is provided in the facility F1.

[0023] The behavior prediction system 1, the positioning system 2, and the mobile terminal 3 will be described below.

[0024] (2.1.1) Positioning System Positioning system 2 is located in facility F1, where GPS (Global Positioning System) signals cannot reach. This is a positioning system called LPS (Local Positioning system) that measures the position of the mobile terminal 3 within the

[0025] The positioning system 2 includes a beacon transmitting unit 35 included in the mobile terminal 3, a plurality of beacon receiving units 20 provided in the facility F1, and a positioning unit 15.

[0026] The mobile terminal 3 has a GPS positioning function that measures its position using, for example, GPS radio waves, and when it detects that it has entered the facility F1 based on positioning information using the GPS, the beacon transmitting unit 35 transmits a beacon signal by radio signal periodically or irregularly. The beacon signal includes an individual terminal ID (e.g., MAC address, etc.) assigned to the mobile terminal 3. The mobile terminal 3 may be configured to transmit a beacon signal when it enters the facility F1 and receives activation signals transmitted at predetermined time intervals from the multiple beacon receiving units 20. In addition, when the mobile terminal 3 receives a signal notifying the entry into the facility F1 from, for example, an entrance / exit management system that manages the entrance / exit to the facility F1, the beacon transmitting unit 35 may periodically wirelessly transmit a beacon signal by radio signal.

[0027] The multiple beacon receivers 20 are installed at intervals on the ceiling or the like of a room used by users in the facility F1. The multiple beacon receivers 20 each receive a beacon signal transmitted from a beacon transmitter 35. The multiple beacon receivers 20 each have a function for measuring the received signal strength indication (RSSI) of the received beacon signal. The beacon receiving unit 20 has a function of outputting the received signal strength of the received beacon signal and the identification information of the beacon receiving unit 20 to the positioning unit 15.

[0028] The positioning unit 15 measures the position of the mobile terminal 3 in the facility F1 based on the multiple beacon signals received by the multiple beacon receivers 20. The positioning unit 15 registers information on the three-dimensional coordinates of the installation location and identification information for each of the multiple beacon receivers 20 in association with each other. When the positioning unit 15 receives the terminal ID of the mobile terminal 3 that is the sender of the beacon signal, the received signal strength of the beacon signal, and the identification information of the beacon receiver 20 from each of the multiple beacon receivers 20, the positioning unit 15 calculates the distance from the multiple beacon receivers 20 to the mobile terminal 3 based on the received signal strength. Then, the positioning unit 15 calculates the three-dimensional position coordinates of the mobile terminal 3 using the three-dimensional coordinates of the installation locations of the multiple beacon receivers 20 and the multiple distances from the multiple beacon receivers 20 to the mobile terminal 3 using the principle of triangulation. When the positioning unit 15 detects the location of the mobile terminal 3, it stores the positioning result, including the location information indicating the location of the mobile terminal 3 and the time information at the time of the location detection, in the first storage unit DB1.

[0029] (2.1.2) Behavioral Prediction System The behavior prediction system 1 includes a processing unit 10, a communication unit 17, a receiving unit 19, a first storage unit DB1, and a second storage unit DB2.

[0030] The communication unit 17 has a communication function for wirelessly communicating with the mobile terminal 3 carried by the user. The communication method between the communication unit 17 and the mobile terminal 3 is, for example, Bluetooth (registered trademark). Note that the communication method between the communication unit 17 and the mobile terminal 3 is not limited to Bluetooth, and may be WiFi (registered trademark) or a specific low-power radio in the 920 MHz band. Furthermore, the communication unit 17 is not limited to one that directly communicates with the mobile terminal 3, and may be one that communicates via a repeater or the like.

[0031] The reception unit 19 receives an input operation from a manager of the behavior prediction system 1 (for example, a person in charge of the management department of a company that uses the facility F1). The reception unit 19 has input devices such as a keyboard, a mouse, a touch pen, a touch panel, etc., receives an input operation from a user, and outputs an operation signal according to the input operation to the processing unit 10. The reception unit 19 receives an input of setting information related to a threshold value used by the evaluation unit 13 to evaluate the accuracy of answer information, for example.

[0032] The processing unit 10 includes, for example, a computer system. The computer system mainly includes one or more processors and one or more memories as hardware. The functions of the question unit 11, the answer receiving unit 12, the evaluation unit 13, the prediction unit 14, the positioning unit 15, the notification unit 16, the setting unit 18, and the like are realized by the one or more processors executing the programs recorded in one or more memories of the computer system. In this embodiment, the processing unit 10 of the behavior prediction system 1 has the function of the positioning unit 15 of the positioning system 2. The program is pre-recorded in one or more memories of the computer system. The program may be provided through an electric communication line, or may be provided by recording it in a non-transitory recording medium such as a memory card, an optical disk, or a hard disk drive that can be read by the computer system. Note that the question unit 11, the answer receiving unit 12, the evaluation unit 13, the prediction unit 14, the positioning unit 15, the notification unit 16, and the setting unit 18 merely indicate the functions realized by the processing unit 10, and do not necessarily indicate the configuration with substance.

[0033] The questioning unit 11 transmits question information to the mobile terminal 3 carried by the user to be predicted via the communication unit 17 at the prediction time. The behavior prediction system 1, for example, sets the working hours of the next day as a first period, and predicts the intention of each of the multiple users to visit the facility F1 during the first period. Therefore, the behavior prediction system 1 sets a time point before the first period, for example, the end of the working hours on the day before the first period (for example, 5 p.m.) as the prediction time point, and the questioning unit 11 transmits question information to the multiple mobile terminals 3 carried by the multiple users via the communication unit 17 at the prediction time point (for example, 5 p.m. on the day before the first period). The network addresses (for example, IP addresses or MAC addresses, etc.) of the multiple mobile terminals 3 carried by the multiple users are registered in advance in the behavior prediction system 1, and the question information is transmitted to each of the multiple mobile terminals 3 using the registered network addresses. The question information may, for example, inquire as to whether the employee will come to the office (i.e., work in the office), stay at home (i.e., work from home), or whether it is undecided on the next day, which is the first period.

[0034] The response receiving unit 12 receives, via the communication unit 17, a plurality of pieces of response information returned from the plurality of mobile terminals 3, respectively.

[0035] The evaluation unit 13 obtains a degree of accuracy representing the accuracy of each of the multiple pieces of answer information received from the multiple mobile terminals 3. The evaluation unit 13 sets the period from three days before the prediction time to the day of the prediction time as the second period, for example. For example, when predicting whether or not the user will come to work on August 5, the evaluation unit 13 sets the period from August 1, three days before the prediction time (August 4), to August 4, the day of the prediction time, as the second period. The evaluation unit 13 evaluates the accuracy of the answer information by comparing the multiple pieces of answer information (for four days) received from the mobile terminal 3 with the attendance record to the facility F1 during this second period. The evaluation unit 13 acquires the positioning result of the positioning system 2 during the second period from the first storage unit DB1, and determines that the user came to work on the day the position of the mobile terminal 3 carried by the user was detected during the second period, and that the user was at home on the day the position of the mobile terminal 3 carried by the user was not detected.

[0036] Table 1 below shows the content of the response information sent from a mobile terminal 3 carried by a certain user A during the second period from August 1st to August 4th, and the visit record of user A. Note that the response information indicating the intention to visit on August 1st is actually information sent on July 31st, the day before August 1st. Similarly, the response information for each day from August 2nd to August 4th is actually information sent on each day from August 1st to August 3rd. Also, in Table 1, office work where you visit facility F1 and work is described as "commute to the office," and telecommuting where you work at home without visiting facility F1 is described as "working from home."

[0037] [Table 1]

[0038] The evaluation unit 13 obtains the attendance record of the user A from the positioning result of the positioning system 2. The first storage unit DB1 stores the positioning results of a plurality of mobile terminals 3 that have been positioned by the positioning system 2 during a predetermined period (for example, a period of several days to several months) until the prediction time. The evaluation unit 13 acquires the positioning result of the positioning system 2 that has been positioned for the position of the mobile terminal 3 carried by the user A during the second period from the first storage unit DB1. If the presence of the mobile terminal 3 is detected within the facility F1 on each day during the second period, the evaluation unit 13 determines that the user A has come to work, and if the presence of the mobile terminal 3 is not detected within the facility F1, the evaluation unit 13 determines that the user A has not come to work, that is, that the user A has been at home. Then, the evaluation unit 13 compares the response information for four days transmitted from the mobile terminal 3 with the attendance record for four days during the second period. The evaluation unit 13 obtains a plurality of first judgment values ​​binarized according to whether the presence or absence of visit for each day indicated by the answer information for four days (i.e., coming to work or staying at home) matches the attendance record for the four days. That is, the evaluation unit 13 obtains a plurality of first judgment values ​​binarized according to whether the presence or absence of visit indicated by each of the plurality of answer information received from the mobile terminal 3 in the second period matches the attendance record, in other words, whether the intention to visit matches the attendance record. For example, the evaluation unit 13 sets the value of the first judgment value to "1" if the presence or absence of visit indicated by the answer information (i.e., the intention to visit) matches the attendance record, and sets the value of the first judgment value to "0" if the presence or absence of visit indicated by the answer information (i.e., the intention to visit) does not match the attendance record.

[0039] Next, the evaluation unit 13 obtains a plurality of second judgment values ​​by weighting each of the plurality of first judgment values ​​with a weighting coefficient according to the elapsed time from the plurality of response time points at which the plurality of pieces of response information were respectively acquired to the prediction time point. The weighting coefficient is a value according to the elapsed time from the response time point of each of the plurality of pieces of response information to the prediction time point, and is set so that the value becomes smaller as the elapsed time from the response time point to the prediction time point becomes longer. For example, the weighting coefficients for the first judgment values ​​on August 4, 3, 2, and 1 in the second period are set to 1.0, 0.95, 0.9, and 0.8, respectively. The evaluation unit 13 obtains a plurality of second judgment values ​​by multiplying each of the plurality of first judgment values ​​by a weighting coefficient, and the plurality of second judgment values ​​are as shown in Table 1. Note that the value of the weighting coefficient is an example, and the value of the weighting coefficient can be appropriately changed within a range of 0 to 1 so that the value becomes smaller as the elapsed time from the response time point to the prediction time point becomes longer.

[0040] Then, the evaluation unit 13 calculates the accuracy based on a cumulative value obtained by accumulating a plurality of second judgment values. Specifically, the evaluation unit 13 calculates an average value by dividing the cumulative value obtained by accumulating a plurality of second judgment values ​​by the number of second judgment values, and calculates this average value as the accuracy. In the example of Table 1, the evaluation unit 13 calculates the accuracy of the answer information of user A to be 0.44. Note that the evaluation unit 13 calculates the average value of the plurality of second judgment values ​​as the accuracy, but the calculation method of the accuracy can be changed as appropriate, and for example, the median value of the plurality of second judgment values ​​may be calculated as the accuracy.

[0041] The prediction unit 14 predicts whether the user will come (i.e., whether they will come to work or stay at home) during the first period (e.g., August 5) based on the result of comparing the accuracy obtained by the evaluation unit 13 with a threshold value and the response information received from the mobile terminal 3 at the time of prediction (e.g., August 4). The prediction unit 14 stores the prediction result in the first storage unit DB1.

[0042] For example, when the accuracy is higher than the threshold, the prediction unit 14 predicts whether or not the person will come to work during the first period according to the presence or absence of attendance indicated by the answer information (i.e., coming to work or being at home), and when the accuracy is equal to or lower than the threshold, the prediction unit 14 predicts whether or not the person will come to work during the first period as being the opposite of the presence or absence of attendance indicated by the answer information. For example, when the threshold is set to 0.4, the accuracy obtained by the evaluation unit 13 is greater than the threshold, so if the content of the answer information acquired at the time of prediction is coming to work (coming to work), the prediction unit 14 predicts that the person will come to work during the first period. On the other hand, when the threshold is set to 0.5, the accuracy obtained by the evaluation unit 13 is less than the threshold, so if the content of the answer information acquired at the time of prediction is coming to work (coming to work), the prediction unit 14 predicts that the person will not come to work (coming to work) during the first period, that is, will be at home. In this way, the prediction unit 14 predicts whether or not the user will visit facility F1 during the first period (i.e., whether the user will come to work or stay at home) based on the response information obtained at the time of prediction and the accuracy obtained by the evaluation unit 13, which has the advantage of improving the accuracy of predictions as to whether or not the user will visit facility F1.

[0043] The notification unit 16 executes a notification step of notifying the mobile terminal 3 of the prediction result of the prediction unit 14. The notification unit 16 transmits information indicating the prediction result to the mobile terminal 3 of the notification destination via the communication unit 17, for example. The notification unit 16 also has a function of transmitting screen information displaying the locations of the multiple mobile terminals 3 (i.e., multiple users) in the facility F1, that is, the current attendance status, on a map, to the mobile terminal 3 of the notification destination via the communication unit 17, based on the locations of the multiple mobile terminals 3 measured by the positioning unit 15.

[0044] The setting unit 18 sets a threshold value to be compared with the accuracy of the answer information in the prediction step by the prediction unit 14. The setting unit 18 sets the threshold value based on the setting information accepted by the acceptance unit 19, for example, and can set the threshold value based on the setting information input by the user.

[0045] The first storage unit DB1 stores, for example, the position information of a plurality of portable terminals 3 whose positions are measured by the positioning unit 15.

[0046] The second storage unit DB2 stores reference information that can be used by the evaluation unit 13 when evaluating the accuracy of the answer information. The reference information includes information that influences the user's decision of whether to go to work, that is, whether to choose to work in the office or to work from home. For example, the reference information includes meteorological information about the weather in the area including the user's home and the facility F1, traffic-related information about traffic between the user's home and the facility F1, company information related to the company to which the user belongs or the facility F1, etc. The traffic-related information may include, for example, the operation status of public transportation between the user's home and the facility F1, the congestion status of roads between the user's home and the facility F1, etc. The company information includes the work schedules of multiple users, information about the department to which the user belongs, information about projects to which the user is involved, information about the lunch menu provided at the facility F1, etc.

[0047] (2.1.3) Mobile Devices The mobile terminal 3 includes a processing unit 30, a communication unit 33, a UI unit 34, and a beacon transmission unit 35. In this embodiment, the mobile terminal 3 is, for example, a smartphone carried by a user, but may be a computer terminal such as a tablet terminal having a communication function.

[0048] The communication unit 33 has a communication function for wirelessly communicating with the communication unit 17 of the behavior prediction system 1. The communication unit 33 is not limited to a unit that directly communicates with the communication unit 17 of the behavior prediction system 1, and may communicate via a repeater or the like installed in the facility F1. The communication method of the communication unit 33 is, for example, Bluetooth (registered trademark). The communication method of the communication unit 33 is not limited to Bluetooth, and may be WiFi (registered trademark) or a specific low-power radio in the 920 MHz band, or the like.

[0049] The UI unit 34 includes, for example, a touch panel 34A (see FIGS. 4 and 5) that combines a display device such as a liquid crystal display with an input device such as a touch pad. The UI unit 34 displays question information transmitted from the behavior prediction system 1, information showing the prediction result, screen information showing a map of the current attendance situation, and the like. When the UI unit 34 receives an operation of the user touching the screen of the display device, it outputs an operation signal according to the user's operation to the processing unit 30.

[0050] The processing unit 30 includes, for example, a computer system. The computer system mainly comprises one or more processors and one or more memories as hardware. The one or more processors execute the programs recorded in one or more memories of the computer system to realize the functions of the response sending unit 31, the display control unit 32, the attendance status request unit 36, and the like. The programs are pre-recorded in one or more memories of the computer system. The programs may be provided through an electric communication line, or may be provided by recording them in a non-transitory recording medium such as a memory card, an optical disk, or a hard disk drive that is readable by the computer system. Note that the response sending unit 31, the display control unit 32, and the attendance status request unit 36 ​​merely indicate the functions realized by the processing unit 30, and do not necessarily indicate the configurations with substance.

[0051] The display control unit 32 controls the display of the display device of the UI unit 34. The display control unit 32 displays a question screen (see FIG. 4) asking about the intention of visiting the facility F1 on the screen of the UI unit 34 based on the question information transmitted from the behavior prediction system 1. The display control unit 32 also displays a prediction result of the attendance situation to the facility F1 in a first period (for example, the next day) on the screen of the UI unit 34 based on the information indicating the prediction result transmitted from the behavior prediction system 1. The display control unit 32 also displays an image (see FIG. 5) showing the position of the user with a marker or the like on a floor plan of the facility F1 on the screen of the UI unit 34 (i.e., the screen of the touch panel 34A) based on the screen information for displaying a map of the current attendance situation transmitted from the behavior prediction system 1.

[0052] The answer sending unit 31 causes the communication unit 33 to send answer information, which is received by the UI unit 34 and includes an answer to the question information, to the behavior prediction system 1.

[0053] The attendance status request unit 36 ​​causes the communication unit 33 to transmit to the behavior prediction system 1 request information, which is received by the UI unit 34 and requests a display of the current attendance status to the facility F1, that is, the attendance status.

[0054] For example, when the mobile terminal 3 is present inside the facility F1, the beacon transmission unit 35 periodically transmits a beacon signal including the terminal ID of the mobile terminal 3 by radio wave signal. The mobile terminal 3 has, for example, a GPS positioning function that measures the current position using a GPS. The mobile terminal 3 can measure the current position by the GPS positioning function when outdoors where radio waves from a GPS satellite can be received. When the processing unit 30 detects that the mobile terminal 3 has entered the facility F1 based on the positioning result of the GPS positioning function, it causes the beacon transmission unit 35 to periodically transmit a beacon signal including the identification information of the mobile terminal 3.

[0055] (2.2) Operation Description Next, the operation of the behavior prediction system 1 of the present embodiment will be described with reference to Fig. 2 and Fig. 3. Note that the sequence diagrams shown in Fig. 2 and Fig. 3 are merely examples of the behavior prediction method performed by the behavior prediction system 1, and the order of the processes may be changed as appropriate, and processes may be added or omitted as appropriate.

[0056] When the prediction time arrives for predicting a user's attendance, the questioning unit 11 of the behavior prediction system 1 causes the communication unit 17 to transmit question information to the mobile terminal 3 carried by the user to be predicted, asking about the user's intention to attend during the first period (step S1).

[0057] When the communication unit 33 of the mobile terminal 3 receives the question information from the behavior prediction system 1, the display control unit 32 displays the question information on the screen of the UI unit 34 (the screen of the touch panel 34A) (step S2). FIG. 4 shows an example of the question information displayed on the screen of the UI unit 34 (the screen of the touch panel 34A). A display frame A1 is provided on the upper side of the screen of the UI unit 34 (the screen of the touch panel 34A) to display a message indicating the content of the question (for example, "Are you going to work tomorrow?"). In addition, three selection buttons B1 to B3 are displayed on the screen of the UI unit 34 (the screen of the touch panel 34A) below the display frame A1. The selection buttons B1, B2, and B3 are buttons operated to answer the question information with "Yes," "No," and "Undecided." When the user touches any of the selection buttons B1 to B3 (step S3), the response transmission unit 31 transmits the response information corresponding to the selection button touched by the user from the communication unit 33 to the behavior prediction system 1 (step S4).

[0058] When the answer receiving unit 12 receives the answer information transmitted from the mobile terminal 3 via the communication unit 17, the evaluation unit 13 executes an evaluation step (step S5) of evaluating the accuracy of the answer information transmitted from the mobile terminal 3.

[0059] The evaluation unit 13 obtains a plurality of first judgment values ​​by binarizing a plurality of pieces of answer information received from the mobile terminal 3 during a second period prior to the prediction time according to whether or not the presence or absence of attendance indicated by each of the plurality of pieces of answer information matches the attendance record. The evaluation unit 13 obtains a plurality of second judgment values ​​by weighting each of the plurality of first judgment values ​​with a weighting coefficient according to the elapsed time. The evaluation unit 13 then obtains an average value by dividing an accumulated value of the plurality of second judgment values ​​by the number of second judgment values, and obtains this average value as the probability.

[0060] The evaluation unit 13 may obtain the accuracy by adding a correction value based on the reference information stored in the second storage unit DB2 to the average value of the second judgment values. The reference information is information that affects the visit of users, and includes, for example, weather information, traffic-related information, company information, and the like in the first period. For example, when the weather in the first period is "cloudy" or "rainy," the user is more likely to actually work from home despite the content of the answer information being office work, compared to when the weather in the first period is "sunny." Therefore, it is preferable that the evaluation unit 13 corrects the accuracy by adding a correction value that is a negative value to the average value of the second judgment values. Also, if the content of the traffic-related information is that public transportation will be delayed in the first period, the user is more likely to actually work from home despite the content of the answer information being office work. Therefore, it is preferable that the evaluation unit 13 corrects the accuracy by adding a correction value that is a negative value to the average value of the second judgment values. Furthermore, if the content of the company information is that the work (e.g., a meeting) that the user to be predicted will perform at facility F1 is scheduled for the first period, the user is likely to work in the office as he or she answered that he or she will work in the office, so it is preferable for the evaluation unit 13 to correct the accuracy by adding a correction value, which is a positive value, to the average value of the second judgment values. In this way, the evaluation unit 13 can more accurately determine the accuracy of the answer information by correcting the accuracy of the answer information in consideration of the reference information.

[0061] When the evaluation unit 13 obtains the degree of accuracy, the prediction unit 14 compares the degree of accuracy with a threshold value. Then, the prediction unit 14 executes a prediction step of predicting whether or not the user will visit the facility, i.e., whether or not the user will come to work, during the first period, based on the answer information received by the answer receiving unit 12 and the result of comparing the degree of accuracy of the answer information with the threshold value (step S6).

[0062] Here, if the accuracy of the answer information is higher than the threshold, the user is likely to behave in accordance with the content of the answer information, so the prediction unit 14 predicts the user's behavior in the first period according to the content of the answer information. On the other hand, if the accuracy of the answer information is equal to or lower than the threshold, the user is likely to behave in a manner opposite to the content of the answer information, so the prediction unit 14 predicts the user's behavior in the first period opposite to the content of the answer information.

[0063] Furthermore, when the content of the response information is undecided, the prediction unit 14 may predict whether or not the user to be predicted will visit the facility F1 in the first period based on the probability of the user to be predicted visiting the facility F1 in the third period prior to the prediction time. The prediction unit 14 calculates the probability of the user to be predicted visiting the facility F1 in the third period (for example, the past n days) based on the visit record stored in the first storage unit DB1. For example, when n=100, if the number of days that the user to be predicted visited (came to work) the facility F1 in the third period including the past 100 working days is 42 days, the prediction unit 14 calculates the visit probability of the user to be predicted as 42%. Then, if the visit probability in the third period is higher than a predetermined reference value, the prediction unit 14 predicts that the user will come to work in the first period, and if the visit probability is equal to or lower than the reference value, the prediction unit 14 predicts that the user will be at home in the facility F1.

[0064] Furthermore, when the content of the response information is undecided, the prediction unit 14 may predict whether or not the prediction target user will visit in the first period based on the visit probability of the multiple users in a third period prior to the prediction time point. The first storage unit DB1 stores data on the visit records of the multiple users. The prediction unit 14 obtains the visit probability of the multiple users in the third period (e.g., the most recent one week, one month, etc.) based on the visit records of the multiple users stored in the first storage unit DB1. Then, if the visit probability in the third period is higher than a predetermined judgment value, the prediction unit 14 predicts that the prediction target user will come to work in the first period, and if the visit probability is equal to or lower than the judgment value, the prediction unit 14 predicts that the prediction target user will be at home in the facility F1.

[0065] In this embodiment, since multiple users use the facility F1, the behavior prediction system 1 performs a process of predicting whether or not each of the multiple users will visit the facility during the first period. That is, the questioning unit 11 of the behavior prediction system 1 transmits question information to multiple mobile terminals 3 carried by the multiple users. When the response receiving unit 12 of the behavior prediction system 1 receives multiple pieces of response information from the multiple mobile terminals 3, the evaluation unit 13 evaluates the accuracy of each of the multiple pieces of response information, and the prediction unit 14 predicts whether or not each of the multiple users will visit the facility during the first period based on the result of comparing the level of each of the multiple pieces of response information with a threshold and each of the multiple pieces of response information. Then, the prediction unit 14 performs a counting step of counting the results of the prediction of whether or not each of the multiple users will visit the facility during the first period (step S7), and obtains the attendance rate of the entire facility F1 during the first period. The prediction unit 14 stores the prediction result of the overall attendance rate in the first storage unit DB1.

[0066] Here, a prediction method for predicting the attendance rate of a group consisting of multiple users by the behavior prediction system 1 will be described below. For simplicity of explanation, an example will be described in which the attendance rate is calculated for a group consisting of three users A, B, and C.

[0067] The questioning unit 11 transmits question information to three mobile terminals 3 carried by three users A, B, and C. When the answer receiving unit 12 acquires answer information from the three mobile terminals 3, the prediction unit 14 predicts whether or not each of the three users A, B, and C will visit the facility during the first period based on the answer information received by the answer receiving unit 12 and the result of comparing the accuracy of each user determined by the evaluation unit 13 with a threshold value. That is, the questioning unit 11 transmits question information to a plurality of mobile terminals 3 carried by a plurality of users, and the answer receiving unit 12 receives answer information from the plurality of mobile terminals 3. Then, the prediction unit 14 predicts whether or not each of the plurality of users will visit the facility during the first period, and further predicts the visit status related to the number of visitors who will visit the facility F1 during the first period among the plurality of users. Here, the evaluation unit 13 evaluates the accuracy of the answer information for each of the three users A, B, and C by comparing the answer information received from the mobile terminal 3 with the visit record during the second period.

[0068] Table 2 below shows an example of response information and attendance records received from three mobile terminals 3 carried by three users A, B, and C during the second period. The evaluation unit 13 obtains a plurality of first judgment values ​​binarized for each of the three users A, B, and C according to whether or not the attendance indicated by the response information matches the attendance record, and further obtains a plurality of second judgment values ​​weighted by a weighting coefficient. The evaluation unit 13 then obtains an average value of the plurality of second judgment values, and regards the average value as the accuracy of the response information. Here, the accuracy of user A is obtained as 0.44, the accuracy of user B as 0.71, and the accuracy of user C as 0.20.

[0069] [Table 2]

[0070] Then, when the contents of the answer information given by users A, B, and C at the time of prediction are "come to work," "come to work," and "stay at home," respectively, and the threshold is set to 0.5, the prediction unit 14 predicts whether users A, B, and C will come to work during the first period based on the answer information and the result of comparing the accuracy of the answer information with the threshold. Here, since the accuracy of the answer information of users A and C is equal to or lower than the threshold, the prediction unit 14 predicts that user A will be at home and user C will come to work during the first period, contrary to the answer information. Since the accuracy of the answer information of user B is greater than the threshold, the prediction unit 14 predicts that user B will come to work during the first period, as per the answer information. In other words, the prediction unit 14 predicts that two of users A, B, and C will come to work, and predicts that the overall attendance rate will be 67%.

[0071] In addition, since the possibility that a user will come to work if they answer that they will be at home is low, the prediction unit 14 may predict that the user will be at home according to the content of the answer information even if the accuracy of the answer information is below the threshold value if the content of the answer information indicates that the user will be at home. In this case, the prediction unit 14 predicts that only user B will come to work, and predicts that the overall attendance rate is 33%.

[0072] In addition, since the possibility that a user will come to work if they answer that they will be at home is low, the threshold value when the content of the answer information is coming to work (hereinafter referred to as the first threshold value) and the threshold value when the content of the answer information is coming to work (hereinafter referred to as the second threshold value) may be different values. If the first threshold value when the content of the answer information is coming to work is 0.5 and the second threshold value when the content of the answer information is coming to work is 0.1, the prediction unit 14 compares the accuracy of the answer information with the first threshold value for users A and B who answered that they will be at work, and compares the accuracy of the answer information with the second threshold value for user C who answered that he will be at home. The accuracy of the answer information of user A who answered that he will be at work is 0.44, which is less than the first threshold value, so the prediction unit 14 predicts that user A will be at home in the first period, contrary to the answer information. The accuracy of the answer information of user B who answered that he will be at work is 0.71, which is higher than the first threshold value, so the prediction unit 14 predicts that user B will come to work in the first period as per the answer information. The accuracy of the response information of user C who responded that he / she will be at home is 0.2, which is higher than the second threshold, so the prediction unit 14 predicts that user C will be at home during the first period as per the response information. In other words, the prediction unit 14 predicts that only user B will be at work, and predicts that the overall attendance rate is 33%.

[0073] In addition, when the prediction unit 14 predicts that a certain user will come to work, the prediction unit 14 may predict the accuracy of the answer information of this user as the number of people coming to work, rather than counting the number of people coming to work as 1. For example, when the prediction unit 14 predicts that user B, whose answer information accuracy is 0.71, and user C, whose answer information accuracy is 0.20, will come to work, the prediction unit 14 may predict that the total number of people coming to work will be 0.91 people, which is the sum of the answer information accuracy, and the overall attendance rate will be about 30%.

[0074] Furthermore, the number of users using the facility F1 is not limited to three, and can be changed as appropriate depending on the size of the facility F1. When each of the multiple users using the facility F1 belongs to one of multiple groups, the prediction unit 14 may obtain the attendance rate for each of the multiple groups by aggregating the prediction results for each of the multiple groups. The multiple groups may be groups divided into business departments such as sales, general affairs, and research and development, or may be groups divided into departments that bring together multiple business departments, or may be groups divided into rooms, floors, or buildings that users use.

[0075] When the prediction unit 14 obtains the attendance rate for the entire facility F1 or the attendance rate for each group in the first period, the notification unit 16 causes the communication unit 17 to transmit the predicted results of the attendance rates to each of the multiple mobile terminals 3 (step S8).

[0076] When each of the multiple mobile terminals 3 receives the prediction result transmitted from the notification unit 16 via the communication unit 17, the display control unit 32 executes a display step (step S9) of displaying the prediction result on the screen of the UI unit 34. This allows each of the multiple users to understand the attendance rate in the first period by looking at the prediction result of the attendance rate displayed on the screen of the UI unit 34. For example, a user who responded that they will come to work in the first period can take an action such as changing their working style in the first period from coming to work (office work) to working from home (telecommuting) because their attendance rate in the first period is higher than the target value.

[0077] If the user changes the plan to come to work or stay at home during the first period, it is preferable to operate the mobile terminal 3 to notify the changed response information to the behavior prediction system 1. Upon acquiring the changed response information, the behavior prediction system 1 only needs to recalculate the attendance rate of the individual user and the attendance rate of the group to which the user belongs based on the changed response information, and it is preferable to notify the multiple mobile terminals 3 of the recalculation results.

[0078] Furthermore, when a user wants to know whether another user (hereinafter, referred to as a specific user) will come to work during the first period, the user operates the mobile terminal 3 to transmit a request signal requesting a prediction result for the specific user to the behavior prediction system 1. When the behavior prediction system 1 receives a request signal requesting a prediction result for the specific user from the mobile terminal 3, the notification unit 16 transmits the prediction result for the specific user to the requesting mobile terminal 3. The user can look at the prediction result for the specific user displayed on the screen of the UI unit 34 of the mobile terminal 3 and reconsider whether to work in the office or telecommute during the first period. The notification unit 16 may transmit the prediction result for the specific user and the accuracy of the specific user's answer information to the mobile terminal 3, and the UI unit 34 of the mobile terminal 3 may display the prediction result for the specific user and the accuracy together, so that the user can grasp the degree of accuracy of the specific user coming to work or staying at home.

[0079] The timing for the notification unit 16 to notify the results of the prediction of the attendance rate may be the night before the prediction of whether or not multiple users will attend has been completed, or the morning of the following day, before the start of the working hours.

[0080] Furthermore, the notification unit 16 may provide a notification to the user recommending working in the office or working from home based on the prediction result of the prediction unit 14. For example, if the prediction result of the attendance rate in the first period (the attendance rate of the entire facility F1, or the attendance rate of the group to which the user belongs) exceeds a predetermined target value, the notification unit 16 may transmit notification information such as "Congestion is predicted tomorrow. We recommend working from home" to the mobile terminal 3. If the prediction result of the attendance rate in the first period is equal to or lower than the target value, the notification unit 16 may transmit notification information such as "It looks like you will be able to work in the office tomorrow while maintaining social distance" to the mobile terminal 3.

[0081] In this embodiment, the positioning unit 15 of the behavior prediction system 1 periodically executes a positioning step of measuring the position of the mobile terminal 3 present in the facility F1 (step S10). For example, during working hours, when a user performs an operation to request the visit status using the UI unit 34 of the mobile terminal 3 in order to check the visit status to the facility F1, the visit status request unit 36 ​​executes a request step of transmitting request information requesting transmission of screen information displaying the visit status on a map from the communication unit 33 to the behavior prediction system 1 (step S11). When the communication unit 17 of the behavior prediction system 1 receives the request information from the mobile terminal 3, the communication unit 17 transmits screen information displaying the map of the position of the mobile terminal 3 present in the facility F1 to the requesting mobile terminal 3 based on the positioning result of the positioning unit 15 (step S12). When the communication unit 33 of the mobile terminal 3 receives the screen information of the map display from the behavior prediction system 1, the display control unit 32 displays an image of the map display on the screen of the UI unit 34 (the screen of the touch panel 34A) (step S13). Here, Fig. 5 shows an example of a map display image displayed on the screen of the UI unit 34 (the screen of the touch panel 34A). A floor plan MP1 showing the facility F1 is displayed on the screen of the UI unit 34 (the screen of the touch panel 34A). The floor plan MP1 shows desks TB1-TB5 and the like arranged in the facility F1, and a mark M1 like a black circle representing the user is displayed at the location (e.g., a seat) where the user's mobile terminal 3 is located. A user who has requested the attendance status can check the attendance status to the facility F1 by looking at the floor plan MP1 displayed on the UI unit 34 of the mobile terminal 3.

[0082] (3) Modifications The above embodiment is merely one of various embodiments of the present disclosure. The above embodiment can be modified in various ways depending on the design, etc., as long as the object of the present disclosure can be achieved. Modifications of the embodiment are listed below. The modifications described below can be applied in appropriate combination.

[0083] (3.1) Variation 1 Modification 1 differs from the above embodiment in that the setting unit 18 sets a threshold value based on the user's past visit record, etc. Note that the configuration of the behavior prediction system 1 other than the setting unit 18 is the same as in the above embodiment, so the same reference numerals are used for the common components, and the description thereof will be omitted.

[0084] Here, the setting unit 18 may set an individual threshold for each of the multiple users, or may set a common threshold for a group to which the multiple users belong.

[0085] When the setting unit 18 sets an individual threshold for each user, the setting unit 18 sets the threshold based on the percentage of the number of days the user visited (came to work) the facility F1 or the percentage of the number of days the user did not visit the facility F1 (the number of days the user was at home) in a fourth period before the prediction time. For example, in a fourth period including the past 100 working days, the number of days the user visited the facility F1 was 52 days, and the number of days the user was at home was 48 days. In this case, the setting unit 18 sets the threshold based on the percentage of the number of days the user visited the facility or the percentage of the number of days the user was at home. For example, the setting unit 18 sets the threshold when the content of the answer information is that the user came to work to 52%, which is the percentage of the number of days the user visited the facility F1, and sets the threshold when the content of the answer information is that the user was at home to 48%, which is the percentage of the number of days the user was at home.

[0086] In addition, when the setting unit 18 sets an individual threshold for each user, the setting unit 18 may set the first threshold and the second threshold based on the number of pieces of answer information when the content of the answer information of the user does not match the attendance record. For example, in the answer information of a certain user, among the answer information for the past 100 days when the content of the answer information does not match the attendance record, the number of days when the content of the answer information was coming to work but the user was actually at home was 52 days, and the number of days when the content of the answer information was at home but the user was actually at work was 48 days. In this case, the setting unit 18 sets the first threshold when the content of the answer information is coming to work to 52%, which is the ratio of the number of days when the content of the answer information was coming to work but the user was actually at home. In addition, the setting unit 18 sets the second threshold when the content of the answer information is at home to 48%, which is the ratio of the number of days when the content of the answer information was at home but the user was actually at work.

[0087] In addition, when the setting unit 18 sets a common threshold value for a group to which a plurality of users belong, the setting unit 18 may set the threshold value based on the ratio of the number of days that a plurality of users belonging to the group visited the facility F1 (came to work) or the ratio of the number of days that a plurality of users belonging to the group did not visit the facility F1 (the number of days that they stayed at home) in a predetermined period before the prediction time. In a predetermined period in which the total number of working days of a plurality of users belonging to a group is 1000 days, if the total number of days that a plurality of users visited the facility is 600 days and the total number of days that a plurality of users stayed at home is 400 days, the setting unit 18 may set the threshold value based on the ratio of the total number of days that a plurality of users visited the facility or the ratio of the total number of days that a plurality of users stayed at home. For example, the setting unit 18 sets the first threshold value when the content of the answer information is that the users came to work to 0.6, which is the ratio of the total number of days that a plurality of users stayed at home, and sets the second threshold value when the content of the answer information is that the users stayed at home to 0.4, which is the ratio of the total number of days that a plurality of users stayed at home.

[0088] In addition, when the setting unit 18 sets a common threshold for a group to which a plurality of users belong, the setting unit 18 may set the threshold based on the number of pieces of answer information that do not match the actual visit record to the facility F1 with respect to the number of answers of answer information acquired from a plurality of mobile terminals 3 carried by a plurality of users in the group during a fifth period before the prediction time. For example, among 1000 pieces of answer information in which the content of the answer information does not match the actual visit record in the group, the number of days when the content of the answer information was coming to work but the user was actually at home was 600 days, and the number of days when the content of the answer information was at home but the user was actually at work was 400 days. In this case, the setting unit 18 sets the first threshold when the content of the answer information is coming to work to 0.6, which is the ratio of the number of days when the content of the answer information was coming to work but the user was actually at home. In addition, the setting unit 18 sets the second threshold when the content of the answer information is at home to 0.4, which is the ratio of the number of days when the content of the answer information was at home but the user was actually at work.

[0089] (3.2) Other Modifications The same function as the behavior prediction system 1 may be embodied in a behavior prediction method, a computer program, or a non-transitory recording medium on which a program is recorded. The behavior prediction method according to one embodiment is a behavior prediction method for predicting whether or not a user to be predicted among a plurality of users will visit a facility F1 where a plurality of users visit and perform a predetermined action at a prediction time that arrives regularly or irregularly. The behavior prediction method includes a questioning step, a receiving step, an evaluation step, and a prediction step. In the questioning step, question information inquiring about the intention to visit the facility F1 in a first period after the prediction time is transmitted to a mobile terminal 3 carried by the user to be predicted. In the receiving step, answer information indicating the intention to visit the facility F1 is received from the mobile terminal 3. In the evaluation step, the accuracy of the answer information is evaluated by comparing the plurality of answer information received from the mobile terminal 3 with the visit record to the facility F1 obtained from the positioning result of a positioning system 2 that measures the position of the mobile terminal 3 in the facility F1 in a second period before the prediction time. In the prediction step, based on the answer information received from the mobile terminal 3 at the time of prediction and a result of comparing whether the accuracy is high or low with a threshold, it is predicted whether the user to be predicted will visit the venue during the first period. A (computer) program according to one aspect is a program for causing a computer system to execute a behavior prediction method.

[0090] Modifications of the above embodiment are listed below. The modifications described below can be applied in appropriate combinations.

[0091] The behavior prediction system 1, the positioning system 2, and the mobile terminal 3 in the present disclosure include a computer system. The computer system is mainly composed of a processor and a memory as hardware. The functions of the behavior prediction system 1, the positioning system 2, and the mobile terminal 3 in the present disclosure are realized by the processor executing a program recorded in the memory of the computer system. The program may be pre-recorded in the memory of the computer system, may be provided through an electric communication line, or may be recorded and provided in a non-transitory recording medium such as a memory card, an optical disk, or a hard disk drive that can be read by the computer system. The processor of the computer system is composed of one or more electronic circuits including a semiconductor integrated circuit (IC) or a large scale integrated circuit (LSI). The integrated circuits such as IC or LSI referred to here are called by different names depending on the degree of integration, and include integrated circuits called system LSI, VLSI (Very Large Scale Integration), or ULSI (Ultra Large Scale Integration). Furthermore, the manufacture of LSIs Later programmed, FPGA (Field-Programmable Gate Array) or LSI internal A logic device capable of reconfiguring the connection relationship of an LSI or reconfiguring the circuit partitions within the LSI can also be employed as a processor. The electronic circuits may be integrated into one chip or distributed among multiple chips. The chips may be integrated into one device or distributed among multiple devices. The computer system referred to here includes a microcontroller having one or more processors and one or more memories. Thus, the microcontroller is also composed of one or more electronic circuits including a semiconductor integrated circuit or a large-scale integrated circuit.

[0092] In the above embodiment, the questioning unit 11 may appropriately change the content of the question information to be transmitted to the mobile terminal 3. For example, the questioning unit 11 may acquire a work schedule from a schedule server that manages the work schedule of the user, and change the content of the question information based on the work schedule. If the work schedule includes a schedule for inspecting arriving luggage as a task performed by the user in the first period, the questioning unit 11 may transmit, for example, question information such as "Luggage is scheduled to arrive tomorrow. Will you come to work?" to the mobile terminal 3. If the work schedule includes a schedule for Mr. C, who is in charge of the same project as the user, to come to work, the questioning unit 11 may transmit, for example, question information such as "Mr. C is scheduled to come to work tomorrow. It looks like the work on the project registered in the schedule server will progress, so we recommend that you come to work." to the mobile terminal 3. The questioning unit 11 may also change the content of the question information according to the schedule in the facility F1 in the first period, such as the lunch menu provided at the cafeteria in the facility F1, events held in the facility F1, and weather forecasts.

[0093] Furthermore, if the attendance records of multiple users who use facility F1 are accumulated and the expected attendance rate in the first period can be calculated simply before the user to be predicted sends the question information to the mobile terminal 3, the question unit 11 may change the content of the question information based on the simple expected attendance rate. For example, if the attendance rate in the first period is expected to exceed a target value based on the past attendance records, the question unit 11 may notify the mobile terminal 3 of question information such as "It is expected that many people will come to work tomorrow. Will you come to work?", and the user can be encouraged to choose telecommuting.

[0094] In the present embodiment, the behavior prediction system 1 may be provided in the facility F1, or may be provided outside the facility F1.

[0095] In this embodiment, the behavior prediction system 1 has the function of the positioning unit 15 of the positioning system 2, but the behavior prediction system 1 and the positioning system 2 may be provided in separate housings. In this case, the behavior prediction system 1 only needs to obtain the positioning result of the mobile terminal 3 from the positioning system 2.

[0096] Furthermore, it is not essential for the behavior prediction system 1 that multiple functions are integrated into one housing, and the components of the behavior prediction system 1 may be distributed across multiple housings. Furthermore, at least some of the functions of the behavior prediction system 1, for example, some of the functions of the evaluation unit 13 and the prediction unit 14, may be realized by the cloud (cloud computing) or the like.

[0097] In the above embodiment, the case where the space used by the user is the work area where the user performs work has been described, but the space used is not limited to the work area. The behavior prediction system 1 may be applied to a facility such as a membership-based sports gym where the user is a member, and may predict the number of users who will use the facility on the next day.

[0098] In the above embodiment, facility F1 was a facility where users come and work, but facility F1 may also be a members-only sports facility where users come and exercise, or a cultural facility where users come and learn, etc.

[0099] In the above embodiment, in a comparison of two values ​​such as measurement data, "greater than" may be "greater than or equal to." In other words, in a comparison of two values, whether or not the two values ​​are equal can be arbitrarily changed depending on the setting of a reference value, etc., so there is no technical difference between "greater than" and "greater than or equal to." Similarly, "less than or equal to" may be "less than."

[0100] (summary) As described above, the first aspect of the behavior prediction method is a behavior prediction method for predicting whether or not a user to be predicted among a plurality of users will visit a facility (F1) where a plurality of users visit and perform a predetermined action at a prediction time that arrives regularly or irregularly. The first aspect of the behavior prediction method includes a questioning step, a receiving step, an evaluating step, and a predicting step. In the questioning step, question information inquiring about the intention to visit the facility (F1) in a first period after the prediction time is transmitted to a mobile terminal (3) carried by the user to be predicted. In the receiving step, answer information indicating the intention to visit the facility (F1) is received from the mobile terminal (3). In the evaluating step, the accuracy of the answer information is evaluated by comparing the plurality of answer information received from the mobile terminal (3) with the visit record to the facility (F1) obtained from the positioning result of the positioning system (2) in a second period before the prediction time. The positioning system (2) measures the position of the mobile terminal (3) within the facility (F1). In the prediction step, based on the answer information received from the mobile terminal (3) at the time of prediction and the result of comparing whether the accuracy is high or low against a threshold, it is predicted whether the user to be predicted will visit the venue during the first period.

[0101] According to this embodiment, it is possible to improve the accuracy of predicting whether or not a user will visit the facility (F1).

[0102] In the behavior prediction method of the second aspect, in the first aspect, in the prediction step, if the degree of certainty is higher than a threshold, the presence or absence of visit during the first period is predicted in accordance with the presence or absence of visit indicated by the answer information. In the prediction step, if the degree of certainty is equal to or lower than the threshold, the presence or absence of visit during the first period is predicted to be the opposite of the presence or absence of visit indicated by the answer information.

[0103] According to this aspect, by predicting whether or not a person will attend the event taking into consideration the accuracy of the response information, it is possible to improve the prediction accuracy.

[0104] In the behavior prediction method of the third aspect, in the first or second aspect, in the evaluation step, a plurality of first judgment values ​​are obtained by binarizing a plurality of pieces of answer information received from the mobile terminal (3) during the second period according to whether or not the intention to visit the facility (F1) matches the visit record. In the evaluation step, a plurality of second judgment values ​​are obtained by weighting each of the plurality of first judgment values ​​with a weighting coefficient. In the evaluation step, a degree of accuracy is obtained based on a cumulative value obtained by accumulating the plurality of second judgment values. The weighting coefficient is a value according to the elapsed time from the answer time when each of the plurality of pieces of answer information was acquired to the prediction time.

[0105] According to this embodiment, the accuracy of the answer information can be evaluated based on a plurality of pieces of answer information and the attendance record during the second period.

[0106] In the fourth aspect of the behavior prediction method, in any of the first to third aspects, if the content of the answer information received from the mobile terminal (3) at the time of prediction is undecided, the prediction step predicts whether or not the user to be predicted will visit the facility (F1) in the first period based on the probability of the user to be predicted visiting the facility (F1) in a third period prior to the time of prediction.

[0107] According to this embodiment, it is possible to improve the accuracy of predicting whether or not a user will visit the facility (F1).

[0108] In the fifth aspect of the behavior prediction method, in any of the first to third aspects, if the content of the answer information received from the mobile terminal (3) at the time of prediction is undecided, the prediction step predicts whether or not the user to be predicted will visit the facility (F1) in the first period based on the probability of multiple users visiting the facility (F1) in a third period prior to the time of prediction.

[0109] According to this embodiment, it is possible to improve the accuracy of predicting whether or not a user will visit the facility (F1).

[0110] The behavior prediction method of the sixth aspect is any one of the first to fifth aspects, further including a receiving step of receiving input of setting information related to a threshold, and a setting step of setting the threshold based on the setting information.

[0111] According to this aspect, the threshold value for comparing with the accuracy of the answer information can be set to an arbitrary value.

[0112] The behavior prediction method of the seventh aspect is any one of the first to fifth aspects, further including a setting step. In the setting step, a threshold is set based on the percentage of the number of days that the user to be predicted visited the facility (F1) during a fourth period prior to the prediction time point, or the percentage of the number of days that the user to be predicted did not visit the facility (F1).

[0113] According to this aspect, the threshold value for comparing with the accuracy of the answer information can be set to an arbitrary value.

[0114] In the behavior prediction method of the eighth aspect, in any one of the first to fifth aspects, the mobile terminal is one of a plurality of mobile terminals (3) carried by a plurality of users, and further includes a setting step of setting a threshold based on the number of pieces of answer information that do not match the actual visit record to the facility (F1) among the plurality of pieces of answer information acquired from the plurality of mobile terminals (3) carried by the plurality of users during a fifth period prior to the prediction time point.

[0115] According to this aspect, the threshold value can be set based on the proportion of the number of pieces of answer information provided by a plurality of users that do not match the attendance record.

[0116] In the behavior prediction method of the ninth aspect, in any of the first to eighth aspects, the mobile terminal is one of a plurality of mobile terminals (3) carried by a plurality of users. In the questioning step, question information is transmitted to a plurality of mobile terminals (3) carried by a plurality of users. In the receiving step, a plurality of pieces of answer information are received from the plurality of mobile terminals (3). In the predicting step, whether or not each of the plurality of users will visit the facility (F1) during a first period is predicted, and a visit status related to the number of visitors from the plurality of users who will visit the facility (F1) during the first period is further predicted.

[0117] According to this embodiment, it is possible to predict the attendance status of a plurality of users who use the facility (F1).

[0118] The behavior prediction method of the tenth aspect is any one of the first to ninth aspects, further including a notification step of notifying the mobile terminal (3) of the prediction result in the prediction step.

[0119] According to this embodiment, a user carrying the mobile terminal (3) can change whether or not to visit the facility (F1) during the first period, taking into account the prediction result.

[0120] A program according to an eleventh aspect is a program for causing one or more processors to execute the behavior prediction method according to any one of the first to tenth aspects.

[0121] According to this embodiment, it is possible to improve the accuracy of predicting whether or not a user will visit the facility (F1).

[0122] The behavior prediction system (1) of the twelfth aspect is a behavior prediction system (1) that predicts whether or not a user to be predicted among a plurality of users will visit a facility (F1) where a plurality of users visit and perform a predetermined action at a prediction time that arrives regularly or irregularly. The behavior prediction system includes a questioning unit (11), a response receiving unit (12), an evaluation unit (13), and a prediction unit (14). The questioning unit (11) transmits question information inquiring about the intention to visit the facility (F1) in a first period after the prediction time to a mobile terminal (3) carried by the user to be predicted. The response receiving unit (12) receives response information indicating the intention to visit the facility (F1) from the mobile terminal (3). The evaluation unit (13) evaluates the accuracy of the response information based on a plurality of pieces of response information received from the mobile terminal (3) in a second period before the prediction time and a record of visits to the facility (F1) in the second period obtained from the positioning result of the positioning system (2). The positioning system (2) measures the position of the mobile terminal (3) within the facility (F1). The prediction unit (14) predicts whether or not the user to be predicted will visit the facility during the first period based on the answer information received from the mobile terminal (3) at the time of prediction and the result of comparing the accuracy with a threshold value.

[0123] According to this embodiment, it is possible to improve the accuracy of predicting whether or not a user will visit the facility (F1).

[0124] In the behavior prediction system (1) of the thirteenth aspect, in the twelfth aspect, the positioning system (2) includes a beacon transmitting unit (35) provided in the mobile terminal (3), a plurality of beacon receiving units (20) provided in the facility (F1), and a positioning unit (15). The beacon transmitting unit (35) transmits a beacon signal. The plurality of beacon receiving units (20) each receive the beacon signal transmitted from the beacon transmitting unit (35). The positioning unit (15) determines the position of the mobile terminal (3) within the facility (F1) based on the plurality of beacon signals received by the plurality of beacon receiving units (20), respectively.

[0125] According to this embodiment, the positioning system (2) measures the position of the mobile terminal (3) within the facility (F1), and thus the position of the user carrying the mobile terminal (3) can be measured.

[0126] Not limited to the above aspects, various configurations (including modified examples) of the behavior prediction system (1) according to the embodiment can be embodied as a behavior prediction method of the behavior prediction system (1), a (computer) program, or a non-transitory recording medium having a program recorded thereon, etc.

[0127] The configurations according to the second to tenth aspects are not essential for the behavior prediction method and may be omitted as appropriate. The configuration according to the thirteenth aspect is not essential for the behavior prediction system (1) and may be omitted as appropriate. [Explanation of symbols]

[0128] 1. Behavioral Prediction System 2 Positioning System 3. Mobile devices 11 Question Section 12. Response Receiving Department 13 Evaluation Section 14 Prediction Department 15 Positioning unit 20 Beacon receiver 35 Beacon transmitter

Claims

1. A behavior prediction method executed by one or more processors, comprising: The behavior prediction method is a method for predicting whether or not a user to be predicted will visit a facility where a plurality of users visit and perform a predetermined behavior at a prediction time that arrives regularly or irregularly, a questioning step of transmitting question information inquiring about an intention to visit the facility in a first period after the prediction time point to a mobile terminal carried by the user to be predicted; a receiving step of receiving answer information indicating an intention to visit the facility from the mobile terminal; an evaluation step of evaluating accuracy of the answer information by comparing the plurality of pieces of answer information received from the mobile device with a visit record to the facility obtained from a positioning result of a positioning system that measures a position of the mobile device within the facility during a second period before the prediction time point; and a prediction step of predicting whether or not the user to be predicted will visit the facility during the first period based on the answer information received from the mobile terminal at the time of prediction and a result of comparing the accuracy with a threshold value. Behavioral prediction methods.

2. In the prediction step, If the degree of certainty is higher than the threshold value, predicting whether or not there will be attendance during the first period according to the presence or absence of attendance indicated by the response information; If the degree of certainty is equal to or less than the threshold value, predicting whether or not the customer will attend the event during the first period to be the opposite of whether or not the customer will attend the event indicated by the response information. The behavior prediction method according to claim 1 .

3. In the evaluation step, determining a plurality of first determination values ​​by binarizing the plurality of pieces of answer information received from the mobile terminal during the second time period according to whether or not the intention to visit the facility coincides with the past visit record; determining a plurality of second determination values ​​by weighting each of the plurality of first determination values ​​with a weighting coefficient; The degree of certainty is calculated based on a cumulative value obtained by accumulating the plurality of second determination values, The weighting coefficient is a value according to the elapsed time from the response time when each of the plurality of pieces of answer information was acquired to the prediction time. The behavior prediction method according to claim 1 .

4. When the content of the answer information received from the mobile terminal at the prediction time point is undetermined, the prediction step predicts whether or not the user to be predicted will visit the facility in the first period based on a probability of the user to be predicted visiting the facility in a third period prior to the prediction time point. The behavior prediction method according to any one of claims 1 to 3.

5. When the content of the answer information received from the mobile terminal at the prediction time point is undetermined, the prediction step predicts whether or not the prediction target user will visit the facility in the first period based on the probability of the plurality of users visiting the facility in a third period prior to the prediction time point. The behavior prediction method according to any one of claims 1 to 3.

6. a receiving step of receiving input of setting information related to the threshold value; and setting the threshold value based on the setting information. The behavior prediction method according to any one of claims 1 to 5.

7. and setting the threshold value based on a ratio of the number of days that the user to be predicted visited the facility or a ratio of the number of days that the user to be predicted did not visit the facility during a fourth period prior to the prediction time point. The behavior prediction method according to any one of claims 1 to 5.

8. the mobile terminal is one of a plurality of mobile terminals carried by each of the plurality of users; and setting the threshold value based on the number of pieces of answer information that do not match the actual visit record to the facility among the plurality of pieces of answer information acquired from the plurality of mobile terminals carried by the plurality of users during a fifth period prior to the prediction time point. The behavior prediction method according to any one of claims 1 to 5.

9. the mobile terminal is one of a plurality of mobile terminals carried by each of the plurality of users; In the questioning step, the question information is transmitted to the plurality of mobile terminals carried by the plurality of users, In the receiving step, a plurality of pieces of answer information are received from the plurality of mobile terminals, In the prediction step, the presence or absence of attendance of each of the plurality of users during the first period is predicted, and a visitation status related to the number of visitors among the plurality of users who will visit the facility during the first period is further predicted. The behavior prediction method according to any one of claims 1 to 8.

10. The method further includes a notification step of notifying the mobile terminal of a prediction result in the prediction step. The behavior prediction method according to any one of claims 1 to 9.

11. One or more processors, To execute the behavior prediction method according to any one of claims 1 to 10, program.

12. 1. A behavior prediction system for predicting whether a user to be predicted will visit a facility where a plurality of users visit and perform a predetermined behavior at a prediction time that arrives regularly or irregularly, comprising: a questioning unit that transmits question information inquiring about an intention to visit the facility in a first period after the prediction time point to a mobile device carried by the user who is the prediction target; a response receiving unit that receives response information indicating an intention to visit the facility from the mobile terminal; an evaluation unit that evaluates accuracy of the answer information based on a plurality of pieces of answer information received from the mobile device during a second period prior to the prediction time point and a visit record to the facility during the second period obtained from a positioning result of a positioning system that measures a position of the mobile device within the facility; and a prediction unit that predicts whether or not the user to be predicted will visit the facility during the first period based on the answer information received from the mobile terminal at the time of prediction and a result of comparing the accuracy with a threshold value. Behavioral prediction system.

13. the positioning system includes a beacon transmitting unit included in the mobile terminal, a plurality of beacon receiving units provided in the facility, and a positioning unit; the beacon transmitter transmits a beacon signal; the plurality of beacon receivers each receive the beacon signal transmitted from the beacon transmitter; The positioning unit determines a position of the mobile terminal within the facility based on the beacon signals received by the beacon receiving units, respectively. The behavior prediction system according to claim 12.

Images