JP5331719B2 - Method, apparatus and program for determining action situation - Google Patents

Description

  The present invention relates to an action situation determination method, a determination apparatus, and a determination program for determining, for example, whether a user's action pattern corresponds to a weekday or a holiday.

  In recent years, various services have been proposed for determining a user's behavioral situation and providing a comment or the like suitable for the determined behavioral state, and one of the services is as follows. That is, the user's pre-registered action time information such as his / her wake-up time, attendance time, return time, sleep time, etc., user's stay location data obtained using GPS (Global Positioning System), and the stay location data On the other hand, based on the annotation information given by the user, two kinds of situation transition models for weekdays and holidays are learned and created. In this state, the current situation of the user is estimated based on the acquired position data and time information of the user and the situation before the user transition, and a comment corresponding to the estimated situation is provided to the user. (For example, see Patent Document 1).

JP 2009-182862 A

  By the way, in the above conventional method, a weekday / holiday determination is performed by comparing a time series change of a transition of a specific staying place obtained from acquired staying position data with a predetermined weekday / holiday model. In this method, all the stay position data obtained for each time zone are handled under the same conditions. However, the user's behavior pattern for each time zone includes various patterns such as a pattern in which the user goes to various places and the stay places are dispersed, and a pattern in which the user continues to stay in a predetermined place. Therefore, if the behavior pattern for each time zone is handled under uniform conditions, it is difficult to accurately determine weekdays / holidays.

  The present invention has been made paying attention to the above circumstances, and its purpose is to reduce the influence even when the user's behavior pattern changes variously for each time zone, and to increase the judgment of weekdays / holidays. An object of the present invention is to provide an action situation determination method, a determination apparatus, and a determination program that can be accurately performed.

In order to achieve the above object, according to one aspect of the present invention, user's stay position data is collected and stored in a position data storage means . In this state, first, the day is divided into a plurality of time sections and the map information including the stay range of the user is divided into a plurality of areas. Based on the stay position data stored in the position data storage unit , For each of the divided time sections, a probability density distribution representing a user's stay status in each divided area of the map information is calculated and stored in the probability density storage means . Next, a weight for each of the divided time intervals for the calculated probability density distribution is calculated and stored in a weight storage means . Then, the probability density and its weight in the time interval and the divided area including the user's stay position data collected on the date to be determined and stored in the position data storage means are stored in the probability density storage means and the weight storage means. Search based on the stored probability density distribution and weight, respectively, and use the searched probability density and the weight to calculate the probability that the determination target day is a weekday or a holiday, Based on the probability, it is determined whether the day to be determined is a weekday or a holiday . Further, when calculating the weight, for each of the divided time sections, the average of the probability density of each divided area in the time section is calculated for the divided area in which the stay position data has been mapped in the past in the time section. By dividing by the number, the weight for the probability density of the time interval is calculated, the weight calculated for each of the divided time intervals is normalized based on the maximum value, and this normalized frequency distribution is The weight for the probability density for each time interval is used.

Specifically, the following aspects are provided.
First, when calculating the probability density distribution, for each divided time section, the user's stay position data collected in the same same time section is mapped to map information, and the map on which this stay position data is mapped The information is divided into a plurality of areas in a mesh form, and a frequency distribution of the stay position data for each of the divided areas is created, and the created frequency distribution is normalized based on the maximum value and this normalization is performed. The obtained frequency distribution is set as the probability density in the divided area for each of the divided time intervals.

  Furthermore, when calculating the probability that the date to be determined is a weekday or a holiday, the user's stay location data collected and stored on the date to be determined is mapped to the map information, and the stay location data is Based on the mapped map information, the probability density in the time interval and the divided area including the stay position data is searched from the calculated probability density distribution, and based on the map information on which the stay position data is mapped. In addition, the weight for the probability density in the time interval and the divided area including the stay position data is searched from the calculated weight. Then, the average of the probability density of the position data collected on the date to be determined is calculated using the searched probability density and its weight, and the average value of the calculated probability density becomes the determination target. The probability that the day is a weekday or a holiday.

  Therefore, according to one aspect of the present invention, a weight that contributes to weekday / holiday determination is calculated for each time zone from the distribution density and spread of past position data, and the weekday / holiday is calculated in consideration of the calculated weight. A determination is made. As a result, in consideration of the distribution density and spread of the past position data, the weights contributing to weekday / holiday determination are automatically changed for each time interval. It is possible to determine weekdays / holidays in consideration of the difference. For this reason, for example, the user's location data collected in the past is not dispersed in various regions, and the time interval in which the user is considered likely to have stayed in a specific place, and the user's location data collected in the past are various. Even if the user stays in a place that is not in the stay history, it is possible to accurately determine the weekday / holiday by distinguishing it from the time period that would otherwise not affect the judgment process. It becomes.

  In other words, according to the present invention, even when the user's behavior pattern changes variously for each time zone, the behavior situation determination method, the determination device, and the determination of the weekday / holiday can be performed with high accuracy by reducing the influence A determination program can be provided.

It is a figure which shows the whole structure of the system which implements the action condition determination method concerning one Embodiment of this invention. It is a block diagram which shows the function structure of the system shown in FIG. It is a flowchart which shows the processing procedure and processing content of the learning phase by the system shown in FIG. It is a flowchart which shows the processing procedure and processing content of the weekday / holiday determination processing phase by the system shown in FIG. It is a flowchart which shows the procedure and processing content of the reading / selection process of the position data in the learning phase shown in FIG. It is a figure which shows the result of having mapped the position data collected by the same specific time area in the past obtained by the reading / selection process of the position data shown in FIG. 4 on map data. It is a flowchart which shows the procedure and process content of the probability density calculation process in the learning phase shown in FIG. It is a figure which shows an example of the map data meshed used by the probability density calculation process shown in FIG. It is a figure which shows an example of the frequency distribution for every mesh obtained by the probability density calculation process shown in FIG. It is a flowchart which shows the procedure and processing content of the weight calculation process with respect to the probability density of each time slot | zone in the learning phase shown in FIG. 5 is a flowchart showing the procedure and processing contents of a weekday / holiday determination process in the weekday / holiday determination process phase shown in FIG.

Embodiments according to the present invention will be described below with reference to the drawings.
FIG. 1 is a schematic configuration diagram of a system for realizing a situation determination method according to an embodiment of the present invention. This system includes a mobile terminal MS possessed by a user who is a situation determination target, a database server DSV, a weekday / holiday determination processing device PS, a user terminal UT that is a notification destination of a weekday / holiday determination result, and a plurality of applications. Servers AP1 to APn are provided, and these can communicate with each other via a communication network. The communication network includes, for example, an IP network represented by the Internet and an access network for accessing the IP network.

FIG. 2 is a block diagram showing a functional configuration of the system shown in FIG.
The mobile terminal MS has an input / output function as a man / machine interface, a voice communication function, and a browser function as basic functions, and further includes a sensor data measuring unit 11 and functions necessary for carrying out the present invention. A weekday annotation input unit 12 is provided. All of these functions are realized by causing a CPU (Central Processing Unit) in the mobile terminal MS to execute the program.

  The sensor data measuring unit 11 has a GPS (Global Positioning System) receiver, receives GPS signals transmitted from a plurality of GPS satellites (not shown) by the GPS receiver, and is represented by latitude and longitude. MS position data is calculated. Then, a process of transmitting the calculated position data to the database server DSV together with a time stamp indicating the measurement time is performed.

  The weekday annotation input unit 12 displays an input guidance screen for weekday / holiday annotation information on the display unit. In this state, the user operates the input unit to determine whether the day is a weekday with respect to the past days when the position data is measured. Alternatively, when annotation information indicating a holiday is input, a process of receiving the annotation information and transmitting it to the database server DSV is performed.

  The database server DSV includes a communication unit that performs communication between a large-capacity storage medium and the portable terminal MS and the weekday / holiday determination processing device PS, and further, as a storage control function necessary for carrying out the present invention. , A sensor data storage unit 21, a weekday annotation storage unit 22, a probability density storage unit 23, and a weight storage unit 24 for storing a weight for the probability density. All of these storage control functions are realized by causing the CPU in the database server DSV to execute the program.

  The sensor data storage unit 21 receives the position data of the mobile terminal MS transmitted from the mobile terminal MS and stores it in the position data storage area in the storage medium, and stores the stored position data on weekdays / holidays. A process of selectively reading in response to a request from the determination processing device PS and transmitting it to the requesting weekday / holiday determination processing device PS is performed.

  The weekday annotation storage unit 22 receives the annotation information transmitted from the mobile terminal MS and stores it in a predetermined storage area in the storage medium, and the stored annotation information from the weekday / holiday determination processing device PS. Is read in response to the request and transmitted to the requesting weekday / holiday determination processing device PS.

  The probability density storage unit 23 receives information representing the probability density calculated by a probability density calculation processing unit 31 (to be described later) in the weekday / holiday determination processing device PS and stores the information in a predetermined storage area in the storage medium. The information representing the stored probability density is read in response to a request from the weekday / holiday determination processing device PS, and transmitted to the requesting weekday / holiday determination processing device PS.

  The weight storage unit 24 for the probability density receives information indicating the weight for the probability density calculated by the weight calculation processing unit 32 for the probability density (to be described later) in the weekday / holiday determination processing device PS, and performs predetermined storage in the storage medium. Processing for storing in the area, and processing for reading information indicating the weight for the stored probability density in response to a request from the weekday / holiday determination processing device PS and transmitting the information to the requesting weekday / holiday determination processing device PS Do.

  The weekday / holiday determination processing device PS is composed of, for example, a Web server. As processing functions necessary for carrying out the present invention, a position data reading / selection processing unit (not shown), a probability density calculation processing unit 31, A weight calculation processing unit 32 for probability density and a weekday / holiday determination processing unit 33 are provided. All of these processing functions are realized by causing the CPU in the database server DSV to execute the program.

  The position data reading / selection processing unit reads the annotation information from the weekday annotation storage unit 22 of the database server DSV, and selectively reads the daily position data annotated as weekdays from the annotation information from the sensor data storage unit 21. . A map in which one day is divided into a plurality of time intervals, the selectively read position data is mapped on a map for each divided time interval, and the position data is mapped for each time interval. A process of transmitting information to the database server DSV and storing it in the sensor data storage unit 21 is performed.

  The probability density calculation processing unit 31 reads map information in which position data is mapped for each time interval from the sensor data storage unit 21 of the database server DSV. Then, the map information is divided into a plurality of areas in a mesh shape, and the distribution of the user's stay probability density for each of the divided areas for each of the time intervals is set to the number of mapping position data in the read map information. And calculate. Then, the information representing the calculated probability density distribution is transmitted to the database server DSV and stored in the probability density storage unit 23.

  The weight calculation processing unit 32 for the probability density reads the position data mapped on the map for each time interval from the sensor data storage unit 21 of the database server DSV, and from the probability density storage unit 23 for each time interval for each time interval. Reads information representing the probability density distribution for an area. Then, based on the read information, a weight for the probability density for each time interval is calculated, and information representing the weight for the calculated probability density for each time interval is transmitted to the database server DSV to obtain the probability density. The process which memorize | stores in the weight memory | storage part 24 is performed.

  The weekday / holiday determination processing unit 33 reads the position data collected on the weekday / holiday determination target from the sensor data storage unit 21 of the database server DSV. At the same time, information representing the probability density distribution and information representing the weight are read from the probability density storage unit 23 and the probability density storage unit 24 of the database server DSV, respectively. The read position data is subjected to weekday / holiday determination using information indicating the read probability density distribution and information indicating its weight, and information indicating the determination result is used as the user terminal UT or the application server AP1. ~ Process to transmit to APn.

  The user terminal UT or the application servers AP1 to APn includes a weekday / holiday notification receiving unit 41 as a control function according to the present invention. The weekday / holiday notification receiving unit 41 is realized by causing the CPU to execute a program. The weekday / holiday judgment processing unit 33 of the weekday / holiday judgment processing device PS transmits the weekday / holiday judgment result. Information to be received is received and stored as one parameter used for determining the user's situation.

  Next, the weekday / holiday determination method by the system configured as described above will be described by dividing it into a probability density distribution and its weight calculation phase and a weekday / holiday determination phase.

(1) Probability density distribution and its weight calculation phase The probability density distribution and its weight are calculated as follows. FIG. 3 is a flowchart showing the processing procedure and processing contents. Here, it is assumed that the user position data measured over a certain past period is already stored in the sensor data storage unit 21 in the database server DSV together with a time stamp indicating the measurement time. .

(1-1) Weekday Annotation Input Processing In the mobile terminal MS, the weekday annotation input unit 12 is activated in step S1, and annotation information indicating whether the day is a weekday or a holiday with respect to the past days when the position data is measured. Input processing is performed. This input process is performed in an interactive manner between the mobile terminal MS and the user.

  For example, a calendar-type GUI is displayed on the display of the mobile terminal MS, and the user first selects and inputs the day when the position data is measured, and further selects and inputs the day of the week or the day by day. In addition to the above GUI format input, a question such as “Is the holiday on a calendar?” Is displayed, and a rough holiday annotation process is executed from the answer entered by the user for this question. It is also possible to do.

  Then, the weekday annotation input unit 12 of the mobile terminal MS transmits the input annotation information indicating the weekday / holiday to the database server DSV. As a result, the input annotation information is stored in the weekday annotation storage unit 22 of the database server DSV.

(1-2) Position Data Reading / Selection Process The weekday / holiday determination processing device PS first activates the position data reading / selection processing section in step S2 and measures the day when all position data is annotated as weekdays. A process for selectively reading the obtained position data is executed. FIG. 5 is a flowchart showing the processing procedure and processing contents.

That is, first, in step S21, annotation information is read from the weekday annotation storage unit 22 of the database server DSV, and daily position data annotated as weekdays by this annotation information is selectively read from the sensor data storage unit 21. Next, one day is divided into a plurality of time intervals τi (i = 1, 2,..., L), and the weekday position data selectively read in step S22 to step S26 is converted to the time stamp tj. Based on the above, each divided time interval τi is mapped on the map. Then, the map information for each time interval τi to which the position data is mapped is transmitted to the database server DSV in step S27 and stored in the sensor data storage unit 21.
FIG. 6 shows an example of a map in which position data measured in a time section from 8 o'clock to 8:30 is mapped when a day is divided every 30 minutes. Note that the length of the time interval and the start time of the time interval delimiter are arbitrary.

(1-3) Probability Density Distribution Calculation Processing The weekday / holiday determination processing device PS next activates the probability density calculation processing unit 31 in step S3, and the user for each divided area on the map for each time interval τi. The process of calculating the distribution of the stay probability density is executed as follows. FIG. 7 is a flowchart showing the processing procedure and processing contents.

  The probability density calculation processing unit 31 first initializes the number i of the time interval τi to = 1 in step S31, and stores the position data mapped to the map information of the time interval τi in step S32 in the sensor data storage unit of the database server DSV. 21.

  In step S33, the map information is first divided into a plurality of areas each having a predetermined size in a mesh shape. The size of each divided area can be arbitrarily set. FIG. 8 shows an example of map information when the map on which the position data shown in FIG. 7 is mapped is divided into meshes. Subsequently, the number of position data mapped to each divided area on the map is counted, and a frequency distribution is created on the map. FIG. 9 shows an example of map information when the number of position data mapped for each divided area is entered and this is used as a frequency distribution.

ここで、Ｃ（τi ，x,y）は時間区間τi 、座標(x,y) で示される分割エリアでカウントされたれた位置データの数、Ｃmax(τi)は時間区間τiに各分割エリアでカウントされた位置データのうち最大のカウント数と示す。また、Ｌは１日を一定の時間長で区切ったときの時間区間の総数であり、例えば１つの時間区間の長さが３０分であれば、Ｌ＝４８となる。 Next, in step S34, the frequency distribution of each divided area is normalized using the maximum number of mapping values in the same time interval τi. Then, the probability density p (τi, x, y) in the divided area of the time interval τi and coordinates (x, y) is calculated by the following equation (1) based on the normalized frequency distribution.

Here, C (τi, x, y) is a time interval τi, the number of position data counted in the divided area indicated by coordinates (x, y), and Cmax (τi) is a time interval τi in each divided area. This is indicated as the maximum count number among the counted position data. L is the total number of time sections when one day is divided by a certain time length. For example, if the length of one time section is 30 minutes, L = 48.

The processes in steps S32 to S34 are repeated every time i is incremented in step S36. Then, the probability density p (τi, x, y) for all time intervals τi (i = 1, 2,..., L) is calculated, and it is detected in step S35 that i has reached L. Then, the probability density calculation process ends.
The calculated probability density p (τi, x, y) is transmitted to the database server DSV and stored in the probability density storage unit 23.

(1-4) Weight Calculation Processing for Probability Density Distribution The weekday / holiday determination processing device PS next activates the weight calculation processing unit 32 for probability density in step S4, and weights for the probability density of each area for each time interval τi. The process of calculating is executed as follows. FIG. 10 is a flowchart showing the processing procedure and processing contents.

ここで、ｎiは時間区間τiに位置データのカウントが確認された総分割エリア数とする。
すなわち、特定の時間区間τiにおける確率密度の平均を、特定の時間区間τiにおいて過去に位置データがマッピングされた分割地域の数で割り算する。 The weight calculation processing unit 32 for probability density first initializes the number i of the time interval τi to = 1 in step S41, and then in step S42, from the sensor data storage unit 21 of the database server DSV for each time interval τi. The position data mapped on the map is read, and the probability density p (τi, x, y) for each area for each time interval τi is read from the probability density storage unit 23. In step S43, the weight w (τi) is calculated by the following equation (2) based on the read data.

Here, ni is the total number of divided areas in which the position data count is confirmed in the time interval τi.
That is, the average of the probability density in a specific time interval τi is divided by the number of divided areas to which position data has been mapped in the past in the specific time interval τi.

  The weight calculation processing unit 32 for the probability density calculates the weight w (τi) for the probability density p (τi, x, y) for each time interval τi in step S43, and increments the value of i in step S46. Repeatedly. Meanwhile, in step S44, the weight max w (τj) (j = 1, 2,..., L) taking the maximum value is saved while being updated.

すなわち、上記（２）式により算出された値を、すべての時間区間の中で最大となる重みmax ｗ（τj）（j=1,2,…,L）を用いて正規化した値を、特定の時間区間τiの位置データに対する重みＷ（τi）とする。 When it is detected in step S45 that the value of i has reached L, the weight calculation processing unit 32 for the probability density moves to step S47. Then, after initially setting the number i of the time interval τi to = 1, in step S48, the weight w (τi) calculated for each time interval τi and the maximum value max w (τj) (j = 1) , 2,..., L), the weight W (τi) for the probability density distribution in the time interval τi is calculated using the following equation (3).

That is, a value obtained by normalizing the value calculated by the above equation (2) using the weight max w (τj) (j = 1, 2,..., L) that is the maximum in all time intervals, A weight W (τi) for position data in a specific time interval τi is used.

  The calculation process of the weight W (τi) in step S48 is repeated every time the value of i is incremented in step S50, and ends when it is detected in step S49 that the value of i has reached L. . Information representing the weight W (τi) for the calculated probability density distribution for each time interval τi is transmitted to the database server DSV in step S5 and stored in the weight storage unit 24 for probability density.

(2) Weekday / Holiday Determination Phase Next, the processing of the weekday / holiday determination phase is performed as follows. FIG. 4 is a flowchart showing the processing procedure and processing contents.
In the weekday / holiday determination processing device PS, the weekday / holiday determination processing unit 33 first reads the position data collected on the weekday / holiday determination target day from the sensor data storage unit 21 of the database server DSV in step S6. . At the same time, information representing the probability density p (τi, x, y) and information representing the weight W (τi) are read from the probability density storage unit 23 and the weight storage unit 24 for the probability density of the database server DSV.

  In step S7, a probability density distribution suitable for collating the read position data is searched. This search process corresponds to searching for a time interval τi including the read position data. When the search for the probability density appropriate for this comparison is completed, the weights for the frequency distribution in the same time interval are searched similarly in step S8.

  When these search processes are completed, the weekday / holiday determination processing unit 33 then uses the weights for the searched probability density and frequency distribution as parameters in step S9 to determine whether the weekday / holiday determination target day is a weekday or The process of determining whether it is a holiday is executed as follows. FIG. 11 is a flowchart showing the processing procedure and processing contents.

ここで、ｍはデータ取得日における、平日／休日判定処理開始時点までに得られた最終データ収集時刻、ｌはデータ取得日におけるデータ収集開始時刻である。また、Ｎは時間区間ｍ−ｌ間に収集した位置データの個数を示し、Ｗ（τi）は時刻ｔの該当する時間区間τiにおける確率密度（τi ，x,y）に対する重みを示す。さらに、ｘt，ｙtは、時刻ｔにおいて計測された緯度、経度を示す。
すなわち、各位置データの確率密度の平均値が、位置データの計測日が平日である確率Ｐ（ｍ）となる。 That is, first in step S91, the probability P (m) that the day is a weekday is calculated for the weekday / holiday target day by the following equation (4).

Here, m is the last data collection time obtained until the start of the weekday / holiday determination process on the data acquisition date, and l is the data collection start time on the data acquisition date. N indicates the number of position data collected during the time interval ml, and W (τi) indicates the weight for the probability density (τi, x, y) in the corresponding time interval τi at time t. Further, xt and yt indicate the latitude and longitude measured at time t.
That is, the average value of the probability density of each position data is the probability P (m) that the measurement date of the position data is a weekday.

ここで、しきい値Ｔhはシステム設計者やユーザが手動で設定してもよい。また、過去に蓄積された位置データをテストデータとして読み込んで、一定の精度を保つしきい値を計算により選択する自動しきい値設定処理を採用してもよい。 Next, the weekday / holiday determination processing unit 33 compares the probability P (m) with the threshold Th according to the following equation (5) in steps S92 to S94, so that the determination target day is a weekday or a holiday. Determine whether.

Here, the threshold value Th may be set manually by the system designer or user. Alternatively, automatic threshold setting processing may be employed in which position data accumulated in the past is read as test data and a threshold value that maintains a certain accuracy is selected by calculation.

Finally, in step S10, the weekday / holiday determination processing unit 33 transmits the determination result obtained by the weekday / holiday determination process in step S9 to the user terminal UT or the application servers AP1 to APn.
Under the control of the weekday / holiday notification receiving unit 41, the user terminal UT or the application servers AP1 to APn receives and stores information representing the weekday / holiday determination result transmitted from the weekday / holiday determination processing device PS. . Information representing the determination result of the weekday / holiday is used as attribute information for determining the user's situation. For example, information on work is distributed to the user on weekdays, and information on leisure is distributed on holidays.

  As described in detail above, in this embodiment, in the weekday / holiday determination processing device PS, the map information in which the position data is mapped for each time interval is read from the sensor data storage unit 21 of the database server DSV, and this map information is read. A mesh is divided into a plurality of areas, and a distribution of the user's stay probability density for each divided area for each time interval is calculated based on the map information divided into the areas. Also, based on the information representing the calculated probability density distribution, a weight for the probability density in each divided area is calculated for each time interval. Then, for the position data collected on the weekday / holiday determination target day, it is determined whether it is a weekday or a holiday using the calculated probability density distribution and its weight.

That is, a weight that contributes to weekday / holiday determination is calculated for each time zone from the distribution density and spread of past position data, and the weekday / holiday determination is performed in consideration of the calculated weight.
Therefore, in consideration of the distribution density and spread of the past position data, the weight contributing to the weekday / holiday determination is automatically changed for each time interval, and the quality of the position data varies depending on the time zone. It is possible to determine weekdays / holidays in consideration of the difference. As a result, for example, the user's location data collected in the past is not dispersed in various regions and the user can be considered to have a high probability of staying in a specific place, and the user's location data collected in the past varies. Even if the user stays in a place that is not in the stay history, it is possible to accurately determine the weekday / holiday by distinguishing it from the time period that would otherwise not affect the judgment process. It becomes.

  The present invention is not limited to the above embodiment. For example, in the above embodiment, the database server DSV and the weekday / holiday determination processing device PS are provided separately, but they may be provided in one server device. In the above embodiment, the user designates and inputs weekday / holiday annotation information. However, when weekdays / holidays are intended for users according to a calendar, input of weekday / holiday annotation information is omitted. Also good.

  In addition, the types and functions of the user terminal, the database server, and the weekday / holiday determination processing device, the processing procedure and contents for inputting weekday / holiday annotation information, the probability density distribution and the processing procedure and processing of the phase for calculating the weight thereof The contents, the weekday / holiday determination phase processing procedure, the processing contents, and the like can be variously modified and implemented without departing from the gist of the present invention.

  In short, the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. Further, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

  MS ... portable terminal, DSV ... database server, PS ... weekday / holiday determination processing device, UT ... user terminal, AP1 to APn ... application server, 11 ... sensor data measuring unit, 12 ... holiday ratio input unit, 21 ... sensor data storage , 22 ... holiday ratio storage unit, 23 ... feature vector storage unit, 24 ... weekday / holiday identification function storage unit, 31 ... feature vector calculation processing unit, 32 ... weekday / holiday discrimination function learning processing unit, 33 ... weekday / Holiday determination processing unit, 41 ... Weekday / holiday notification receiving unit.

Claims (7)

A situation determination method performed by a situation determination apparatus including sensor data storage means, probability density calculation processing means, weight calculation processing means, and weekday / holiday determination processing means,
The sensor data storage means collects the user's stay position data and stores it in the position data storage means ;
The probability density calculation processing unit divides one day into a plurality of time sections, divides map information including the stay range of the user into a plurality of areas, and stores the stay position data stored in the position data storage unit . A probability density storage means for calculating a probability density distribution representing a user's stay status with respect to each divided area of the map information for each of the divided time intervals based on the stay position data of the day annotated as a weekday The probability density distribution calculation process stored in
A weight calculation process in which the weight calculation processing means calculates a weight for each of the divided time intervals for the calculated probability density distribution , and stores the weight in a weight storage means ;
The weekday / holiday determination processing means, the probability density at time intervals and the divided areas include staying position data stored in the position data storage means collected on the day to be determined user and their weights, the probability density A search is made based on the probability density distribution and the weight stored in the storage means and the weight storage means, respectively, and using the searched probability density and the weight, the probability that the day to be judged is a weekday or a holiday is obtained. The process of calculating,
A step of determining whether the day to be determined is a weekday or a holiday based on the calculated probability , the weekday / holiday determination processing means ;
Equipped with,
The weight calculation process includes:
For each of the divided time sections, the average of the probability density of each divided area in the time section is divided by the number of divided areas in which the stay position data has been previously mapped to the time section. Calculating weights for probability density;
Normalizing a weight calculated for each of the divided time intervals based on the maximum value, and setting the normalized frequency distribution as a weight for the probability density for each time interval;
Situation determining method comprising Rukoto equipped with. The probability density distribution calculation process includes:
For each of the divided time intervals, mapping the user's stay location data collected in the past same time interval and stored in the location data storage means to map information;
Dividing the map information on which the stay position data is mapped into a plurality of areas in a mesh shape, and creating a frequency distribution of the stay position data for each of the divided areas;
And normalizing the created frequency distribution based on the maximum value, and making the normalized frequency distribution a probability density in the divided area for each of the divided time sections. The situation determination method according to claim 1. The process of calculating the probability that the date to be determined is a weekday or a holiday,
Mapping the user's stay location data collected on the date to be determined and stored in the location data storage means to the map information;
Based on the map information to which the stay position data is mapped, a process of searching the probability density in the time interval and the divided area including the stay position data from the probability density distribution stored in the probability density storage means ;
Based on the map information to which the stay position data is mapped, a process of searching the weight for the probability density in the time interval and the divided area including the stay position data from the weight stored in the weight storage means ;
Using the searched probability density and its weight, the average of the probability density of the position data collected on the date to be determined is calculated, and the average value of the calculated probability density is calculated as the date to be determined. The situation determination method according to claim 1, further comprising: a step of setting a probability that is a weekday or a holiday. Sensor data storage means for collecting user location data and storing it in the position data storage means ;
A day that is divided into a plurality of time intervals, map information that includes the stay range of the user is divided into a plurality of areas, and a day that is annotated as a weekday in the stay position data stored in the position data storage unit A probability density distribution calculating means for calculating a probability density distribution representing a user's stay status in each divided area of the map information for each of the divided time intervals based on the stay position data , and storing the probability density distribution means in a probability density storage means When,
A weight calculation processing unit that calculates a weight for each of the divided time intervals with respect to the probability density distribution stored in the probability density storage unit, and stores the weight in the weight storage unit ;
The probability density and its weight in the time interval and the divided area including the user's stay position data collected on the date to be determined and stored in the position data storage means are stored in the probability density storage means and the weight storage means, respectively. A search based on the probability density distribution and the weight, and using the searched probability density and the weight, means for calculating a probability that the day to be determined is a weekday or a holiday;
Means for determining whether the day to be determined is a weekday or a holiday based on the calculated probability ,
The weight calculation processing means includes
For each of the divided time sections, the average of the probability density of each divided area in the time section is divided by the number of divided areas in which the stay position data has been previously mapped to the time section. Means for calculating weights for probability density;
Means for normalizing the weight calculated for each of the divided time intervals based on the maximum value, and setting the normalized frequency distribution as a weight for the probability density for each time interval;
Situation determining apparatus according to claim Rukoto equipped with. The probability density distribution calculating means includes:
Means for mapping the user's stay location data collected in the past same time interval and stored in the location data storage means to map information for each of the divided time intervals;
Means for dividing the map information to which the stay position data is mapped into a plurality of areas in a mesh shape, and creating a frequency distribution of the stay position data for each of the divided areas;
And means for normalizing the created frequency distribution based on the maximum value, and setting the normalized frequency distribution as a probability density in the divided area for each of the divided time sections. The situation determination apparatus according to claim 4 . Means for calculating the probability that the determination target day is a weekday or a holiday,
It means for mapping the stay position data of users stored collected on the day to be determined in said positional data storage unit to the map information,
Based on the map information to which the stay position data is mapped, a means for searching the probability density in the time interval and the divided area including the stay position data from the probability density distribution stored in the probability density storage means ;
Based on the map information to which the stay position data is mapped, a means for searching the weight for the probability density in the time interval and the divided area including the stay position data from the weight stored in the weight storage means ;
Using the searched probability density and its weight, the average of the probability density of the position data collected on the date to be determined is calculated, and the average value of the calculated probability density is calculated as the date to be determined. 6. The situation determination apparatus according to claim 4, further comprising a means for setting a probability that is a weekday or a holiday. The situation determination program which makes the computer with which the said condition determination apparatus performs the process which implement | achieves the means with which the condition determination apparatus in any one of Claim 4 thru | or 6 is provided.

Images