JP2012226390A - Evaluation prediction system and evaluation prediction method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an evaluation prediction system capable of extracting a correlation model capable of accurately evaluating and predicting a prescribed event on the basis of data transmitted from a mobile unit.SOLUTION: In an RNC 300 of an evaluation prediction system 10, a position information communication part 301 specifies position information of a mobile unit 100, and transmits the specified position information to an information analyzer 600. In the information analyzer 600, a position information reception part 601 receives the position information transmitted from the position information communication part 301. A dynamic state data calculation part 602 calculates dynamic state data indicating a dynamic state of the mobile unit 100 on the basis of the received position information. A correlation model extraction part 604 extracts a correlation model of the dynamic state data and external data indicating a prescribed event. An evaluation prediction part 605 evaluates or predicts the external data on the basis of the extracted correlation model and the dynamic state data. Then, an output part 606 outputs an evaluation result or a prediction result by the evaluation prediction part 605.

Description

  The present invention relates to an evaluation prediction system and an evaluation prediction method for extracting a correlation model capable of evaluating or predicting external data and dynamic data representing a predetermined event.

  Conventionally, for example, Patent Document 1 discloses a technique for performing statistical processing based on information acquired from a mobile device and performing predetermined numerical prediction. Here, information on population distribution is collected based on location registration information transmitted from mobile terminals (mobile devices), or data on future population distribution (short-term future prediction data) based on collected population distribution information Is calculated.

JP 2002-342557 A

  However, with the technology described in Patent Document 1, it is only possible to predict current and past population distributions and future population distributions, and these population distributions and predetermined events such as store sales, for example. The relationship with is not considered. Therefore, it has been impossible to accurately predict or evaluate a future predetermined event based on data transmitted from the mobile device.

  In view of the above problems, the present invention provides an evaluation prediction system and an evaluation prediction method capable of extracting a correlation model that can accurately evaluate and predict a predetermined event based on data transmitted from a mobile device. The purpose is to do.

  In order to solve the above problems, an evaluation prediction system of the present invention includes a wireless network control device that performs wireless connection control of a mobile device, and an information analysis device that is capable of transmitting and receiving information to and from the wireless network control device. The radio network controller includes a location information communication unit that identifies location information of the mobile device and transmits the identified location information to the information analysis device, The information analysis device includes: a position information receiving unit that receives the position information transmitted from the position information communication unit; and a dynamic state of the mobile device based on the position information of the mobile device received by the position information receiving unit. A kinetic data calculation unit for calculating kinetic data indicating a correlation, and a correlation model between the kinetic data calculated by the kinetic data calculation unit and external data representing a predetermined event It includes a correlation model extraction unit for extracting, and an output unit for outputting the correlation model extracted by the correlation model extracting portion.

  In addition, the evaluation prediction method of the present invention includes an evaluation including a wireless network control device that performs wireless connection control of a mobile device and an information analysis device that is capable of transmitting and receiving information to and from the wireless network control device. A prediction information method executed by a prediction system, wherein the wireless network control device specifies location information of the mobile device and transmits the specified location information to the information analysis device; The information analysis device receives the location information transmitted from the wireless network control device, and the information analysis device receives the location information of the mobile device based on the received location information of the mobile device. A kinetic data calculating step for calculating kinetic data indicating the kinetics of the information, and the information analysis device includes the calculated kinetic data and an external data representing a predetermined event. A correlation model extracting a correlation model with the data, the information analyzer includes an output step of outputting the extracted the correlation model, the.

  According to this invention, the information analysis device calculates the dynamic data based on the location information of the mobile device transmitted from the wireless network control device, and calculates a correlation model between the calculated dynamic data and external data representing a predetermined event. Extract. In this way, it is possible to extract a correlation model indicating the correlation between dynamic data and external data.

  The evaluation prediction system of the present invention is an evaluation prediction system including a mobile device and an information analysis apparatus capable of communicating with the mobile device, and the mobile device includes the mobile device. A position information communication unit that acquires the position information of the position information and transmits the acquired position information to the information analysis device, wherein the information analysis device receives the position information transmitted from the position information communication unit. Calculated by the information receiving unit, the dynamic data calculating unit calculating dynamic data indicating the dynamics of the mobile device based on the positional information of the mobile device received by the positional information receiving unit, and the dynamic data calculating unit. A correlation model extraction unit for extracting a correlation model between the kinetic data and external data representing a predetermined event, and an output unit for outputting the correlation model extracted by the correlation model extraction unit. Eteiru.

  The evaluation prediction method of the present invention is an evaluation prediction method executed by an evaluation prediction system including a mobile device and an information analysis device capable of communicating with the mobile device. A position information communication step in which the mobile device acquires position information of the mobile device, and transmits the acquired position information to the information analysis device; and the position at which the information analysis device is transmitted from the mobile device. A positional information receiving step for receiving information, a dynamic data calculating step for calculating dynamic data indicating dynamics of the mobile device based on the received positional information of the mobile device, and the information analysis A correlation model extraction step in which a device extracts a correlation model between the calculated dynamic data and external data representing a predetermined event; and the information analysis device outputs the extracted correlation model. And it includes an output section for, a.

  According to this invention, the information analysis device calculates dynamic data based on the location information of the mobile device directly transmitted from the mobile device, and extracts a correlation model between the calculated dynamic data and external data representing a predetermined event To do. In this way, it is possible to extract a correlation model indicating the correlation between dynamic data and external data.

  Further, in the evaluation prediction system according to the present invention, the dynamic data calculation unit, based on the location information of the mobile device, in addition to the user of the mobile device, the total population including people other than the user of the mobile device It is preferable to calculate kinetic data indicating the kinetics of.

  According to the present invention, it is possible to extract a correlation model indicating a correlation between dynamic data of a total population including people other than a user of a mobile device and external data.

  The evaluation prediction system according to the present invention further includes an evaluation prediction unit that evaluates or predicts the external data based on the correlation model extracted by the correlation model extraction unit and the dynamic data. The unit preferably outputs an evaluation result or a prediction result by the evaluation prediction unit.

  According to this invention, the external data is evaluated or predicted using the extracted correlation model and dynamic data, and the evaluation result or the prediction result is output. As described above, it is possible to accurately evaluate and predict the external data representing a predetermined event based on the correlation model and the dynamic data, and output the evaluation result or the prediction result of the external data.

  Further, in the evaluation prediction system according to the present invention, the information analysis device includes a high-impact data extraction unit that extracts kinetic data based on the degree of influence on the external data from the kinetic data calculated by the kinetic data calculation unit. Further, it is preferable that the correlation model extraction unit extracts a correlation model between the dynamic data extracted by the high influence data extraction unit and the external data.

  According to the present invention, a correlation model based on the degree of influence on external data to be evaluated or predicted is extracted, and external data can be more accurately evaluated and predicted using this correlation model and dynamic data. it can.

  In the evaluation prediction system according to the present invention, the correlation model extraction unit extracts the correlation model using a multiple regression function, the dynamic data is used as an explanatory variable of the multiple regression function, and the external data is It is preferable to use the objective variable of the multiple regression function.

  According to the present invention, it is possible to accurately evaluate and predict external data using a multiple regression function.

  The evaluation prediction system according to the present invention further includes an evaluation prediction unit that evaluates or predicts the dynamic data based on the correlation model extracted by the correlation model extraction unit and the external data, and the output unit Preferably outputs an evaluation result or a prediction result by the evaluation prediction unit.

  According to this invention, the dynamic data is evaluated or predicted using the extracted correlation model and external data, and the evaluation result or the prediction result is output. In this way, it is possible to accurately evaluate and predict the dynamic data based on the correlation model and the external data, and output the evaluation result or the prediction result of the dynamic data.

  In the evaluation prediction system according to the present invention, the information analysis device further includes a high-impact data extraction unit that extracts external data from the external data based on the degree of influence on the dynamic data, and the correlation model extraction unit Preferably, a correlation model between the external data extracted by the high influence data extraction unit and the dynamic data is extracted.

  According to the present invention, a correlation model based on the degree of influence on dynamic data to be evaluated or predicted is extracted, and dynamic data can be more accurately evaluated and predicted using this correlation model and external data. it can.

  In the evaluation prediction system according to the present invention, the correlation model extraction unit extracts the correlation model using a multiple regression function, the external data is used as an explanatory variable of the multiple regression function, and the dynamic data is It is preferable to use the objective variable of the multiple regression function.

  According to this invention, dynamic data can be accurately evaluated and predicted using a multiple regression function.

  According to the present invention, a correlation that can accurately evaluate or predict a predetermined event based on dynamic data calculated based on data transmitted from a mobile device and external data representing the predetermined event. A model can be extracted.

It is a system configuration figure of an evaluation prediction system of this embodiment. It is a functional block block diagram of an evaluation prediction system. It is a figure which shows the data flow at the time of external data evaluation and prediction. It is a figure which shows the example of dynamic data. It is a figure which shows a multiple regression function. It is a flowchart which shows the processing content performed with an evaluation prediction system. It is a figure which shows dynamic data and external data.

Embodiments of the present invention will be described with reference to the accompanying drawings. Where possible, the same parts are denoted by the same reference numerals, and redundant description is omitted.
[Configuration of evaluation prediction system]
FIG. 1 is a system configuration diagram of an evaluation prediction system 10 of the present embodiment. As shown in FIG. 1, the evaluation prediction system 10 includes a mobile device 100, a BTS (base station) 200, an RNC (radio network control device) 300, an exchange 400, and a management center 500. The management center 500 includes a social sensor unit 501, a petamining unit 502, a mobile demography unit 503, and a visualization solution unit 504.

  The exchange 400 collects the location information of the mobile device 100 via the BTS 200 and the RNC 300. The RNC 300 can measure the position of the mobile device 100 using the delay value in the RRC connection request signal when communication connection is established with the mobile device 100. The exchange 400 can receive the position information of the mobile device 100 measured in this way when the mobile device 100 performs communication connection. The exchange 400 stores the received position information, and outputs the collected position information to the management center 500 at a predetermined timing or in response to a request from the management center 500. Here, in general, the RNC 300 is composed of about 1,000 pieces and is arranged throughout Japan. On the other hand, about 300 exchanges 400 are arranged in Japan.

  As described above, the management center 500 includes the social sensor unit 501, the petamining unit 502, the mobile demography unit 503, and the visualization solution unit 504. Each unit uses the position information of the mobile device 100. Perform statistical processing.

  The social sensor unit 501 is a server device that collects data including location information of the mobile device 100 from each exchange 400. The social sensor unit 501 is configured to receive data periodically output from the exchange 400 or to acquire data from the exchange 400 according to a predetermined timing in the social sensor unit 501.

  The petamining unit 502 is a server device that converts data received from the social sensor unit 501 into a predetermined data format. For example, the petamining unit 502 performs a sorting process using a user ID as a key, or performs a sorting process for each area.

  The mobile demography unit 503 is a server device that performs aggregation processing on the data processed in the petamining unit 502, that is, count processing for each item. For example, the mobile demography unit 503 can count the number of users located in a certain area, and can total the distribution of the located areas.

  The visualization solution unit 504 is a server device that processes the data aggregated in the mobile demography unit 503 so as to be visible. For example, the visualization solution unit 504 can map the aggregated data on a map. Data processed by the visualization solution unit 504 is provided to companies, government offices or individuals, and is used for store development, road traffic surveys, disaster countermeasures, environmental countermeasures, and the like. It should be noted that the information statistically processed in this way is processed so that individuals are not specified so as not to infringe privacy.

  The social sensor unit 501, the petamining unit 502, the mobile demography unit 503, and the visualization solution unit 504 are all configured by the server device as described above, and although not shown, the basic configuration of a normal information processing device Needless to say, it includes a CPU, a RAM, a ROM, an input device such as a keyboard and a mouse, a communication device that communicates with the outside, a storage device that stores information, and an output device such as a display and a printer.

[Function block configuration of evaluation prediction system]
Next, a functional block configuration of the evaluation prediction system according to the present embodiment will be described. FIG. 2 shows a functional block configuration of the evaluation prediction system 10, and FIG. 3 shows a data flow at the time of external data evaluation / prediction. As shown in FIG. 2, the evaluation prediction system 10 includes, as elements according to the present invention, a plurality of mobile stations 100 located in sectors controlled by a plurality of BTSs (base stations) 200, an RNC 300, and information analysis. And an apparatus 600. The information analysis apparatus 600 corresponds to the mobile demography unit 503 and the visualization solution unit 504 of FIG. Since the exchange 400, the social sensor unit 501 and the petamining unit 502 in FIG. 1 are devices not directly related to the present invention, these notations are omitted in FIG.

  The RNC 300 includes a location information communication unit 301 that communicates with the transmission unit 101 of the mobile device 100, identifies the location of the mobile device 100, and transmits the identified location information to the information analysis device 600. The position of the mobile device 100 is obtained, for example, by a location registration signal transmitted when the mobile device 100 moves across the location registration area, or position registration in the location registration area every certain period (for example, 54 minutes). It is specified when the RNC 300 receives a location registration signal as a signal. Further, the location information of the mobile device 100 is not limited to being specified by the RNC 300. For example, when the mobile device 100 has a function capable of acquiring location information such as a GPS function, the location information acquired by the GPS function or the like. Information can also be transmitted directly to the information analysis device 600 by a position information communication unit for GPS function or the like provided in the mobile device 100.

  Based on the location information receiving unit 601 that receives the location information of the mobile device 100 and the location information of the mobile device 100 received by the location information receiving unit 601, the information analysis device 600 includes, in addition to the user of the mobile device 100, Among the dynamic data calculated by the dynamic data calculating unit 602, the dynamic data calculating unit 602 that calculates the dynamic data indicating the dynamics of the total population (for example, Japanese nationals) including people other than the user of the mobile device 100, A high-impact data extraction unit 603 that extracts dynamic data having a high degree of influence on external data to be evaluated or predicted by an evaluation prediction unit 605 described later. Also, the information analysis apparatus 600 includes a correlation model extraction unit 604 that extracts a correlation model based on the external data and the dynamic data extracted by the high influence data extraction unit 603, and external data using the extracted correlation model. An evaluation prediction unit 605 that performs evaluation or prediction of the above and an output unit 606 that outputs an evaluation result or a prediction result by the evaluation prediction unit 605. Needless to say, the “output” here includes display output and print output widely. That is, the evaluation result or the prediction result may be displayed and output on a display or the like, may be printed out from a printer or the like, or may be output both in display and printing. The information analysis apparatus 600 further includes an external data holding unit 607 that holds external data.

  The dynamic data calculation unit 602 uses the existing technology based on the location information of the mobile device 100 to change the dynamics of the total population including people other than the user of the mobile device in addition to the user of the mobile device. The kinetic data showing is calculated. As shown in FIG. 4, the dynamic data includes a population distribution, a distribution by attribute (a distribution by gender, a distribution by age, etc.), a flow rate of people, a behavior area of people, and the like.

  In addition, the external data holding unit 607 stores store information (for example, sales information, store visitor information, warehousing information), weather information (for example, hourly rainfall information), and taxi boarding information (for example, by time of day) as external data. Number of achievements) and data having geographical information such as traffic jam information. These external data are, for example, statistically processed by an external device (not shown) and the like, and are input to the external data holding unit 607 and represent current or past events.

  As shown in FIG. 3, the high influence data extraction unit 603 analyzes the dynamic data calculated by the dynamic data calculation unit 602 and external data similar to the external data to be evaluated or predicted by the evaluation prediction unit 605. As a result, the kinetic data having a high degree of influence (correlation coefficient) on the external data to be evaluated or predicted by the evaluation prediction unit 605 is extracted. This extraction is performed based on an existing statistical processing method.

  The correlation model extraction unit 604 extracts a correlation model using a multiple regression function. The multiple regression function used in the correlation model extraction unit 604 and the evaluation prediction unit 605 described later is represented in FIG.

  As shown in FIG. 3, the correlation model extraction unit 604 uses the external data held in the external data holding unit 607 as the objective variable Y, and uses the dynamic data extracted by the high influence data extraction unit 603 as the explanatory variable X as a correlation model. To extract. The correlation model extraction unit 604 calculates the variable b as a correlation model by increasing or decreasing the number of explanatory variables X according to the number of types of dynamic data extracted by the high influence data extraction unit 603.

  Here, as an example of the correlation between the external data and the dynamic data, for example, the correlation between the store information as the external data and the dynamic data is that the area where there are many people has a lot of sales (number of customers etc.) There are many sales when there are many people in the area, and there are many sales at stores where the population is changing rapidly (when the flow of people is large or the fluctuation is large). The correlation model extraction unit 604 extracts a correlation model representing the correlation between such external data and dynamic data. As another example of the correlation, for example, the correlation between the weather information (rainfall information by hour) as the external data and the dynamic data indicates that the raining area has few people and it is raining. Adjacent areas have more people than normal, areas with fewer people are more likely to rain. In addition, as another example of the correlation, for example, the correlation between taxi boarding information (actual number of rides by time zone) as external data and dynamic data is a place where there are few passengers and many people There are many passengers, and there are many passengers in areas where the population is changing rapidly (such as when the flow of people is large or the fluctuation is large).

  The evaluation predicting unit 605 performs the current evaluation of the external data or the future prediction using the multiple regression function. Specifically, as shown in FIG. 3, the dynamic data extracted by the high influence data extraction unit 603 is used as the explanatory variable X of the multiple regression function, and this dynamic data and the correlation model extracted by the correlation model extraction unit 604 are used. Is used to evaluate the current state of external data or to predict the future. For example, in the case of evaluation of external data, expected external data is calculated by introducing current or past dynamic data related to the position of the external data to be evaluated into explanatory variables, and the value and actual external data are calculated. Can be compared and evaluated. For example, in the case of prediction of external data, expected external data can be calculated and predicted by introducing future dynamic data related to the position of the external data to be predicted into an explanatory variable. Here, prediction of future dynamic data can be performed by, for example, estimating future dynamic data from the correlation between the weather information and the dynamic data using the weather information at the time of prediction from the weather forecast or the like.

[Processes executed by the evaluation prediction system]
Next, a series of processes executed by the evaluation prediction system 10 will be described with reference to FIG. Here, a case where sales (external data) of a newly installed store X is predicted using dynamic data will be described as an example.

  Transmitting section 101 of mobile device 100 transmits a location registration signal to RNC 300 (step S1 in FIG. 6). This operation is performed in each of a large number of mobile devices 100. As a result, a large number of location registration signals reach the RNC 300 via the BTS 200.

  In the RNC 300, the location information communication unit 301 receives a large number of location registration signals (step S2), identifies the location of the mobile device 100 using the received location registration signal as a trigger (step S3), and analyzes the identified location information. It transmits to the apparatus 600 (step S4). The transmission in step S4 may be executed periodically at a predetermined time interval, may be executed at a preset time, or may be executed using a transmission request from the information analysis apparatus 600 as a trigger. Good.

  In the information analysis device 600, when the position information receiving unit 601 receives the position information transmitted from the RNC 300 (step S5), the dynamic data calculating unit 602 calculates the dynamic data of the total population using the received positional information. It performs (step S6). Here, the dynamic data calculation unit 602 includes basic dynamic data including dynamic population data including a population in a predetermined area, a population difference in the predetermined area, a flow of people in the predetermined area, and predetermined area information. calculate. This basic dynamic data is specifically represented as shown in FIG. Further, the dynamic data calculation unit 602 calculates basic dynamic data at predetermined time intervals for each predetermined area.

  Moreover, the dynamic data calculation unit 602 calculates the peripheral dynamic data and the dynamic data of the preceding and following time zones using the basic dynamic data. Specifically, as illustrated in FIG. 7A, when the basic dynamic data is calculated for the area name “Hikari no Oka”, the dynamic data calculation unit 602 includes the areas A and B around the “Hikari no Oka”. The basic dynamic data for... Is calculated as the peripheral dynamic data for the area name “Hikari no Oka”. Also, as shown in FIG. 7A, when the basic dynamic data is calculated for the area name “Hikari no Oka”, the dynamic data calculation unit 602 is adjacent to the area name “Hikari no Oka” in time series. The three basic dynamic data are calculated as dynamic data of the time zone before and after the area name “Hikari no Oka”. In addition, when calculating the surrounding dynamic data, the basic dynamic data of the area adjacent to the predetermined area is used, but in addition to this, depending on the characteristics of the external data to be evaluated / predicted, for example, In the case of a target, a range that can be moved on foot (for example, 4 km for 1 hour) is extracted as a peripheral area, and the peripheral dynamic data can be calculated based on the basic dynamic data in the extracted peripheral area. . Similarly, the time zone may be calculated in consideration of the characteristics of the external data, in addition to calculating the two basic dynamic data adjacent in time series as the dynamic data of the preceding and following time zones.

  Next, the high influence data extraction unit 603 selects a parameter having a high degree of influence on the external data to be evaluated or predicted in step S9 described later, from among the parameters of the various movement data calculated by the movement data calculation unit 602. Extract (step S7). Specifically, for example, when trying to evaluate and predict sales of a store, parameters included in basic dynamic data and parameters included in peripheral dynamic data as parameters of dynamic data highly correlated with the sales of the store And extract. A multiple regression function can also be used when extracting a parameter that has a high influence on external data. For example, when evaluating / predicting store sales, it is possible to extract parameters having a high influence on store sales using store information and a multiple regression function shown in FIG. In addition, a parameter having a high influence level can be extracted based on an empirical rule or the like.

  Next, the correlation model extraction unit 604 extracts a correlation model using a multiple regression function between the parameters of the dynamic data extracted by the high influence data extraction unit 603 and the sales of the store (step S8). Here, in the external data holding unit 607, as shown in FIG. 7B, information on the sales of the store A, the store B, the store C, and the installation area of each store is stored in the store information A, the store information. B, store information C... Specifically, as shown in FIG. 7C, the correlation model extraction unit 604 uses the store information held in the external data holding unit 607 as the objective variable of the multiple regression function and extracts it in the high influence data extraction unit 603. The correlation model is extracted using the obtained dynamic data (basic dynamic data and peripheral dynamic data) as explanatory variables of the multiple regression function. The extraction of the correlation model using this multiple regression function is performed based on the existing statistical processing method.

  The evaluation predicting unit 605 predicts sales at the store X using the correlation model extracted by the correlation model extracting unit 604 (step S9). Specifically, as illustrated in FIG. 7D, when the installation area of the store X is the Miura coast, the correlation model extraction unit 604 performs basic motion data regarding the Miura coast and peripheral motion data regarding the surrounding area of the Miura coast. Are used as explanatory variables of the multiple regression function, and the sales of the store information X, which is an objective variable, is calculated based on the explanatory variables and the correlation model.

  The output unit 606 outputs the calculation result of the evaluation prediction unit 605 (step S10).

  Thus, by extracting the correlation model from the external data such as the store information A and the dynamic data, the extracted correlation model and the dynamic data of the total population calculated based on the mobile device 100 position information are used. It is possible to evaluate or predict external data such as store sales.

  In this embodiment, external data is evaluated or predicted using a correlation model. However, as a modification, dynamic data can be evaluated or predicted using a correlation model. In this case, the high influence data extraction unit 603 extracts external data having a high degree of influence on the dynamic data to be evaluated or predicted from the external data held in the external data holding unit 607. The correlation model extraction unit 604 extracts a correlation model using dynamic data as an objective variable and external data extracted by the high influence data extraction unit 603 as an explanatory variable. The evaluation prediction unit 605 uses the external data extracted by the high influence data extraction unit 603 as an explanatory variable of the multiple regression function, and uses this external data and the correlation model extracted by the correlation model extraction unit 604 to Make an assessment or prediction. In this way, dynamic data can be evaluated or predicted using external data and a correlation model.

  In addition to the above examples, external data includes information such as various statistical information (national census information, person trip information, etc.), disaster information, shelter location information, stores / automatic sales, etc. The machine installation position information, the fixed advertisement position information, the information about the fixed advertisement position and its effect, the information about the event position and the number of event visitors, the position information of the base station of the mobile device 100, and the like can be used. External data includes taxi probe information (passenger information, sales information, etc.), mobile advertisement location information and information on its effect, mobile store location information and sales information, and car rental Information (information on vehicle allocation and load capacity), position information of the mobile base station of the mobile device 100, and the like can be used.

  In addition, even if the data granularity is different, such as when external data is pinpoint information (information in units of several meters) and dynamic data is macro information (information in units of several hundreds of meters), the correlation Models can be calculated and external or dynamic data can be evaluated and predicted. For example, in the case of a taxi ride, the dynamic data is macro information, while the micro information indicates which side of the road the taxi can be picked up. Even in this case, it is possible to extract a correlation model capable of evaluating and predicting micro information by using macro information (population, population change, number of surrounding areas, etc.). Also, for example, in the case of traffic congestion information of VICS, by extracting a correlation model using traffic congestion information (micro information) and dynamic data (macro information) for each road, only road information can be obtained from macro information (dynamic data). It is also possible to predict traffic jam information (micro information).

  The dynamic data used in the correlation analysis between external data and dynamic data uses uniform dynamic data that does not depend on the position information included in the external data (for example, the same dynamic data is used regardless of the position of the external data). Or use dynamic data based on the position information included in the external data (for example, re-aggregate and use the dynamic data according to the position of the external data) Is calculated))).

  In the above-described embodiment, an example in which a multiple regression function is used when extracting a correlation model has been described. However, the function for extracting a correlation model is not limited to the multiple regression function, and other functions are used. You can also.

[Action and effect]
Next, the operation and effect of the evaluation prediction system 10 of this embodiment will be described.

  The RNC 300 of the evaluation prediction system 10 according to the present embodiment includes a position information communication unit 301 that specifies the position information of the mobile device 100 and transmits the specified position information to the information analysis apparatus 600. Based on the location information receiving unit 601 that receives the location information transmitted from the information communication unit 301 and the location information of the mobile device 100 received by the location information receiving unit 601, dynamic data indicating the dynamics of the mobile device 100 is calculated. To be extracted by a correlation model extraction unit 604, a correlation model extraction unit 604 that extracts a correlation model between the dynamic data calculated by the dynamic data calculation unit 602 and external data representing a predetermined event. And an output unit 606 for outputting the correlation model. Therefore, a correlation model can be extracted based on dynamic data and external data.

  In addition, the mobile device 100 of the evaluation prediction system 10 according to the present embodiment includes a position information communication unit that acquires the position information of the mobile device 100 and transmits the acquired position information to the information analysis device 600. Reference numeral 600 denotes a position information receiving unit 601 that receives position information transmitted from the position information communication unit, and a dynamic state indicating the dynamics of the mobile device 100 based on the position information of the mobile device 100 received by the position information receiving unit 601. Dynamic data calculation unit 602 that calculates data, correlation model extraction unit 604 that extracts a correlation model between the dynamic data calculated by the dynamic data calculation unit 602 and external data representing a predetermined event, and a correlation model extraction unit 604 And an output unit 606 for outputting the correlation model extracted by. Therefore, a correlation model can be extracted based on dynamic data and external data.

  The dynamic data calculation unit 602 calculates dynamic data indicating the dynamics of the total population including not only the user of the mobile device 100 but also people other than the user of the mobile device 100 based on the location information of the mobile device 100. To do. Thereby, it is possible to extract a correlation model indicating the correlation between the dynamic data of the total population including people other than the user of the mobile device 100 and the external data.

  In addition, an evaluation prediction unit 605 that evaluates or predicts external data based on the correlation model extracted by the correlation model extraction unit 604 and the dynamic data is further provided, and the output unit 606 is an evaluation result of the evaluation prediction unit 605. Alternatively, the prediction result is output. Accordingly, it is possible to accurately evaluate and predict external data representing a predetermined event based on the correlation model and the dynamic data, and output the external data evaluation result or prediction result.

  The information analysis apparatus 600 further includes a high influence data extraction unit 603 that extracts movement data based on the degree of influence on external data from the movement data calculated by the movement data calculation unit 602. The correlation model extraction unit 604 includes: A correlation model between the dynamic data extracted by the high influence data extraction unit 603 and the external data is extracted. Thereby, a correlation model based on the degree of influence on external data to be evaluated or predicted is extracted, and external data can be evaluated and predicted more accurately.

  The correlation model extraction unit 604 extracts a correlation model using a multiple regression function. By using dynamic data as explanatory variables for the multiple regression function and external data as an objective variable for the multiple regression function, multiple correlation models are extracted. External data can be accurately evaluated and predicted using a regression function.

  In addition, an evaluation prediction unit 605 that evaluates or predicts dynamic data based on the correlation model extracted by the correlation model extraction unit 604 and external data is further provided, and the output unit 606 outputs an evaluation result by the evaluation prediction unit 605 or By outputting the prediction result, it is possible to accurately evaluate and predict the dynamic data based on the correlation model and the external data, and output the evaluation result or the prediction result of the dynamic data.

  The information analysis apparatus 600 further includes a high-impact data extraction unit 603 that extracts external data from external data based on the degree of influence on dynamic data, and the correlation model extraction unit 604 is operated by the high-impact data extraction unit 603. A correlation model between the extracted external data and dynamic data is extracted. Thereby, a correlation model based on the degree of influence on dynamic data to be evaluated or predicted is extracted, and the dynamic data of the mobile device 100 can be evaluated and predicted more accurately.

  Further, the correlation model extraction unit 604 extracts a correlation model using a multiple regression function. By using external data as explanatory variables for the multiple regression function and dynamic data as an objective variable for the multiple regression function, multiple correlation models are extracted. It is possible to accurately evaluate and predict dynamic data using a regression function.

  DESCRIPTION OF SYMBOLS 10 ... Evaluation prediction system, 100 ... Mobile device, 300 ... RNC, 301 ... Location information communication part, 600 ... Information analysis apparatus, 601 ... Position information reception part, 602 ... Dynamic data calculation part, 603 ... High influence data extraction part, 604 ... correlation model extraction unit, 605 ... evaluation prediction unit, 606 ... output unit, 607 ... external data holding unit.

Claims (11)

An evaluation prediction system including a wireless network control device that performs wireless connection control of a mobile device and an information analysis device that is capable of transmitting and receiving information to and from the wireless network control device,
The wireless network control device includes a location information communication unit that identifies location information of the mobile device and transmits the identified location information to the information analysis device,
The information analyzer is
A position information receiving unit that receives the position information transmitted from the position information communication unit;
Based on the location information of the mobile device received by the location information reception unit, a dynamic data calculation unit that calculates dynamic data indicating the dynamics of the mobile device;
A correlation model extraction unit for extracting a correlation model between the dynamic data calculated by the dynamic data calculation unit and external data representing a predetermined event;
An output unit that outputs the correlation model extracted by the correlation model extraction unit;
An evaluation prediction system comprising: An evaluation prediction system configured to include a mobile device and an information analysis device capable of communicating with the mobile device,
The mobile device includes a location information communication unit that acquires location information of the mobile device and transmits the acquired location information to the information analysis device,
The information analyzer is
A position information receiving unit that receives the position information transmitted from the position information communication unit;
Based on the location information of the mobile device received by the location information reception unit, a dynamic data calculation unit that calculates dynamic data indicating the dynamics of the mobile device;
A correlation model extraction unit for extracting a correlation model between the dynamic data calculated by the dynamic data calculation unit and external data representing a predetermined event;
An output unit that outputs the correlation model extracted by the correlation model extraction unit;
An evaluation prediction system comprising:   The dynamic data calculation unit calculates dynamic data indicating dynamics of a total population including persons other than the user of the mobile device in addition to the user of the mobile device based on the location information of the mobile device. Item 3. The evaluation system according to Item 1 or 2. An evaluation prediction unit that evaluates or predicts the external data based on the correlation model extracted by the correlation model extraction unit and the dynamic data;
The evaluation prediction system according to claim 1, wherein the output unit outputs an evaluation result or a prediction result by the evaluation prediction unit. The information analysis device further includes a high-impact data extraction unit that extracts kinetic data based on the degree of influence on the external data from the kinetic data calculated by the kinetic data calculation unit,
The evaluation prediction system according to any one of claims 1 to 4, wherein the correlation model extraction unit extracts a correlation model between the dynamic data extracted by the high influence data extraction unit and the external data.   The correlation model extraction unit is configured to extract the correlation model using a multiple regression function, wherein the dynamic data is an explanatory variable of the multiple regression function, and the external data is an objective variable of the multiple regression function. The evaluation prediction system according to any one of Items 1 to 5. An evaluation prediction unit that evaluates or predicts the dynamic data based on the correlation model extracted by the correlation model extraction unit and the external data;
The evaluation prediction system according to claim 1, wherein the output unit outputs an evaluation result or a prediction result by the evaluation prediction unit. The information analysis device further includes a high-impact data extraction unit that extracts external data based on the degree of influence on the dynamic data from the external data,
The evaluation prediction system according to claim 1, wherein the correlation model extraction unit extracts a correlation model between the external data extracted by the high influence data extraction unit and the dynamic data.   The correlation model extraction unit is configured to extract the correlation model using a multiple regression function, wherein the external data is an explanatory variable of the multiple regression function, and the dynamic data is an objective variable of the multiple regression function. Item 9. The evaluation prediction system according to Item 1, 2, 3, 7, or 8. Evaluation prediction executed by an evaluation prediction system including a wireless network control device that performs wireless connection control of a mobile device and an information analysis device that is capable of transmitting and receiving information to and from the wireless network control device A method,
The wireless network control device specifies the location information of the mobile device, and transmits the specified location information to the information analysis device;
A location information receiving step in which the information analysis device receives the location information transmitted from the wireless network control device;
A dynamic data calculation step in which the information analyzer calculates dynamic data indicating the dynamics of the mobile device based on the received location information of the mobile device;
A correlation model extraction step in which the information analysis apparatus extracts a correlation model between the calculated dynamic data and external data representing a predetermined event;
An output step in which the information analyzer outputs the extracted correlation model;
An evaluation prediction method comprising: An evaluation prediction method executed by an evaluation prediction system configured to include a mobile device and an information analysis apparatus capable of communicating with the mobile device,
A location information communication step in which the mobile device acquires location information of the mobile device and transmits the acquired location information to the information analysis device;
A position information receiving step in which the information analyzer receives the position information transmitted from the mobile device;
A dynamic data calculation step in which the information analyzer calculates dynamic data indicating the dynamics of the mobile device based on the received location information of the mobile device;
A correlation model extraction step in which the information analysis apparatus extracts a correlation model between the calculated dynamic data and external data representing a predetermined event;
An output unit that outputs the extracted correlation model;
An evaluation prediction method comprising:

Images