JP2016066243A - Recommendation effect visualization system, method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a recommendation effect visualization system capable of achieving the promotion of the utilization of recommendation information.SOLUTION: Virtual target setting means 81 sets a virtual information provision target which acts in accordance with recommendation information. Action content calculation means 82 calculates the action content of the virtual information provision target on the basis of the recommendation information. Virtual result calculation means 83 calculates the action result of the virtual information provision target from the action of the information provision target in an actual environment. Cumulative result calculation means 84 calculates the cumulative result of the action results in a predetermined period.SELECTED DRAWING: Figure 5

Description

  The present invention relates to a recommendation effect visualization system, a recommendation effect visualization method, and a recommendation effect visualization program that can visualize the effect of recommended information.

  Various techniques for performing analysis based on learning data and generating various recommended information (hereinafter referred to as recommendation information) have been proposed. For example, Patent Document 1 describes an optimal taxi placement system that can efficiently acquire customers. In the system described in Patent Document 1, a taxi operator's business area is divided into a plurality of placement destination districts, and a demand forecast is calculated based on actual vehicle performance information for each dispatch destination district. Place the required number of empty vehicles in the district.

JP 2014-6890 A

  Assuming taxi dispatch as described above, for example, a place where a customer is likely to get on, such as a shortage of taxis, may be generated as recommendation information. However, although the recommendation information is useful information, it is difficult to know whether or not the vehicle allocation according to the recommendation information leads to sales, and may not be effectively utilized. Therefore, the driver often uses his intuition and experience, word-of-mouth information between drivers, and the like, rather than the objectively analyzed recommendation information described in Patent Document 1.

  However, the recommendation information is information generated based on various information other than the driver's intuition and experience. Therefore, in order to effectively use a large amount of accumulated information, a mechanism that promotes the use of recommendation information generated based on the information is necessary.

  Therefore, an object of the present invention is to provide a recommendation effect visualization system, a recommendation effect visualization method, and a recommendation effect visualization program that can promote the use of recommendation information.

  The recommendation effect visualization system according to the present invention includes a virtual object setting unit that sets a virtual information provision target that acts in accordance with recommendation information, and an action content calculation unit that calculates the action content of a virtual information provision target based on the recommendation information. , A virtual result calculation means for calculating a virtual performance result of the information provision target from a behavior of the information provision target in the real environment, and a cumulative result calculation means for calculating a cumulative result of the behavior result in a predetermined period, The means calculates the action content of the virtual information provision target based on the information based on the information provision target action in the real environment, and the cumulative result calculation means acts based on each recommendation information for a predetermined period. It is characterized in that the cumulative result of the action results to be provided is calculated.

  The recommendation effect visualization method according to the present invention sets a virtual information provision target to act according to recommendation information, and based on the recommendation information, based on information based on the behavior of the information provision target in the real environment, the behavior of the virtual information provision target Calculate the content, calculate the actual performance of the information provision target from the behavior of the information provision target in the real environment, and accumulate the results of the behavior of the virtual information provision target that acts based on each recommendation information in a predetermined period Is calculated.

  The recommendation effect visualization program according to the present invention is a virtual object setting process for setting a virtual information provision target that acts on a computer according to recommendation information, and an action content calculation that calculates the action content of a virtual information provision target based on the recommendation information. Process, a virtual result calculation process for calculating a virtual result of action of the information provision target from an action of the information provision target in the real environment, and a cumulative result calculation process for calculating a cumulative result of the action result in a predetermined period In the content calculation process, the action content of the virtual information provision target is calculated based on the information based on the behavior of the information provision target in the real environment, and the virtual result of the cumulative result calculation process is based on each recommendation information in the predetermined period. It is characterized in that the result of accumulating the action results of the information provision target is calculated.

  According to the present invention, it is possible to promote the use of recommendation information.

It is a block diagram which shows 1st Embodiment of the ranking presentation system by this invention. It is explanatory drawing which shows the relationship between a virtual taxi and a real taxi. It is a flowchart which shows the operation example of the ranking presentation system in 1st Embodiment. It is a flowchart which shows the operation example of the ranking presentation system in 2nd Embodiment. It is a block diagram which shows the outline | summary of the recommendation effect visualization system by this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. This embodiment demonstrates the example at the time of applying the recommendation effect visualization system of this invention to the system (ranking presentation system) which shows the ranking of a taxi. Since the taxi of this embodiment is provided with recommendation information from a ranking presentation system, it can be called an information provision object.

Embodiment 1. FIG.
FIG. 1 is a block diagram showing a first embodiment of a ranking presentation system. The ranking presentation system 10 of the present embodiment includes a data collection unit 11, a situation determination unit 12, a movement location determination unit 13, a navigation unit 14, a recommendation presentation unit 15, a cumulative data storage unit 16, and a virtual taxi calculation. Means 17, actual taxi calculating means 18, ranking calculating means 19, and ranking presenting means 20 are provided.

  The ranking presentation system 10 communicates with the taxi 30 via the wireless communication network 100 or the like. The taxi 30 periodically transmits its position information, business conditions (empty car, actual car (rental), forwarding, pick-up), and fares indicating sales if possible, to the ranking presentation system 10. Such information may be transmitted, for example, by a taxi-mounted device or by a portable terminal such as a tablet or a smartphone.

  The ranking presentation system 10 receives these pieces of information from the taxi 30, generates recommendation information, and transmits it to the taxi 30. In addition, since these data transmitted from the taxi 30 are used for the production | generation of the recommendation information mentioned later, they can be called analysis data or learning data. Further, the ranking presentation system 10 creates a ranking of the taxis 30 based on the analysis data, and transmits it to each taxi 30.

  The data collection unit 11 collects analysis data for generating recommendation information. For example, the data collection unit 11 may receive analysis data from the taxi 30 via the wireless communication network 100. The data collection means 11 may receive taxi sales from not only the taxi 30 but also from another system (not shown).

  The situation judgment unit 12 makes various judgments from the current situation. Specifically, the situation determination unit 12 performs taxi demand prediction from the current situation. The method of forecasting taxi demand is arbitrary. For example, the situation determination unit 12 may identify a point where a taxi is frequently used (a so-called hot spot) from the riding results of the taxi 30 collected by the data collection unit 11, and predict that the demand at the point is high. Good.

  Moreover, the information used for demand prediction is not limited to the information which shows the present condition. For example, the situation determination unit 12 may statistically make a current demand prediction based on information acquired retroactively for a certain period. In addition, when predicting information that changes from moment to moment, such as demand forecast for taxis, it is better to make various decisions based on information indicating the current situation as in this embodiment. This is preferable because it is possible.

  The travel location determination unit 13 generates recommendation information to be presented to the taxi 30 based on the demand prediction by the situation determination unit 12 and the current position of the taxi 30. The method for generating the recommendation information is arbitrary. The movement location determination means 13 may generate the position predicted to be more demanded as recommendation information. Moreover, the movement location determination means 13 may generate recommendation information according to the current position of the taxi 30 and the distance to the position where it is predicted that there is a demand. Further, the movement location determination means 13 may generate information including a plurality of positions as recommendation information.

  In the present embodiment, the position where demand is predicted may be a specific position such as a taxi stand, or may be an arbitrary position where a so-called sink taxi places a user. In addition, when there are a plurality of taxis 30 that transmit recommendation information, it is preferable that the moving location determination unit 13 generates recommendation information indicating different positions so that recommended positions do not overlap as much as possible.

  The navigation means 14 calculates a route to the position indicated by the recommendation information, based on the current position of the taxi 30 that transmits the recommendation information, the current traffic situation, and the like. The navigation means 14 may calculate the shortest route to that position. In addition, when the recommendation information includes a plurality of positions (that is, when there are a plurality of places where the user is likely to be placed), the navigation means 14 calculates a route appropriately passing through the positions. Also good. The route calculation method is arbitrary, and a generally known navigation system may be used.

  The recommendation presentation unit 15 transmits the recommendation information to the taxi 30. When the navigation means 14 is calculating the route, the recommendation presenting means 15 may transmit the recommendation information including information indicating the route.

  The accumulated data storage unit 16 collects and holds information indicating the operation status of the taxi 30 (hereinafter referred to as an operation log) as needed. The accumulated data storage unit 16 holds an operation log in accordance with an instruction from the data collection unit 11, for example. In the operation log, position information of the taxi 30 and information indicating the business state are accumulated for each time. In addition, the cumulative data storage unit 16 may associate and hold a fare that is generated each time a user gets on the vehicle.

  The virtual taxi calculating means 17 sets a virtual information provision target that acts according to the recommendation information. In this embodiment, the virtual taxi calculation means 17 determines the virtual taxi which acts according to recommendation information. Here, the virtual target means a target that acts completely according to the recommendation information when the recommendation information exists. The virtual taxi in the present embodiment means a taxi that moves in accordance with the recommendation information when the recommendation information is received in a state where no user is on the vehicle (for example, an empty vehicle state).

  In the present embodiment, it is assumed that a virtual taxi that is virtually moved according to the recommendation information performs the same action as a taxi that exists in the real environment at the corresponding destination (hereinafter referred to as a real taxi). That is, the action content of the virtual taxi is determined based on the action of the actual taxi.

  Therefore, in this embodiment, the virtual taxi calculation means 17 first selects a taxi to be handled as a virtual taxi from real taxis. Specifically, the virtual taxi calculation means 17 selects an arbitrary real taxi that is in a real vehicle state. Here, the actual vehicle means that a business vehicle such as a taxi carries passengers. The virtual taxi calculating means 17 may select, for example, a taxi that has entered the actual vehicle state earliest after the taxi opening hours have started.

  Thereafter, the virtual taxi calculating means 17 treats the movement of the virtual taxi as the movement of the real taxi until the selected real taxi drops the user (that is, the real car of the real taxi is completed). And the virtual taxi calculation means 17 calculates the action performance of a virtual taxi from the action of a real taxi.

  Specifically, the virtual taxi calculating means 17 calculates the actual performance of the virtual taxi from the movement status of the actual taxi. For example, the virtual taxi calculating means 17 may calculate the sales of the actual taxi as the sales of the virtual taxi. In addition, the virtual taxi calculating means 17 may calculate the distance and the travel distance for each actual vehicle in order to calculate the cumulative result of the actual vehicle rate and the travel distance.

  When the actual taxi notifies the sales for each actual vehicle, the virtual taxi calculation means 17 may use the notified sales as the sales of the virtual taxi. The virtual taxi calculation means 17 may calculate the actual taxi sales based on the actual taxi travel distance, travel time, initial fare, and the like, and the calculated sales may be used as the virtual taxi sales.

  When the actual taxi becomes empty (that is, when the actual taxi is completed), the virtual taxi calculating means 17 receives the recommendation information from the moving location determining means 13. Note that the virtual taxi calculating means 17 may request recommendation information from the moving location determining means 13 when the actual taxi becomes empty. At this time, the virtual taxi calculating unit 17 may receive route information from the navigation unit 14.

  The virtual taxi calculating means 17 specifies the destination of the virtual taxi and the travel time to the destination based on the received recommendation information and route information. Then, the virtual taxi calculating means 17 again selects an actual taxi that has been actually boarded at the destination after the specified travel time has elapsed as a virtual taxi.

  As described above, the virtual taxi calculating means 17 calculates the action content of the virtual taxi based on the new recommendation information every time the actual taxi completes the actual vehicle as the predetermined action. Then, the virtual taxi calculation means 17 detects the actual taxi that has entered the actual vehicle state at the corresponding position in the real environment as a result of causing the virtual taxi to act in the real environment according to the calculated action content. Is determined as a virtual taxi.

  The virtual taxi calculating means 17 calculates the actual result of the virtual taxi from the selected actual taxi movement status, and further calculates the cumulative result of the actual results within a predetermined period. For example, the virtual taxi calculating means 17 may calculate the cumulative result of sales since the business hours of the day started.

  The cumulative result of the results calculated by the virtual taxi calculating means 17 is not limited to the cumulative result of sales. The virtual taxi calculating means 17 may calculate the actual vehicle rate based on the taxi business conditions. In addition, the virtual taxi calculating means 17 may calculate the maximum value or the average value of the moving distance when the user is on the vehicle as the accumulated result.

  Thereafter, each time a real taxi vehicle is completed, the moving location determination means 13 generates recommendation information, and the virtual taxi calculation means 17 determines a virtual taxi based on the recommendation information, and calculates the results and cumulative results. , Repeated within a predetermined period (for example, business hours).

  FIG. 2 is an explanatory diagram showing the relationship between a virtual taxi and a real taxi. The virtual taxi moves to a position where the user can be placed according to the recommendation information, and uses the actual movement of the actual taxi as the actual movement. In addition, the taxi shown with the thick line in FIG. 2 shows the actual taxi selected as a virtual taxi, and the taxi shown with the broken line in FIG. 2 shows a virtual taxi.

  First, in the initial state, the virtual taxi calculating means 17 selects an actual taxi T1 on which the user has boarded from a plurality of actual taxis (see FIG. 2A). While the selected real taxi T1 carries the user, the virtual taxi V follows the movement of the real taxi T1. That is, the movement of the actual taxi T1 is regarded as the movement of the virtual taxi V. In the example shown in FIG. 2A, the movement of the solid line arrow by the actual taxi T1 is regarded as the movement of the virtual taxi V.

  At the timing when the selected real taxi T1 drops the user, the virtual taxi calculating means 17 virtually moves the virtual taxi V according to the recommendation information. In the example shown in FIG. 2B, as the recommendation information, a position indicated by an asterisk is recommended as a movement destination, and a route indicated by a dashed arrow is calculated.

After moving the virtual taxi V according to the recommendation information, the virtual taxi calculating means 17 determines whether there is an actual taxi on which the user is boarded at the destination after the movement time has elapsed. This is because if the taxi is moved according to the recommendation information, the user is likely to have used his own taxi.
For this reason, even if there is an actual taxi T2 that is in a real vehicle state at a position different from the destination, the virtual taxi calculating means 17 does not select the actual taxi T2 (see FIG. 2C).

  When the virtual taxi V moves according to the recommendation information, for example, the virtual taxi calculation means 17 has a predetermined range (for example, 10 m) between the destination and the location where the actual taxi has taken the user, and the expected arrival When the difference between the time and the time when the actual taxi T2 gets on the user is within a predetermined period (for example, 5 minutes), the actual taxi may be regarded as the virtual taxi V.

  The virtual taxi calculation means 17 again selects the actual taxi T3 that has entered the actual vehicle state at the destination after the movement time has elapsed, as the virtual taxi V (see FIG. 2D). Thereafter, the operations shown in FIGS. 2A to 2D are repeated for a predetermined period. Thus, the virtual taxi calculation means 17 of this embodiment calculates the action content according to recommendation information and the action content based on the movement of an actual taxi as the action content of a virtual taxi.

  The actual taxi calculation means 18 calculates the actual taxi performance. Specifically, the actual taxi calculating means 18 calculates the same contents as the results and cumulative results calculated by the virtual taxi calculating means 17. The calculation method of the results and the accumulated results is the same as the method used by the virtual taxi calculation means 17.

  The ranking calculation means 19 generates a ranking of taxi performance. At this time, the ranking calculation means 19 generates a ranking obtained by adding a virtual taxi to an actual taxi in real time. For example, the ranking calculation means 19 may rank the actual taxi and the virtual taxi in descending order of sales, the maximum or average value of the moving distance of the actual vehicle, and the actual vehicle rate.

  The ranking presentation unit 20 transmits the ranking calculated by the ranking calculation unit 19 to the taxi 30. The ranking presenting means 20 may transmit the ranking to the taxi 30 periodically in real time, or may transmit it according to a request from the taxi 30.

  As described above, the ranking transmitted to the taxi 30 is a ranking obtained by adding a virtual taxi to an actual taxi. Therefore, the driver of the taxi 30 can compare the results when acting according to the recommendation information and when acting based on his / her own thoughts. At this time, when the ranking of the virtual taxi is higher than the own ranking, the driver is motivated to use the recommendation information, and as a result, the use of the recommendation information can be promoted.

  Data collection means 11, situation determination means 12, movement location determination means 13, navigation means 14, recommendation presentation means 15, virtual taxi calculation means 17, actual taxi calculation means 18, ranking calculation means 19, The ranking presentation means 20 is realized by a CPU of a computer that operates according to a program (recommendation effect visualization program).

  For example, the program is stored in a storage unit (not shown) included in the ranking presentation system, and the CPU reads the program, and according to the program, the data collection unit 11, the situation determination unit 12, the moving place determination unit 13, the navigation The means 14, recommendation presentation means 15, virtual taxi calculation means 17, actual taxi calculation means 18, ranking calculation means 19, and ranking presentation means 20 may be operated.

  Further, the data collecting means 11, the situation judging means 12, the moving place determining means 13, the navigation means 14, the recommendation presenting means 15, the virtual taxi calculating means 17, the actual taxi calculating means 18, and the ranking calculating means 19 And each of the ranking presentation means 20 may be realized by dedicated hardware. The accumulated data storage unit 16 is realized by, for example, a magnetic disk.

  Next, the operation of the ranking presentation system of this embodiment will be described. In the following description of the operation, it is assumed that the data collection unit 11 periodically collects taxi operation logs and stores them in the cumulative data storage unit 16. Further, the situation determination means 12 performs demand prediction as needed, and the moving place determination means 13 generates recommendation information indicating a recommended position (stop) based on the taxi position and the demand prediction as required. To do. In addition, the navigation unit 14 appropriately calculates a route to the position determined by the movement location determination unit 13.

  FIG. 3 is a flowchart showing an operation example of the ranking presentation system of the present embodiment. First, the virtual taxi calculating means 17 determines a virtual taxi to act according to the recommendation information from the actual taxis (step S11). In the initial state, the virtual taxi calculation means 17 may select, for example, a taxi that has entered the actual vehicle state earliest after the taxi opening hours have started.

  When the selected real taxi becomes an empty car, the virtual taxi calculating means 17 calculates the action result and cumulative result of the virtual taxi from the action of the actual taxi (step S12).

  If the business hours have already ended (Yes in step S13), the virtual taxi calculating means 17 ends the process. On the other hand, if the business hours have not yet ended (No in step S13), the virtual taxi calculation means 17 inquires the travel location determination means 13 for recommendation information, and based on the recommendation information, the destination and movement of the virtual taxi Time is calculated (step S14). That is, the virtual taxi calculating means 17 calculates the moving destination and moving time of the virtual taxi based on the new recommendation information every time the actual vehicle of the actual taxi is completed.

  The virtual taxi calculating means 17 determines the actual taxi that has entered the actual vehicle state at the destination after the travel time has elapsed as a virtual taxi (step S15). Henceforth, the process after step S12 is repeated.

  In this example of operation, the case where the virtual taxi calculating means 17 determines the destination of the virtual taxi is within business hours, but the virtual taxi calculating means 17 selects the actual taxi within the business hours. It may be determined whether or not. In this case, the virtual taxi calculating means 17 may select an actual taxi when the business hours have not ended yet.

  As described above, in the present embodiment, the virtual taxi calculation unit 17 sets a virtual taxi, and calculates the action details (specifically, the destination and the travel time) of the virtual taxi based on the recommendation information. And the virtual taxi calculation means 17 calculates the accumulation result of the action result of a virtual taxi and the action result in a predetermined period from the action of a real taxi. At that time, the virtual taxi calculating means 17 uses the information based on the actual taxi action (for example, the actual vehicle state of the actual taxi to be moved), and the action content of the virtual information provision target (specifically, the same as the actual taxi) Motion).

  Specifically, in the present embodiment, as a result of the virtual taxi calculating means 17 causing the virtual taxi to act in the real environment according to the calculated action content, a predetermined action (that is, a real vehicle) at a corresponding position in the real environment. ) When a real taxi that is in the state is detected, the real taxi is set as a virtual taxi. And the virtual taxi calculation means 17 calculates the action performance of a virtual taxi from the action of the real taxi. With such a configuration, use of recommendation information can be promoted.

  In general, the recommendation information is not actively used in many cases because it is difficult to know whether or not the vehicle allocation according to the recommendation information leads to sales. In addition, since there is no driver that follows all the recommendation information, the ranking of the results depends on the driver's intuition and experience, and the effect of the recommendation is difficult to appear.

  In the ranking presentation system of this embodiment, the use of recommendation information is promoted by the driver by incorporating gamification-like elements. For example, in the present embodiment, the virtual taxi calculating means 17 calculates the results when the recommendation information is followed, and the ranking calculating means 19 calculates the ranking including the virtual taxi. Therefore, the effect when the recommended information is followed can be visualized.

  In addition, the driver may actively use the recommendation information in order to compete with the virtual taxi and catch up with the results. Furthermore, if the improvement of the ranking becomes the driver's status, the recommendation information can be motivated more actively. In addition, by utilizing the recommendation information, even a driver with little intuition and experience can expect to achieve a certain result.

  In general, the actual performance of the day is often grasped after the day ends. However, in this embodiment, the virtual taxi calculating means 17, the actual taxi calculating means 18 and the ranking calculating means 19 total the results in real time and generate a ranking. Therefore, the driver can always grasp the trends of other drivers.

  In the present embodiment, the case where the virtual information provision target that acts according to the recommendation information is a taxi that is a moving body is exemplified, but the subject that acts according to the recommendation information is not limited to a taxi. For example, when recommending a route to be used according to road conditions, the virtual information provision target may be a general automobile.

  In this case, for example, the moving location determining means 13 transmits the recommended route as recommended information to the general car, and the virtual taxi calculating means 17 determines the virtual general car and moves based on the recommended information. You may calculate the action performance of a general car.

  Next, a modification of this embodiment will be described. In the above embodiment, the method of moving a virtual taxi according to the recommendation information and determining an actual taxi at the destination has been described. In this modification, a method for separately transmitting recommendation information during the movement to the destination will be described. The configuration of this modification is the same as that of the above embodiment. In addition, it is assumed that the taxi 30 regularly transmits the current position to the ranking presentation system 10.

  The situation determination means 12 performs taxi demand prediction at any time. And when more effective demand prediction is performed, the movement location determination means 13 produces | generates the recommendation information shown to the taxi 30 based on the demand prediction and the present position of the taxi 30. FIG. Then, the recommendation presentation unit 15 transmits the generated recommendation information to the taxi 30.

  At this time, if the new taxi taxi receives new recommendation information while moving, the virtual taxi calculation means 17 specifies the destination of the virtual taxi and the travel time to the destination based on the new recommendation information. The subsequent processing is the same as in the above embodiment.

  According to this modification, when the situation determination unit 12 performs more effective demand prediction, the moving place determination unit 13 newly generates recommendation information based on the demand prediction and the current position of the taxi 30. Specifically, when the virtual taxi calculation means 17 obtains new recommendation information during the action of the virtual taxi in the real environment according to the calculated action content (destination and travel time), The action content of the virtual taxi is calculated based on the recommendation information.

  Thus, by creating the recommendation information as needed, more effective recommendation information can be provided to the taxi 30. Moreover, since the recommendation information effective in real time can be transmitted also to the taxi which does not necessarily move according to recommendation information, the effect of recommendation information can be improved.

Embodiment 2. FIG.
Next, a second embodiment of the ranking presentation system of the present invention will be described. The configuration of the ranking presentation system of this embodiment is the same as that of the first embodiment. However, the virtual taxi calculation means 17 of this embodiment differs from 1st Embodiment in the point which determines the action content of a virtual taxi using the statistical characteristic of a real taxi.

  Specifically, in the first embodiment, the virtual taxi calculating unit 17 calculates the action content according to the recommendation information and the action content based on the movement of the actual taxi as the action details of the virtual taxi. On the other hand, in this embodiment, the virtual taxi calculating means 17 calculates the action content according to the recommendation information and the action content based on the statistical information of the actual taxi as the action details of the virtual taxi. Other contents are the same as those in the first embodiment.

  The virtual taxi calculation means 17 sets a virtual taxi. In the initial state, the virtual taxi calculating means 17 virtually sets the position of the virtual taxi. The virtual taxi calculating means 17 may be set, for example, when a virtual taxi exists at a taxi establishment.

  Next, the virtual taxi calculating means 17 receives the recommendation information and the route information from the moving place determining means 13. The virtual taxi calculating means 17 specifies the destination of the virtual taxi and the travel time to the destination based on the received recommendation information and route information.

  The virtual taxi calculating means 17 estimates the actual state of the virtual taxi based on the statistical information after moving the virtual taxi based on the recommendation information. And the virtual taxi calculation means 17 calculates the action performance of a virtual taxi based on the estimated actual vehicle situation. That is, in the present embodiment, it is assumed that the virtual taxi has placed a user on the basis of the estimated actual vehicle situation, and the virtual taxi movement location (movement distance) is also determined based on the estimated actual vehicle situation. To do.

  The statistical information used to estimate the actual vehicle status of the virtual taxi uses the statistical characteristics of the virtual taxi such as the taxi operation results in the actual environment. In addition, the calculation method of statistical information is arbitrary. The virtual taxi calculating means 17 is at least one statistical information of the travel information when the vehicle information is obtained by the boarding information indicating the actual vehicle state for each position and time, the empty vehicle information indicating the empty vehicle state for each position and time, and the actual vehicle for each position and time. Is used to calculate the action details of a virtual taxi.

  The virtual taxi calculating means 17 moves the virtual taxi based on the recommendation information, and then based on the statistical information (empty vehicle information or boarding information), the taxi empty / actual vehicle state in the time zone and position after the movement is actually Estimate as a situation. Specifically, the virtual taxi calculating means 17 is based on statistical information such as which taxi is in an empty car or an actual car at which time zone, how much it travels until a user is put on it, and what destination it travels to. To estimate. At this time, the virtual taxi calculating means 17 may estimate the occurrence probability of the customer in the time zone and position after movement as the actual vehicle situation based on the statistical information.

  When it is estimated that the virtual taxi is in the actual vehicle state from the estimated actual vehicle situation, the virtual taxi calculation means 17 estimates the taxi destination at the time and position in the actual vehicle state based on the statistical information. That is, the virtual taxi calculating means 17 estimates the travel distance and destination of the virtual taxi based on the statistical information when the virtual taxi becomes a real vehicle. Furthermore, the virtual taxi calculating means 17 calculates the estimated travel time to the destination.

  The virtual taxi calculating means 17 considers that the actual vehicle of the virtual taxi has been completed after the calculated travel time has elapsed, and calculates the behavior of the virtual taxi according to the estimated travel to the destination. The virtual taxi calculation means 17 may calculate the sales of the virtual taxi as an action result based on the travel distance, travel time, initial fare, etc. of the virtual taxi, for example. That is, it can be said that the action result is calculated from the action of an actual taxi.

  When the virtual taxi becomes an empty vehicle (that is, the virtual taxi is completed), the virtual taxi calculating means 17 receives new recommendation information and route information from the moving location determining means 13. Then, the virtual taxi calculating means 17 specifies the next destination of the virtual taxi and the travel time to the destination based on the received recommendation information and route information. And the virtual taxi calculation means 17 estimates again the actual vehicle situation of a virtual taxi based on statistical information after progress of the specified movement time.

  Thus, the virtual taxi calculating means 17 of this embodiment calculates the action content of a virtual taxi based on new recommendation information every time a virtual taxi completes a real vehicle. Then, the virtual taxi calculating means 17 moves the virtual taxi according to the action content calculated based on the recommendation information, and estimates the virtual taxi moving destination in the moved position and time zone based on the statistical information. .

  Next, the operation of the ranking presentation system of this embodiment will be described. The operations of the data collection unit 11, the situation determination unit 12, the movement location determination unit 13, and the navigation unit 14 are the same as the contents illustrated in FIG.

  FIG. 4 is a flowchart illustrating an operation example of the ranking presentation system in the second embodiment. First, the virtual taxi calculating means 17 sets a virtual taxi that acts in accordance with the recommendation information (step S21). In the initial state, the virtual taxi calculation means 17 may be set to indicate that a virtual taxi exists at an arbitrary position.

  Next, the virtual taxi calculating means 17 inquires the recommendation information to the moving place determining means 13, and calculates the moving destination and moving time of the virtual taxi based on the recommended information (step S22).

  Based on the statistical information, the virtual taxi calculating means 17 estimates the actual vehicle situation of the virtual taxi at the destination after the movement time has elapsed (step S23). When it is estimated from the estimated actual vehicle situation that the virtual taxi is in the actual vehicle state, the virtual taxi calculating means 17 estimates the movement destination of the virtual taxi at the time and position when the actual vehicle state is reached based on the statistical information ( Step S24). Further, the virtual taxi calculating means 17 calculates the estimated travel time to the destination (step S25).

  The virtual taxi calculation means 17 calculates the action results and cumulative results of the virtual taxi according to the estimated movement to the destination (step S26).

  If the business hours have already ended (Yes in step S27), the virtual taxi calculating means 17 ends the process. On the other hand, when business hours have not ended yet (No in Step S27), virtual taxi calculation means 17 repeats processing after Step S22.

  As described above, according to the present embodiment, compared to the first embodiment, the virtual taxi calculation means 17 calculates the action content of the virtual taxi using the statistical characteristics of the actual taxi. Thus, by using statistical information, in addition to the effects of the first embodiment, it is possible to suppress the action performance of the virtual taxi from being calculated with unique information. Moreover, according to this embodiment, even if it is a case where the movement of a real taxi cannot be grasped | ascertained, the action performance of a virtual taxi can be estimated.

  Next, the outline of the present invention will be described. FIG. 5 is a block diagram showing an outline of the recommendation effect visualization system according to the present invention. The recommendation effect visualization system according to the present invention is based on virtual object setting means 81 (for example, virtual taxi calculation means 17) for setting a virtual information providing object (for example, virtual taxi) that acts according to recommendation information, and on the recommendation information. Action content calculation means 82 (for example, virtual taxi calculation means 17) for calculating the action contents (for example, destination and travel time) of the virtual information provision target, and the behavior of the information provision target (for example, real taxi) in the real environment From the virtual result calculation means 83 (for example, the virtual taxi calculation means 17) for calculating the action result of the virtual information provision target, and the accumulated results of the action results (for example, sales for a day) Cumulative result calculation means 84 (for example, virtual taxi calculation means 17) for calculating the actual vehicle rate, travel distance, etc. It is provided.

  The action content calculation unit 82 calculates the action content of the virtual information provision target based on information based on the behavior of the information provision target in the real environment (for example, actual taxi movement, statistical information), and the cumulative result calculation unit 84 In a predetermined period, a cumulative result of action results of virtual information provision targets that act based on each recommendation information is calculated.

  With such a configuration, use of recommendation information can be promoted. That is, according to the configuration illustrated in FIG. 5, each information providing target can grasp the actual performance of the virtual information providing target that acts based on the recommendation information. Therefore, the effect of the recommendation information can be grasped by comparing with its own results, and as a result, the use of the recommendation information can be promoted.

  Moreover, the action content calculation means 82 may calculate the action content of a virtual information provision object using the statistical characteristic of the information provision object in a real environment. Specifically, the action content calculation means 82 includes taxi boarding information (for example, actual vehicle state for each position and time), empty vehicle information (for example, empty state for each position and time), and travel distance (for position and time). The action content of the virtual information provision target may be calculated using at least one of the distance traveled in the case of a real vehicle every time).

  According to such a structure, it can suppress that the action performance of a virtual information provision object is calculated by the information of a specific information provision object. Further, even if the movement of the information providing target in the real environment cannot be grasped, the action performance of the virtual information providing target can be predicted.

  On the other hand, as a result of causing the virtual information provision target to act in the real environment according to the calculated action content, the virtual target setting unit 81 determines a predetermined action state (for example, a user by taxi) at a corresponding position in the real environment. When an information provision target in the real environment that is in the state of “the actual vehicle” is detected, the information provision target in the real environment may be set as a virtual information provision target. And the virtual performance calculation means 83 may calculate the action performance of a virtual information provision object from the action of the information provision object in a real environment.

  According to such a configuration, it is possible to easily calculate the behavior record of the virtual information provision target from the behavior of the information provision target in the real environment.

  Further, the action content calculation means 82 is a virtual information based on new recommendation information each time a predetermined action (for example, “actual car” in which a user is on a taxi) is completed in an actual environment. You may calculate the action content of provision object. In addition, when the action content calculation unit 82 obtains new recommendation information while the virtual information provision target is acting in the real environment according to the calculated action content, the action content calculation unit 82 calculates the virtual content based on the recommendation information. The action content of the information provision target may be calculated.

  With such a configuration, more effective recommendation information can be provided to the information providing target person. Moreover, since the recommendation information effective in real time can be transmitted also to the information provision target person who has not necessarily performed the action along recommendation information, the effect of recommendation information can be improved.

  Specifically, the action content calculation means 82 calculates the travel destination and travel time of a virtual taxi that acts in accordance with the recommended information based on the new recommended information every time the actual taxi vehicle in the real environment is completed. Also good. Further, the virtual target setting unit 81 may set a taxi that is in a real vehicle state at the destination after the travel time has elapsed from among taxis in the real environment as a virtual taxi. Moreover, the virtual performance calculation means 83 may calculate the virtual taxi behavior record from the taxi behavior in the real environment. And the accumulation result calculation means 84 may calculate the accumulation result of the action performance of the virtual taxi which acts based on each recommendation information in a predetermined period.

  In addition, the cumulative results of action results for each information provision target in the real environment (for example, sales results for each actual taxi) are totaled, and the cumulative results of action results for the virtual information provision target (for example, sales results for virtual taxis) are calculated. You may provide the ranking production | generation means (for example, ranking calculation means 19) which produces | generates the ranking of the information provision object included.

  With such a configuration, each information provision target person can grasp the rank among the information provision target persons and can also compare with a virtual information provision target.

  Moreover, you may provide the recommendation information production | generation means (For example, the situation judgment means 12, the movement location determination means 13) which produces recommendation information based on the present condition. With such a configuration, when information that changes from moment to moment is predicted, more accurate prediction can be made by making various decisions based on information indicating the current situation.

  The present invention is preferably applied to a recommendation effect visualization system that can visualize the effect of recommended information. The present invention is also preferably applied to a system for evaluating the effectiveness when acting based on recommendation information.

DESCRIPTION OF SYMBOLS 10 Ranking presentation system 11 Data collection means 12 Status judgment means 13 Movement place determination means 14 Navigation means 15 Recommendation presentation means 16 Accumulated data storage means 17 Virtual taxi calculation means 18 Actual taxi calculation means 19 Ranking calculation means 20 Ranking presentation means 30 Taxi

Claims (15)

Virtual target setting means for setting a virtual information provision target that acts according to the recommendation information;
Action content calculating means for calculating the action content of the virtual information provision target based on recommendation information;
A virtual result calculating means for calculating the action result of the virtual information providing target from the action of the information providing target in the real environment;
A cumulative result calculating means for calculating a cumulative result of the action results in a predetermined period,
The action content calculation means calculates the action content of the virtual information provision target based on information based on the behavior of the information provision target in the real environment,
The cumulative result calculating means calculates a cumulative result of action results of virtual information provision targets that act based on each recommendation information in the predetermined period. The recommendation effect visualization system. The recommendation effect visualization system according to claim 1, wherein the behavior content calculation means calculates the behavior content of the virtual information provision target using statistical characteristics of the information provision target in the real environment. The action content calculation means calculates the action content of a virtual information provision target using at least one of taxi boarding information, empty car information, and travel distance in a real environment. system. The virtual target setting means determines the information provision target in the real environment that is in a predetermined behavior state at the corresponding position in the real environment as a result of causing the virtual information provision target to act in the real environment according to the calculated action content. If detected, set the information provision target in the real environment as a virtual information provision target,
The recommendation effect visualization system according to claim 1, wherein the virtual result calculation means calculates the action result of the virtual information provision target from the behavior of the information provision target in the real environment. The recommendation effect visualization system according to claim 4, wherein the action content calculation means calculates the action content of the virtual information provision target based on the new recommendation information every time the predetermined action of the information provision target in the real environment is completed. The action content calculation means provides virtual information based on the recommended information when new recommendation information is obtained while the virtual information providing target is acting in the real environment according to the calculated action content. The recommendation effect visualization system according to claim 4, wherein the action content of the target is calculated. The action content calculation means calculates the travel destination and travel time of a virtual taxi that acts in accordance with the recommendation information, based on the new recommendation information every time the taxi vehicle in the real environment is completed,
The virtual target setting means sets a taxi that is in a real vehicle state at the destination after the travel time has elapsed, from among taxis in a real environment,
The virtual result calculating means calculates the action result of the virtual taxi from the action of the taxi in the real environment,
The recommendation result visualization system according to any one of claims 4 to 6, wherein the accumulation result calculation means calculates an accumulation result of action results of a virtual taxi that acts based on each recommendation information in a predetermined period. . 2. A ranking generation unit that aggregates the cumulative results of action results of each information provision target in a real environment and generates a ranking of information provision targets including the cumulative result of action results of virtual information provision targets. Item 8. The recommendation effect visualization system according to any one of Items 7 to 9. The recommendation effect visualization system according to any one of claims 1 to 8, further comprising recommendation information generation means for generating recommendation information based on a current situation. Set a virtual information provision target to act according to recommendation information,
Based on the recommendation information, the action content of the virtual information provision target is calculated based on the information based on the behavior of the information provision target in the real environment,
From the behavior of the information provision target in the real environment, the action performance of the virtual information provision target is calculated,
A recommendation effect visualization method characterized in that, during a predetermined period, a cumulative result of action results of virtual information provision targets that act based on each recommendation information is calculated. The recommendation effect visualization method according to claim 10, wherein the action content of the virtual information provision target is calculated using statistical characteristics of the information provision target in the real environment. As a result of causing the virtual information provision target to act in the real environment according to the calculated action content, when the information provision target in the real environment that has entered a predetermined behavior state at the corresponding position in the real environment is detected, Set the information provision target in the real environment as a virtual information provision target,
The recommendation effect visualization method according to claim 10, wherein an action result of the virtual information provision target is calculated from an action of the information provision target in the real environment. On the computer,
A virtual object setting process for setting a virtual information provision target to act according to recommendation information;
An action content calculation process for calculating the action content of the virtual information provision target based on the recommendation information;
A virtual result calculation process for calculating an action result of the virtual information provision target from an action of the information provision target in the real environment; and
A cumulative result calculation process for calculating a cumulative result of the action results in a predetermined period is executed;
In the behavior content calculation process, the behavior content of the virtual information provision target is calculated based on information based on the behavior of the information provision target in the real environment,
The recommendation effect visualization program for calculating the accumulation result of the action performance of the virtual information provision object which acts based on each recommendation information in the predetermined period in the accumulation result calculation process. On the computer,
The recommendation effect visualization program according to claim 13, wherein the behavior content calculation process calculates the behavior content of the virtual information provision target using the statistical characteristics of the information provision target in the real environment. On the computer,
In the virtual target setting process, as a result of causing the virtual information provision target to act in the real environment according to the calculated action content, the information provision target in the real environment that has entered a predetermined behavior state at a corresponding position in the real environment If detected, the information provision target in the real environment is set as a virtual information provision target,
The recommendation effect visualization program according to claim 13, wherein in the virtual result calculation process, the action result of the virtual information provision target is calculated from the behavior of the information provision target in the real environment.

Images