Influencer extraction device, influencer extraction method, program, and recording medium

Abstract

To provide an influencer extraction device capable of extracting an influencer.SOLUTION: In an influencer extraction device 10 according to the present invention: an aggregation data generation unit 12 generates aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor on the basis of visit information and encounter information; a setting unit 13 analyzes the aggregation data to set a prediction expression for calculating a prediction value of the number of encounters and a standard error; a prediction value calculation unit 14 uses the prediction expression to calculate a prediction value of the number of encounters for each visitor; an error calculation unit 15 calculates an error between the number of encounters in the aggregation data and the prediction value for each visitor on the basis of the number of encounters in the aggregation data, the prediction value of the number of encounters, and the standard error; and an influencer extraction unit 16 extracts a visitor having the error that is equal to or more than a preset threshold as an influencer from among a plurality of visitors.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an influencer extraction device, an influencer extraction method, a program, and a recording medium.

Background Art

[0002] "Influencer marketing," which uses influencers to promote a target of promotion, has attracted attention. An influencer is a person who actively communicates with others and has a great influence on society and others. Patent Document 1 reports a technique for extracting influencers from among users in social media.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the technique related to Patent Document 1 has a problem that it cannot find influencers in "places where people communicate with each other" other than social media.

[0005] Therefore, an object of the present invention is to provide an influencer extraction device, an influencer extraction method, a program, and a recording medium capable of extracting influencers in places other than social media.

Means for Solving the Problems

[0006] To achieve the above object, the influencer extraction device of the present invention includes an acquisition unit, an aggregated data generation unit, a setting unit, a predicted value calculation unit, an error calculation unit, and an influencer extraction unit, The acquisition unit acquires visit information and encounter information, The visit information is information associating identification information indicating individual visitors with time information indicating the time when the visitors visited an arbitrary location, The encounter information is information associating the identification information of each of the visitors who encountered each other at the arbitrary location with time information indicating the encounter time, The aggregation data generation unit generates aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the visit information and the encounter information, The setting unit analyzes the aggregation data and sets a prediction formula for calculating a predicted value of the number of encounters and a standard error, The predicted value calculation unit calculates a predicted value of the number of encounters for each visitor using the prediction formula, The error calculation unit calculates the error between the number of encounters in the aggregation data and the predicted value for each visitor based on the number of encounters in the aggregation data, the predicted value of the number of encounters, and the standard error, The influencer extraction unit is a device that extracts, as influencers, those visitors among a plurality of visitors whose error is equal to or greater than a preset threshold value.

[0007] The influencer extraction method of the present invention includes an acquisition step, an aggregation data generation step, a setting step, a predicted value calculation step, an error calculation step, and an influencer extraction step, The acquisition step acquires visit information and encounter information, The visit information is information associating identification information indicating individual visitors with time information indicating the time when the visitors visited an arbitrary location, The encounter information is information associating the identification information of each of the visitors who encountered each other at the arbitrary location with time information indicating the encounter time, The aggregation data generation step generates aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the visit information and the encounter information, The setting process analyzes the aggregated data to set a prediction formula for calculating the predicted value of the encounter frequency and a standard error, The predicted value calculation process calculates the predicted value of the encounter frequency for each visitor using the prediction formula, The error calculation process calculates, for each visitor, the error between the encounter frequency in the aggregated data and the predicted value based on the encounter frequency in the aggregated data, the predicted value of the encounter frequency, and the standard error, The influencer extraction process is a method of extracting, as influencers, those visitors among a plurality of visitors whose error is equal to or greater than a preset threshold value.

Advantages of the Invention

[0008] According to the present invention, influencers in a place other than social media can be extracted.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Mode for Carrying Out the Invention

[0010] In the influencer extraction device of the present invention, for example, The influencer extraction unit may be configured to exclude at least one of the visitors whose error is equal to or greater than a preset upper limit value and the visitors whose number of visits is equal to or greater than a preset lower limit value from the extraction target.

[0011] The influencer extraction device of the present invention, for example, Further includes a filtering unit, The acquisition unit acquires at least one of visitable time zone information and attribute information, The visitable time zone information is information indicating a time zone in which the visitor can visit the arbitrary location, The attribute information is information indicating the attribute of the visitor, The filtering unit performs filtering on the visit information and the encounter information using at least one of the acquired visitable time zone information and the attribute information, The aggregation data generation unit may be configured to generate aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the filtered visit information and encounter information.

[0012] The influencer extraction device of the present invention, for example, Further includes a recording unit, The recording unit may be configured to record at least one of the visit information and the encounter information.

[0013] The influencer extraction device of the present invention, for example, Further includes a congestion information disclosure unit, The congestion information disclosure unit may be configured to generate congestion information indicating the number of visitors for each time zone at the arbitrary location based on the visit information, and disclose the congestion information.

[0014] The influencer extraction device of the present invention, for example, further includes a notification unit, The notification unit may be configured to notify the visitor extracted as an influencer that they have been extracted as an influencer.

[0015] In the influencer extraction method of the present invention, for example, The influencer extraction step may be configured to exclude at least one of the visitors whose error is equal to or greater than a preset upper limit value and the visitors whose number of visits is equal to or greater than a preset lower limit value from the extraction target.

[0016] The influencer extraction method of the present invention, for example, further includes a filtering step, The acquisition step acquires at least one of visitable time zone information and attribute information, The visitable time zone information is information indicating the time zone in which the visitor can visit the arbitrary place, The attribute information is information indicating the attribute of the visitor, The filtering step filters the visit information and the encounter information using at least one of the acquired visitable time zone information and the attribute information, The aggregation data generation step may be configured to generate aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the filtered visit information and encounter information.

[0017] The influencer extraction method of the present invention, for example, further includes a recording step, The recording step may be configured to record at least one of the visit information and the encounter information.

[0018] The influencer extraction method of the present invention, for example, further includes a congestion information disclosure step, The congestion information disclosure step may be configured to generate congestion information indicating the number of visitors for each time period at the arbitrary location based on the visit information, and disclose the congestion information.

[0019] The method for extracting influencers of the present invention includes, for example, further including a notification step, wherein the notification step may be configured to notify the visitor extracted as an influencer that the visitor has been extracted as an influencer.

[0020] The program of the present invention is a program for causing a computer to execute each step of the method of the present invention as a procedure.

[0021] The recording medium of the present invention is a computer-readable recording medium storing the program of the present invention.

[0022] In the present invention, "encounter" means that there is a fact that two or more visitors are present at the same time and in the same place. For example, it should be interpreted in the broadest sense, including passing by, overtaking, crossing, etc., and is not limited in any sense. Further, "encounter" is not limited to mutual recognition such as greeting or having a conversation between each visitor.

[0023] Next, embodiments of the present invention will be described with reference to the drawings. The present invention is not limited to the following embodiments. In the following drawings, the same parts are denoted by the same reference numerals. In addition, unless otherwise specified, the descriptions of the respective embodiments can be mutually referred to, and the configurations of the respective embodiments can be combined unless otherwise specified.

[0024] [Embodiment 1] FIG. 1 is a block diagram showing a configuration of an example of the influencer extraction device 10 according to the present embodiment. As shown in FIG. 1, the device 10 includes an acquisition unit 11, an aggregated data generation unit 12, a setting unit 13, a predicted value calculation unit 14, an error calculation unit 15, and an influencer extraction unit 16. Further, as an optional configuration, the device 10 may further include a filtering unit 17, a recording unit 18, a congestion information disclosure unit 19, and a notification unit 20.

[0025] The device 10 may be, for example, a single device including each of the above units, or a device in which each of the above units can be connected via a communication network. Further, the device 10 can be connected to an external device described later via the communication network. The communication network is not particularly limited, and a known network can be used. For example, it may be wired or wireless. Examples of the communication network include an Internet line, WWW (World Wide Web), a telephone line, a LAN (Local Area Network), a SAN (Storage Area Network), a DTN (Delay Tolerant Networking), an LPWA (Low Power Wide Area), an L5G (local 5G), and the like. Examples of wireless communication include Wi-Fi (registered trademark), Bluetooth (registered trademark), local 5G, LPWA, and the like. The wireless communication may be a form in which each device directly communicates (Ad Hoc communication), infrastructure communication, indirect communication via an access point, or the like. The device 10 may be incorporated into a server as a system, for example. Further, the device 10 may be, for example, a personal computer (PC, for example, a desktop type or a notebook type) installed with the program of the present invention, a smartphone, a tablet terminal, or the like. Further, the device 10 may be in a form such as cloud computing or edge computing, for example, in which at least one of the above units is on a server and the other units are on a terminal.

[0026] Figure 2 illustrates a block diagram of the hardware configuration of the present apparatus 10. The present apparatus 10 includes, for example, a central processing unit (CPU, GPU, etc.) 101, a memory 102, a bus 103, a storage device 104, an input device 105, a display device 106, a communication device 107, etc. Each part of the present apparatus 10 is interconnected via the bus 103 by respective interfaces (I / F).

[0027] The central processing unit 101 is responsible for overall control of the present apparatus 10. In the present apparatus 10, for example, the program of the present invention and other programs are executed by the central processing unit 101, and various information is read and written. Specifically, for example, the central processing unit 101 functions as an acquisition unit 11, an aggregated data generation unit 12, a setting unit 13, a predicted value calculation unit 14, an error calculation unit 15, and an influencer extraction unit 16. Further, the central processing unit 101 also functions as, for example, a filtering unit 17, a recording unit 18, a congestion information disclosure unit 19, and a notification unit 20.

[0028] The bus 103 can be connected to an external device, for example. Examples of the external device include an external storage device (external database, etc.), a printer, an external input device, an external display device, an external imaging device, etc. The present apparatus 10 can be connected to an external network (the communication line network) by, for example, a communication device 107 connected to the bus 103, and can also be connected to other devices via the external network.

[0029] The memory 102 includes, for example, a main memory (main storage device). When the central processing unit 101 performs processing, for example, the memory 102 reads various operation programs such as the program of the present invention stored in the storage device 104 described later, and the central processing unit 101 receives data from the memory 102 and executes the program. The main memory is, for example, a RAM (random access memory). Further, the memory 102 may be, for example, a ROM (read-only memory).

[0030] The memory device 104 is, for example, also referred to as a so-called auxiliary storage device with respect to the main memory (primary storage device). As described above, an operation program including the program of the present invention is stored in the memory device 104. The memory device 104 may be, for example, a combination of a recording medium and a drive for reading and writing to the recording medium. The recording medium is not particularly limited, and may be, for example, an internal type or an external type, and examples include an HD (hard disk), CD-ROM, CD-R, CD-RW, MO, DVD, flash memory, memory card, etc. The memory device 104 may be, for example, a hard disk drive (HDD) in which a recording medium and a drive are integrated, and a solid state drive (SSD).

[0031] In the present apparatus 10, the memory 102 and the memory device 104 can also store various information such as log information, information acquired from an external database (not shown) or an external device, information generated by the present apparatus 10, and information used when the present apparatus 10 executes processing. Note that at least some of the information may be stored in an external server other than the memory 102 and the memory device 104, for example, or may be distributed and stored in a plurality of terminals using blockchain technology or the like. Further, the memory 102 and the memory device 104 may store, for example, various information recorded by the recording unit 18.

[0032] The present apparatus 10 may further include, for example, an input device 105 and a display device 106. The input device 105 is, for example, a touch panel, keyboard, mouse, etc. The display device 106 includes, for example, an LED display, a liquid crystal display, etc.

[0033] Next, an example of the influencer extraction method of the present embodiment will be described based on the flowchart of FIG. 3. The influencer extraction method of the present embodiment is implemented as follows, for example, using the influencer extraction device 10 of FIG. 1. Note that the influencer extraction method of the present embodiment is not limited to the use of the influencer extraction device 10 of FIG. 1. Also, the steps shown in parentheses in FIG. 3 are optional steps.

[0034] First, the acquisition unit 11 acquires visit information and encounter information (S11a, acquisition step). The visit information is information that associates identification information indicating each visitor with time information indicating the time when the visitor visited an arbitrary location. The encounter information is information that associates the identification information of each of the visitors who encountered each other at the arbitrary location with time information indicating the time of the encounter. The acquisition unit 11 may acquire the visit information and the encounter information from an external server, an external database, etc. via the communication network, or may acquire them by reading the visit information and the encounter information stored in the memory 102 and the storage device 104. The identification information may be, for example, the identification information of a carried item carried by the visitor when the visitor visits the arbitrary location. Examples of the carried item include an admission ticket, an IC (integrated circuit) card, a mobile phone, a smartphone, a wearable device, etc. The identification information is, for example, an ID (identification), an email address, biometric information (face, fingerprint, voice, etc.). Note that the arbitrary location is not particularly limited and may be indoors, outdoors, or inside a moving body such as a train. Specifically, examples of the arbitrary location include a garbage station, a park, a supermarket, a convenience store, etc.

[0035] Next, the aggregation data generation unit 12 generates aggregation data in which the number of visits and the number of encounters are aggregated for each visitor based on the visit information and the encounter information (S12, aggregation data generation step). Since it is considered that the number of encounters increases as the number of visits increases, as will be described later, it can be said that there is a positive correlation between the number of visits and the number of encounters.

[0036] Next, the setting unit 13 analyzes the aggregation data and sets a prediction formula for calculating a predicted value of the number of encounters and a standard error (S13, setting step). Specifically, it will be described later.

[0037] Next, the predicted value calculation unit 14 calculates the predicted value of the number of encounters for each of the visitors using the prediction formula (S14, predicted value calculation step).

[0038] Next, the error calculation unit 15 calculates the error between the number of encounters in the aggregated data and the predicted value for each of the visitors based on the number of encounters in the aggregated data, the predicted value of the number of encounters, and the standard error (S15, error calculation step).

[0039] Then, the influencer extraction unit 16 extracts, from the visitor information, the visitors whose error is equal to or greater than a preset threshold as influencers (S16, influencer extraction step), and ends (END). Further, the influencer extraction unit 16 may exclude, for example, at least one of the visitors whose error is equal to or greater than a preset upper limit value and the visitors whose number of visits is equal to or greater than a preset lower limit value from the extraction target. The threshold value, the upper limit value, and the lower limit value are not particularly limited and can be set arbitrarily. An error equal to or greater than a preset upper limit value may, for example, be a potential outlier. Therefore, appropriate influencers can be extracted by excluding the visitors having such an error from the extraction target. Also, visitors whose number of visits is equal to or greater than a preset lower limit value may, for example, not be familiar with the arbitrary location or other visitors even if the error is equal to or greater than the threshold value. Therefore, appropriate influencers can be extracted by excluding such visitors from the extraction target. That is, an influencer can also be said to be a person who has a relatively large number of encounters compared to the number of visits.

[0040] The influencer extraction unit 16 may, for example, extract, from the visitor information, a preset number of the visitors whose error is equal to or greater than a preset threshold and whose error is high in descending order as influencers. The preset number is not particularly limited and can be set arbitrarily. Specifically, when the preset number is 1, for example, the visitor with the highest error is extracted as an influencer.

[0041] As described above, the apparatus 10 may further include, for example, a filtering unit 17. In this case, the acquisition unit 11 acquires at least one of the visitable time zone information and the attribute information (S11b, acquisition step). The visitable time zone information indicates the time zones during which the visitor can visit the arbitrary location. The time zones may be, for example, time zones every 30 minutes or time zones every hour. Also, the time zones may be, for example, time zones by day of the week. The attribute information is the attribute of the visitor, and specifically, for example, there are attributes such as gender, age, hobby, occupation, etc. The visitable time zone information and the attribute information are associated with, for example, the identification information of the visitor. The acquisition unit 11 may acquire the visitable time zone information and the attribute information from an external server, an external database, etc. via the communication network, or may acquire them by reading the visitable time zone information and the attribute information stored in the memory 102 and the storage device 104. The step (S11b) may be executed in parallel with the step (S11a) or may be executed in order, as shown in FIG. 3. The order is not particularly limited.

[0042] After executing the step (S11a) and the step (S11b), the filtering unit 17 performs filtering on the visit information and the encounter information using at least one of the acquired visitable time zone information and the attribute information (S17, filtering step). That is, assuming that the time zone indicated by the visitable time zone information for visitor X is "from 14:00 to 17:00 on Monday to Friday", the filtering unit 17 extracts (filters) the visit information and the encounter information having time information within the time zone of "from 14:00 to 17:00 on Monday to Friday" from the visit information and the encounter information. Also, assuming that the attribute indicated by the attribute information for visitor Y is "in her 40s, female", the filtering unit 17 extracts (filters) the visit information and the encounter information of the visitor having the attribute information of "in her 40s, female" from the visit information and the encounter information.

[0043] After executing the step (S17), in the step (S12), the aggregation data generation unit 12 generates aggregation data in which the number of visits and the number of encounters are aggregated for each visitor based on the filtered visit information and encounter information.

[0044] In this way, by using the visitable time zone information to filter the visit information and the encounter information, for example, it is possible to prevent the error from becoming large due to the convenience of the schedule of visitor X. Also, by using the attribute information to filter the visit information and the encounter information, for example, influencers can be extracted for each attribute of the visitors, and diversity is generated among the extracted influencers. When the influencers diversify, the people influenced by the influencers also diversify, and as a result, it leads to the diversification of the visitors.

[0045] As described above, the apparatus 10 may further include, for example, a recording unit 18. The recording unit 18 records information such as the visit information, the encounter information, the visitable time zone information, and the attribute information (S18, recording step). The step (S18) may be executed, for example, before the steps (S11a) and (S11b) as shown in FIG. 3. The method of recording each piece of information is not particularly limited.

[0046] When recording the visit information, the recording unit 18 may record the visit information, for example, by reading the identification information of the carried item when a visitor carrying the carried item visits the arbitrary location. Also, when recording the visit information, the recording unit 18 may record the visit information, for example, by detecting that a visitor carrying the carried item has entered the arbitrary location by using Wi-Fi (registered trademark), Bluetooth (registered trademark), or the like. Further, when recording the visit information, the recording unit 18 may record the visit information by detecting that the visitor has entered the arbitrary location by using a sensor, a camera, or the like.

[0047] When recording the encounter information, the recording unit 18 may record the encounter information by detecting, for example, that a plurality of visitors carrying the carried items have entered the arbitrary location using Wi-Fi (registered trademark), Bluetooth (registered trademark), or the like. Further, when recording the encounter information, the recording unit 18 may, for example, presume that one visitor and another visitor have encountered each other and record the encounter information when the other visitor has visited within a certain time (for example, within 5 minutes) before and after one visitor has visited at the arbitrary location.

[0048] When recording the visitable time zone information and the attribute information, the recording unit 18 may record the visitable time zone information and the attribute information by receiving an input of the visitable time zone information and the attribute information from the terminal of the visitor (for example, a mobile phone, a smartphone, a PC, a wearable device, etc.).

[0049] As described above, the apparatus 10 may further include, for example, a congestion information disclosure unit 19. The congestion information disclosure unit 19 generates congestion information indicating the number of visitors for each time period at the arbitrary location based on the visit information, and discloses the congestion information (S19, congestion information disclosure step). Specifically, the congestion information disclosure unit 19 may use, for example, a numerical value obtained by calculating the average number of visits for each time period as the congestion information. The time period may be, for example, a 30-minute time period or a 1-hour time period. Also, the time period may be, for example, a time period by day of the week. The method of disclosing the congestion information is not particularly limited. For example, the congestion information may be disclosed by transmitting (notifying) the congestion information to the terminal of the visitor via the communication network, or the congestion information may be disclosed by displaying the congestion information on the display device 106. The step (S19) may be executed, for example, before the step (S11a) as shown in FIG. 3. Also, the step (S19) may be executed, for example, after the step (S18) or before the step (S18) as shown in FIG. 3. As a factor for increasing the error between the encounter count and the predicted value in the aggregated data, for example, it is considered that when the visitor comes to the arbitrary location during the time period when the location is congested, the encounter count becomes larger than the number of visits per time because the number of visits per time is increased. Therefore, by disclosing the congestion status to the visitor before the visitor arrives, it is possible to extract visitors who deliberately come to the arbitrary location during the time period when the location is congested while being aware of communication with others.

[0050] Also, for example, the congestion information may be disclosed to the visitors in advance without relying on the apparatus 10.

[0051] As described above, the device 10 may further include, for example, a notification unit 20. The notification unit 20 notifies the visitor extracted as an influencer that they have been extracted as an influencer (S20, notification step). "Extracted" may also be referred to as, for example, "selected". The method of the notification is not particularly limited. For example, the visitor's terminal may be notified that they have been selected as an influencer, or the fact that they have been selected as an influencer or the visitor selected as an influencer may be displayed on the display device 106. In addition, the notification unit 20 may notify the visitor of the request information, for example, in addition to the fact that they have been selected as an influencer. The request information is information for requesting the promotion of the promotion target. Note that the promotion target is not particularly limited, and includes products, services, places (specific areas, facilities, etc.), etc., and may be tangible or intangible. The request information may include, for example, information about the promotion target and information such as the place and date / time for executing the promotion. Note that the arbitrary place where the influencer is extracted may be different from the place that is the promotion target and the place where the promotion is executed. For example, an influencer in the supermarket may be extracted from among the visitors to the supermarket, and the person may be requested (notified) to also act as an influencer at the garbage station. In this way, by asking the extracted influencer for cooperation, the promotion target can be promoted efficiently and effectively. The step (S20) is executed, for example, following the step (S16) as shown in FIG. 3.

[0052] According to the present embodiment, for example, by calculating the error between the number of encounters in the aggregated data and the predicted value, an influencer in a place other than social media can be extracted.

[0053] [Embodiment 2] An example of the processes related to the generation of aggregated data, the setting of a prediction formula and a standard error, the calculation of a predicted value, the calculation of the error between the number of encounters in the aggregated data and the predicted value, and the extraction of influencers will be described using the influencer extraction device 10 in FIG. 1. Note that these processes are not limited to the use of the influencer extraction device 10 in FIG. 1.

[0054] FIG. 4 is a diagram showing an example of the process related to the generation of aggregated data. An example of the visit information and the encounter information is shown at the upper part of FIG. 4. In the figure, "user" means the identification information of the visitor, and "date" means the respective time information. The aggregated data generation unit 12 generates aggregated data in which the number of visits and the number of encounters are aggregated for each visitor as shown in the middle part of FIG. 4 from the visit information and the encounter information. As described above, since there is a positive correlation between the number of visits and the number of encounters, a scatter diagram as shown in the lower part of FIG. 4 is generated from the aggregated data.

[0055] FIG. 5 is a diagram showing an example of the process related to the setting of a prediction formula and a standard error. The aggregated data shown in FIG. 4 is shown at the upper part of FIG. 5. The setting unit 13 can set a prediction formula and a standard error for the prediction formula by performing linear regression analysis, for example, using the number of encounters in the aggregated data as the target variable and the number of visits as the explanatory variable. In the middle part of FIG. 5, an example of the prediction formula and the standard error set based on the analysis result by performing linear regression analysis using the aggregated data shown at the upper part of FIG. 5 is shown. In other words, it can be said that the setting unit 13 performs a process of obtaining a linear approximation indicated by a dotted line in the scatter diagram shown in the lower part of FIG. 5, for example. In this example, the prediction formula is set by the approach of obtaining a linear approximation as described above, but this is an example, and the setting of the prediction formula by the setting unit 13 is not limited to this. The setting unit 13 may set the prediction formula by another method such as an approach of obtaining a curve approximation. Hereinafter, the description will be made assuming that the setting unit 13 sets the prediction formula by the approach of obtaining a linear approximation.

[0056] FIG. 6 is a diagram showing an example of a process related to the calculation of predicted values. The aggregated data shown in FIG. 6 is data obtained by adding the predicted values of the encounter times to the aggregated data shown in FIG. 4. The predicted values of the encounter times are calculated by the predicted value calculation unit 14 using the prediction formula shown at the lower part of FIG. 5.

[0057] FIG. 7 is a diagram showing an example of a process related to the calculation of the error between the encounter times and the predicted values in the aggregated data. The aggregated data shown in FIG. 7 is data obtained by adding the error between the encounter times and the predicted values to the aggregated data shown in FIG. 6. The error is calculated by the error calculation unit 15 using the standard error shown at the lower part of FIG. 5. Specifically, the error is calculated, for example, by the following formula (1) as also shown in FIG. 7. In other words, it can be said that the error calculation unit 15 performs a process of obtaining the difference between each plot and the linear approximation in the scatter diagram shown at the lower part of FIG. 7. Error = (number of encounters - predicted value of the number of encounters) / standard error ···(1)

[0058] For example, assuming that a preset threshold value is 1.5, the influencer extraction unit 16 extracts a visitor C with an error greater than 1.5 as an influencer from among a plurality of visitors in the aggregated data shown in FIG. 7.

[0059] [Embodiment 3] Based on the verification results implemented by the present inventors, the correlation between the number of visits and the number of encounters will be described.

[0060] In this verification, a recording device for recording the visitor information was installed at the entrance of the experimental site. The recording information is a device that records the identification information of the terminal when the terminal (smartphone) of the visitor is held up. Also, in this verification, visitors whose visit intervals were less than 5 minutes were judged to have "encountered" each other, and the number of visits and the number of encounters were calculated for each visitor. Then, the correlation coefficient between the number of visits and the number of encounters was obtained and the correlation was analyzed. Fig. 8 shows the result of analyzing the correlation between the number of visits and the number of encounters. As shown in Fig. 8, it was found that there is a strong positive correlation between the number of visits and the number of encounters.

[0061] [Embodiment 4] The program of this embodiment is a program for causing a computer to execute each step of the method of the present invention as a procedure. In the present invention, "procedure" may be read as "process". Also, the program of this embodiment may be recorded on, for example, a computer-readable recording medium. The recording medium is not particularly limited, and examples include a read-only memory (ROM), a hard disk (HD), and an optical disk.

[0062] The present invention has been described above with reference to the embodiments, but the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

[0063] [Supplementary Note] Some or all of the above embodiments may be described as follows in the supplementary note, but are not limited thereto. [Supplementary Note 1] including an acquisition unit, an aggregated data generation unit, a setting unit, a predicted value calculation unit, an error calculation unit, and an influencer extraction unit, the acquisition unit acquires visit information and encounter information, the visit information is information that associates identification information indicating an individual visitor with time information indicating the time when the visitor visited an arbitrary place, The encounter information is information that associates the identification information of each visitor who encountered at the arbitrary location with time information indicating the time of encounter. The aggregation data generation unit generates aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the visit information and the encounter information. The setting unit analyzes the aggregation data to set a prediction formula for calculating a predicted value of the number of encounters and a standard error. The predicted value calculation unit calculates a predicted value of the number of encounters for each visitor using the prediction formula. The error calculation unit calculates, for each visitor, an error between the number of encounters in the aggregation data and the predicted value based on the number of encounters in the aggregation data, the predicted value of the number of encounters, and the standard error. The influencer extraction unit is an influencer extraction device that extracts, as influencers, those visitors among a plurality of visitors whose error is equal to or greater than a preset threshold. (Appendix 2) The influencer extraction unit according to Appendix 1 excludes, from the extraction target, at least one of those visitors whose error is equal to or greater than a preset upper limit value and those visitors whose number of visits is equal to or greater than a preset lower limit value. (Appendix 3) Furthermore, it includes a filtering unit. The acquisition unit acquires at least one of visitable time zone information and attribute information. The visitable time zone information is information indicating a time zone in which the visitor can visit the arbitrary location. The attribute information is information indicating the attribute of the visitor. The filtering unit performs filtering on the visit information and the encounter information using at least one of the acquired visitable time zone information and the attribute information. The aggregation data generation unit according to Appendix 1 or 2 generates aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the filtered visit information and encounter information. (Appendix 4) Furthermore, it includes a recording unit, The recording unit records at least one of the visit information and the encounter information, and is the influencer extraction device according to any one of Appendices 1 to 3. (Appendix 5) Furthermore, it includes a congestion information disclosure unit, The congestion information disclosure unit generates congestion information indicating the number of visitors for each time period at the arbitrary location based on the visit information, and discloses the congestion information, and is the influencer extraction device according to any one of Appendices 1 to 4. (Appendix 6) Furthermore, it includes a notification unit, The notification unit notifies the visitor extracted as an influencer that the visitor has been extracted as an influencer, and is the influencer extraction device according to any one of Appendices 1 to 5. (Appendix 7) It includes an acquisition step, an aggregated data generation step, a setting step, a predicted value calculation step, an error calculation step, and an influencer extraction step. The acquisition step acquires visit information and encounter information. The visit information is information that associates identification information indicating an individual visitor with time information indicating the time when the visitor visited an arbitrary location. The encounter information is information that associates the identification information of each of the visitors encountered at the arbitrary location with time information indicating the encounter time. The aggregated data generation step generates aggregated data obtained by aggregating the number of visits and the number of encounters for each visitor based on the visit information and the encounter information. The setting step analyzes the aggregated data to set a prediction formula for calculating a predicted value of the number of encounters and a standard error. The predicted value calculation step calculates a predicted value of the number of encounters for each visitor using the prediction formula. The error calculation step calculates, for each visitor, the error between the number of encounters in the aggregated data and the predicted value based on the number of encounters in the aggregated data, the predicted value of the number of encounters, and the standard error. The influencer extraction step is an influencer extraction method that extracts, as influencers, those visitors among a plurality of visitors for whom the error is greater than or equal to a preset threshold value. (Appendix 8) The influencer extraction step excludes, from the extraction targets, at least one of those visitors for whom the error is greater than or equal to a preset upper limit value and those visitors for whom the number of visits is greater than or equal to a preset lower limit value, in the influencer extraction method described in Appendix 7. (Appendix 9) Furthermore, it includes a filtering step. The acquisition step acquires at least one of visitable time zone information and attribute information. The visitable time zone information is information indicating the time zones during which the visitor can visit the arbitrary location. The attribute information is information indicating the attributes of the visitor. The filtering step performs filtering on the visit information and the encounter information using at least one of the acquired visitable time zone information and the attribute information. The aggregation data generation step generates aggregation data in which the number of visits and the number of encounters are aggregated for each visitor, based on the filtered visit information and encounter information, in the influencer extraction method described in Appendix 7 or 8. (Appendix 10) Furthermore, it includes a recording step. The recording step records at least one of the visit information and the encounter information, in the influencer extraction method described in any one of Appendices 7 to 9. (Appendix 11) Furthermore, it includes a congestion information disclosure step. The congestion information disclosure step generates congestion information indicating the number of visitors for each time zone at the arbitrary location based on the visit information, and discloses the congestion information, in the influencer extraction method described in any one of Appendices 7 to 10. (Appendix 12) Furthermore, it includes a notification step. The notification process is the influencer extraction method according to any one of Appendices 7 to 11, which notifies the visitor extracted as an influencer that they have been extracted as an influencer. (Appendix 13) A program for causing a computer to execute procedures including an acquisition procedure, an aggregation data generation procedure, a setting procedure, a predicted value calculation procedure, an error calculation procedure, and an influencer extraction procedure: The acquisition procedure acquires visit information and encounter information, The visit information is information associating identification information indicating an individual visitor with time information indicating the time when the visitor visited an arbitrary location, The encounter information is information associating the identification information of each of the visitors encountered at the arbitrary location with time information indicating the encounter time, The aggregation data generation procedure generates aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the visit information and the encounter information, The setting procedure analyzes the aggregation data to set a prediction formula for calculating a predicted value of the number of encounters and a standard error, The predicted value calculation procedure calculates a predicted value of the number of encounters for each visitor using the prediction formula, The error calculation procedure calculates, for each visitor, an error between the number of encounters in the aggregation data and the predicted value based on the number of encounters in the aggregation data, the predicted value of the number of encounters, and the standard error, The influencer extraction procedure extracts, as influencers, those visitors among a plurality of visitors whose error is equal to or greater than a preset threshold value. (Appendix 14) The influencer extraction procedure excludes, from the extraction targets, at least one of those visitors whose error is equal to or greater than a preset upper limit value and those visitors whose number of visits is equal to or greater than a preset lower limit value, in the program described in Appendix 13. (Appendix 15) Furthermore, it includes a filtering procedure, The acquisition procedure acquires at least one of visitable time zone information and attribute information, The arrival possible time zone information is information indicating the time zones when the visitor can arrive at the arbitrary location, the attribute information is information indicating the attributes of the visitor, the filtering procedure performs filtering on the arrival information and the encounter information by using at least one of the obtained arrival possible time zone information and the attribute information, the aggregation data generation procedure generates aggregation data obtained by aggregating the number of arrivals and the number of encounters for each visitor based on the filtered arrival information and encounter information, the program described in Appendix 13 or 14. (Appendix 16) Furthermore, it includes a recording procedure, the recording procedure is a program described in any one of Appendices 13 to 15 that records at least one of the arrival information and the encounter information. (Appendix 17) Furthermore, it includes a congestion information disclosure procedure, the congestion information disclosure procedure generates congestion information indicating the number of visitors for each time zone at the arbitrary location based on the arrival information, and discloses the congestion information, the program described in any one of Appendices 13 to 16. (Appendix 18) Furthermore, it includes a notification procedure, the notification procedure is a program described in any one of Appendices 13 to 17 that notifies the visitor extracted as an influencer that they have been extracted as an influencer. (Appendix 19) A computer-readable recording medium recording a program described in any one of Appendices 13 to 18.

Industrial Applicability

[0064] According to the present invention, influencers in places other than social media can be extracted. Therefore, the present invention is useful, for example, when performing influencer marketing.

Explanation of Signs

[0065] 10 Influenza Influencer Extraction Device 11 Acquisition Unit 12 Aggregate Data Generation Unit 13 Setting Unit 14 Predicted Value Calculation Unit 15 Error Calculation Unit 16 Influenza Influencer Extraction Unit 17 Filtering Unit 18 Recording Unit 19 Congestion Information Disclosure Unit 20 Notification Unit 101 Central Processing Unit 102 Memory 103 Bus 104 Storage Device 105 Input Device 106 Display Device 107 Communication Device

Claims

1. including an acquisition unit, an aggregated data generation unit, a setting unit, a predicted value calculation unit, an error calculation unit, and an influencer extraction unit, wherein the acquisition unit acquires visit information, and based on the visit information, if there are other visitors within a certain time before and after the visit time of one visitor, acquires, as encounter information, the pair of the one visitor and the other visitors and the visit time of the one visitor, wherein the visit information is information associating identification information indicating an individual visitor with time information indicating the time when the visitor visited an arbitrary place, wherein the encounter information is information associating the identification information of each of the visitors who encountered at the arbitrary place with time information indicating the encounter time, wherein the aggregated data generation unit generates aggregated data obtained by aggregating the number of visits and the number of encounters for each visitor based on the visit information and the encounter information, wherein the setting unit analyzes the aggregated data to set a prediction formula for calculating a predicted value of the number of encounters and a standard error, wherein the predicted value calculation unit calculates a predicted value of the number of encounters for each visitor using the prediction formula, wherein the error calculation unit calculates, for each visitor, an error between the number of encounters in the aggregated data and the predicted value based on the number of encounters in the aggregated data, the predicted value of the number of encounters, and the standard error, and wherein the influencer extraction unit extracts, as influencers, the visitors among a plurality of visitors whose error is equal to or greater than a preset threshold value, an influencer extraction device.

2. The influencer extraction device according to claim 1, wherein the influencer extraction unit excludes at least one of the visitors whose error is equal to or greater than a preset upper limit value and the visitors whose number of visits is equal to or greater than a preset lower limit value from the extraction target.

3. further including a filtering unit, wherein the acquisition unit acquires at least one of visitable time zone information and attribute information, wherein the visitable time zone information is information indicating a time zone in which the visitor can visit the arbitrary place, wherein the attribute information is information indicating the attribute of the visitor, and wherein the filtering unit performs filtering on the visit information and the encounter information using at least one of the acquired visitable time zone information and the attribute information. The aggregation data generation unit generates aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the filtered visit information and encounter information. The influencer extraction device according to claim 1 or 2.

4. Furthermore, it includes a recording unit, The recording unit records at least one of the visit information and the encounter information. The influencer extraction device according to any one of claims 1 to 3.

5. Furthermore, it includes a congestion information disclosure unit, The congestion information disclosure unit generates congestion information indicating the number of visitors for each time zone at the arbitrary location based on the visit information, and discloses the congestion information. The influencer extraction device according to any one of claims 1 to 4.

6. Furthermore, it includes a notification unit, The notification unit notifies the visitor extracted as an influencer that the visitor has been extracted as an influencer. The influencer extraction device according to any one of claims 1 to 5.

7. It includes an acquisition step, an aggregation data generation step, a setting step, a predicted value calculation step, an error calculation step, and an influencer extraction step. The acquisition step is as follows: Acquire visit information, Based on the visit information, if there are other visitors within a certain time before and after the visit time of one visitor, acquire the pair of the one visitor and the other visitors and the visit time of the one visitor as encounter information. The visit information is information that associates identification information indicating an individual visitor with time information indicating the time when the visitor visited an arbitrary location. The encounter information is information that associates the identification information of each visitor who encountered at the arbitrary location with time information indicating the encounter time. The aggregation data generation step generates aggregation data obtained by aggregating the number of visits and the number of encounters for each visitor based on the visit information and the encounter information. The setting step analyzes the aggregation data and sets a prediction formula for calculating a predicted value of the number of encounters and a standard error. The predicted value calculation step calculates a predicted value of the number of encounters for each visitor using the prediction formula. The error calculation step calculates, for each visitor, the error between the number of encounters in the aggregation data and the predicted value based on the number of encounters in the aggregation data, the predicted value of the number of encounters, and the standard error. The influencer extraction step is an influencer extraction method in which, from among a plurality of visitors, those visitors for whom the error is greater than or equal to a preset threshold are extracted as influencers, and each step is executed by a computer.

8. The influencer extraction step excludes, from the extraction target, at least one of those visitors for whom the error is greater than or equal to a preset upper limit value and those visitors for whom the number of visits is greater than or equal to a preset lower limit value, according to the influencer extraction method described in claim 7.

9. Furthermore, it includes a filtering step. The acquisition step acquires at least one of visitable time zone information and attribute information. The visitable time zone information is information indicating the time zone during which the visitor can visit the arbitrary location. The attribute information is information indicating the attributes of the visitor. The filtering step performs filtering on the visit information and the encounter information using at least one of the acquired visitable time zone information and the attribute information. The aggregation data generation step generates aggregation data in which the number of visits and the number of encounters are aggregated for each visitor, based on the filtered visit information and encounter information, according to the influencer extraction method described in claim 7 or 8.

10. A program for causing a computer to execute a procedure including an acquisition procedure, an aggregation data generation procedure, a setting procedure, a predicted value calculation procedure, an error calculation procedure, and an influencer extraction procedure: The acquisition procedure is as follows. It acquires visit information. Based on the visit information, if there are other visitors within a certain time before and after the visit time of one visitor, it acquires, as encounter information, the pair of the one visitor and the other visitor and the visit time of the one visitor. The visit information is information associating identification information indicating an individual visitor with time information indicating the time at which the visitor visited an arbitrary location. The encounter information is information associating the identification information of each of the visitors who encountered each other at the arbitrary location with time information indicating the encounter time. The aggregation data generation procedure generates aggregation data in which the number of visits and the number of encounters are aggregated for each visitor, based on the visit information and the encounter information. The setting procedure analyzes the aggregation data and sets a prediction formula for calculating a predicted value of the number of encounters and a standard error. The predicted value calculation procedure calculates, for each visitor, the predicted value of the number of encounters using the prediction formula. Based on the number of encounters in the aggregated data, the predicted value of the number of encounters, and the standard error, the error calculation procedure calculates, for each visitor, the error between the number of encounters in the aggregated data and the predicted value. The influencer extraction procedure extracts, as influencers, those visitors among a plurality of visitors whose error is equal to or greater than a preset threshold value.

Images