Store visit determination method, program and system

Abstract

To provide a store visit determination method capable of stably and highly accurately determining whether a user who has accessed a website has visited the store while maintaining the anonymity of the user. Solution: The server compares the web access log (UUID + access information) with the store Wi-Fi detection log (MAC hash, etc.) based on the degree of agreement between time, location, and auxiliary information, and estimates the likelihood that a device linked to the UUID visited the store. The scoring logic uses multiple judgment factors.

Description

[Technical Field]

[0001] The present invention relates to a store visit determination method, a program, and a system for determining whether a user has visited a store. [Background technology]

[0002] Services such as MEO (Map Engine Optimization) measure the effectiveness of MEO by measuring whether users who viewed a store's web page visited the store. Services such as MEO have the ability to measure whether users who viewed a store's webpage visited the store. This measurement allows for a clear measurement of the effectiveness of MEO. Specifically, it tracks the behavior of users who accessed the store's webpage and uses various indicators to confirm whether they actually visited the store. Based on this data, stores can evaluate the effectiveness of their measures and determine which measures were particularly effective.

[0003] Services such as MEO play an important role in promoting business growth by utilizing data obtained through store web pages to accurately grasp customer trends. Store operators are expected to use this information to implement more effective measures and improve customer satisfaction. Measuring the effectiveness of MEO will become increasingly important in future business development.

[0004] To determine (confirm) whether a customer has visited a store, location information services using GPS functions are commonly used. The system obtains real-time location information from the user's smartphone and determines whether the user has arrived at the store. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2024-017897 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the above-mentioned store visit determination using the GPS function has a problem in terms of the accuracy of the location information. That is, the accuracy of GPS varies depending on the environment, and in urban areas in particular, signals can be disrupted by the influence of buildings. For this reason, it may not be possible to determine the user's exact location. As a result, it may be erroneously determined that a user has visited a store simply because they are nearby.

[0007] Additionally, GPS signals are weaker inside buildings and underground, making it difficult to determine whether someone has visited a store. Furthermore, it is difficult for GPS to determine which floor of a building a person is on. Furthermore, collecting user location information may be considered a violation of privacy and may require user consent, which could be a barrier to the widespread adoption of the service. There is also a risk that the collected location information may be accessed illegally.

[0008] The present invention aims to provide a store visit determination method, a program, and a system that can stably and accurately determine whether a user who accessed a website visited the store while maintaining the anonymity of the user. [Means for solving the problem]

[0009] The present invention includes a first step of acquiring first attribute information of a user terminal from the user terminal that has viewed a web page; a second step of generating user identification information that uniquely identifies a user or a user terminal device based on the first attribute information; a third step of storing in a storage means association information that associates the user identification information generated in the second step with the first attribute information acquired in the first step; a fourth step of acquiring second attribute information of the user terminal device from the user terminal device that has wirelessly communicated using a wireless communication router located in a predetermined store; and a device storing the second attribute information of a communication device located within the wireless communication range of the wireless communication router. A store visit determination method executed by a computer includes a fifth step of acquiring location information, a sixth step of comparing the second attribute information acquired in the fourth step with the second attribute information included in the device location information acquired in the fifth step, a seventh step of comparing the first attribute information included in the device location information that satisfies a predetermined first condition through the comparison in the sixth step with the first attribute information indicated by the association information to identify user identification information that satisfies a predetermined second condition, and an eighth step of determining that the user corresponding to the user identification information identified in the seventh step has visited the store.

[0010] Preferably, the first attribute information, the device location information, and the second attribute information are user terminal attribute information indicating attributes of the user terminal device that are made up of a plurality of common indicators.

[0011] Preferably, the sixth step uses the device location information in which time interval information indicating the time difference between the time when the user terminal device accessed the web page and the time when the user terminal device was detected by wireless communication via the wireless communication router satisfies certain conditions.

[0012] Preferably, the sixth step generates matching score information based on the time interval information, the radio wave strength of the wireless communication router received by the user terminal device, and the degree of match between the indices of the user terminal attribute information, and the seventh step determines whether the first condition is met based on the matching score information.

[0013] Preferably, the user terminal attribute information is information indicating at least one of the OS type, OS version, language setting, carrier, and access route information of the user terminal device.

[0014] Preferably, the first attribute information is associated with access time information when the user terminal device accessed the web page, the second attribute information acquired in the fourth step is associated with detection time information indicating the timing when the second attribute information was acquired from the user terminal device, and the fifth step acquires the device location information based on the access time information and the detection time information.

[0015] Preferably, the device location information includes at least one of router identification information of the wireless communication router and wireless communication reception strength.

[0016] Preferably, the computer executes a ninth step of determining whether the user who was determined to have visited the store in the eighth step will return or the number of visits to the store based on the user identification information of the user.

[0017] The present invention includes a first step of acquiring first attribute information of a user terminal from the user terminal that has viewed a web page; a second step of generating user identification information that uniquely identifies a user or a user terminal device based on the first attribute information; a third step of storing in a storage means association information that associates the user identification information generated in the second step with the first attribute information acquired in the first step; a fourth step of acquiring second attribute information of the user terminal device from the user terminal device that has wirelessly communicated using a wireless communication router located in a predetermined store; and a device storing the second attribute information of a communication device located within the wireless communication range of the wireless communication router. The program causes a computer to execute the following steps: a fifth step of acquiring location information; a sixth step of comparing the second attribute information acquired in the fourth step with the second attribute information included in the device location information acquired in the fifth step; a seventh step of comparing the first attribute information included in the device location information that satisfies a predetermined first condition through the comparison in the sixth step with the first attribute information indicated by the association information to identify user identification information that satisfies a predetermined second condition; and an eighth step of determining that a user corresponding to the user identification information identified in the seventh step has visited the store.

[0018] The present invention includes a first step of acquiring first attribute information of a user terminal from the user terminal that has viewed a web page; a second step of generating user identification information that uniquely identifies a user or a user terminal device based on the first attribute information; a third step of storing in a storage means association information that associates the user identification information generated in the second step with the first attribute information acquired in the first step; a fourth step of acquiring second attribute information of the user terminal device from the user terminal device that has wirelessly communicated using a wireless communication router located in a predetermined store; and a fourth step of storing a device containing the second attribute information of a communication device located within the wireless communication range of the wireless communication router. a fifth step of acquiring device location information; a sixth step of comparing the second attribute information acquired in the fourth step with the second attribute information included in the device location information acquired in the fifth step; a seventh step of comparing the first attribute information included in the device location information that satisfies a predetermined first condition through the comparison in the sixth step with the first attribute information indicated by the association information to identify user identification information that satisfies a predetermined second condition; and an eighth step of determining that the user corresponding to the user identification information identified in the seventh step has visited the store. [Effects of the Invention]

[0019] According to the present invention, it is possible to provide a store visit determination method, program, and system that can stably and accurately determine whether a user who accessed a website visited the store while maintaining the anonymity of the user. [Brief explanation of the drawings]

[0020] [Figure 1] FIG. 1 is a diagram illustrating a network environment in which a store visit determination system 11 according to an embodiment of the present invention is used. [Figure 2] FIG. 2 is a diagram for explaining the first attribute information, the second attribute information, and the device location information used by the store visit determination system 11. As shown in FIG. [Figure 3]FIG. 3 is a diagram for explaining the association information of this embodiment. [Figure 4] FIG. 4 is a flowchart illustrating the service start process. [Figure 5] FIG. 5 is a flowchart for explaining the operation of the store visit determination system 11 to acquire user identification information when the user terminal device 13 accesses a store's web page. [Figure 6] FIG. 6 is a flowchart for explaining the operation of the store visit determination system 11 to determine whether a user has visited a store based on the user identification information of the user terminal device 13 acquired at the store. [Figure 7] FIG. 7 is a functional block diagram of the user terminal device 13 shown in FIG. [Figure 8] FIG. 8 is a functional block diagram of the store visit determination system 11 shown in FIG. DETAILED DESCRIPTION OF THE INVENTION

[0021] The store visit determination system 11 according to the embodiment of the present invention will be described below. FIG. 1 is a diagram illustrating a network environment in which a store visit determination system 11 according to an embodiment of the present invention is used. As shown in FIG. 1, the store visit determination system 11 communicates with a user terminal device 13 and a store terminal device 15. The user terminal device 13 accesses the web page of the store operated by the user and displays it to the user. The user may browse the web page and visit the store. The store terminal device 15 communicates with the user terminal device 13 located within the store via a wireless communication router installed within the store.

[0022] The store visit determination system 11 is a system that acquires user web access logs and measures store visits. In particular, it does not rely on GPS but uses web technology to more accurately determine user store visits and provides a highly anonymous service.

[0023] The service provider of the store visit determination system 11 concludes a contract for the store visit determination service with, for example, a store operator that operates the store terminal device 15. The store visit determination system 11 acquires first attribute information from the user terminal 13 that has accessed a web page provided by a store operator, and generates user identification information based on the first attribute information. The store visit determination system 11 may directly acquire this information, or may receive information acquired by another device.

[0024] Specifically, the store visit determination system 11 generates a visitorId (UUID) as user identification information using, for example, FingerprintJS.

[0025] FIG. 2 is a diagram for explaining the first attribute information, the second attribute information, and the device location information used by the store visit determination system 11. As shown in FIG. As shown in FIG. 2, the first attribute information includes user terminal attribute information indicating at least one of the OS type, OS version, language setting, carrier, and access route information of the user terminal device 13. Here, the access route is, for example, ( ). The first attribute information also includes time information when the user terminal device 13 accessed the web page. In addition, the first attribute information includes, for example, the browser's User-Agent information, OS version, language settings, screen resolution, color depth, whether touch support is available, time zone, list of browser plug-ins, list of fonts, WebGL and Canvas rendering information, AudioContext information related to acoustic fingerprints, device memory capacity, number of CPU cores, remaining battery power and charging status (if permitted), whether local storage or session storage is available, whether IndexedDB is available, IP address, click source URL (referrer), and visit date and time (timestamp). This will enable data collection within the scope of laws and regulations, including various fingerprinting information and device characteristic information that will be available in the future.

[0026] The above-mentioned information is acquired by the store visit determination system 11 by including tag information (measurement tag) for acquiring information in advance on a web page provided by the store, or by other methods.

[0027] FingerprintJS is a library for generating browser fingerprints to identify users. In this embodiment, this technology is used to generate a visitorId, which is a unique identifier (UUID) for each user who accesses a store's web page, and identify the user.

[0028] FingerprintJS combines browser information (such as user agent, resolution, and plugin information) to generate a unique fingerprint.

[0029] 3, association information indicating the association between the first attribute information acquired from the store visit determination system 11 and the user identification information thereof is stored in a storage means in a structured data format such as JSON via a lightweight API or a data receiving module. This data is stored on a VPS or an external data storage service, and is stored in a database or storage structure that allows for chronological management.

[0030] A wireless communication router is installed in each store, and second attribute information of user terminal devices 13 within the range of the router is transmitted to the store visit determination system 11 via the store terminal devices 15 .

[0031] As shown in FIG. 2, the second attribute information includes user terminal attribute information indicating at least one of the OS type, OS version, language setting, carrier, and access route information of the user terminal device 13. The second attribute information also includes information about the detection time when the wireless communication router detected the user terminal device 13. The second attribute information also includes a BSSID detection log (hashed terminal identifier+RSSI+detection time).

[0032] The store visit determination system 11 may acquire information from a company that provides data indicating that Wi-Fi has been received. The wireless communication router identifies the user terminal device 13 based on the MAC address and IP address.

[0033] The store visit determination system 11 uses the device location information shown in FIG. 3 to determine whether the user of the user terminal device 13 has visited the store. As shown in Figure 3, the device location information includes router identification information of the wireless communication router and terminal information indicating at least one of the OS type, OS version, language setting, carrier, and access route information of the communication terminal that received radio waves from the communication router. Specifically, the device location information indicates at least one of the SSID, BSSID, wireless communication reception strength, terminal model information, terminal setting information, application name, and terminal connection carrier information.

[0034] The store visit determination system 11 acquires device location information from, for example, a predetermined service provider. At that time, the store visit determination system 11 requests and acquires device location information within the specified range, for example, by specifying the identification information of the communication router and the communication terminals that have detected radio waves within a predetermined period (third condition). This makes it possible to limit the range of device location information, reduce the cost of acquiring information, and perform matching within an appropriately narrowed range.

[0035] The store visit determination system 11 uses, for example, device location information in which time interval information indicating the time between the time when the store visit determination system 11 accessed a web page and the time when it was detected by wireless communication via a wireless communication router satisfies certain conditions. Specifically, after accessing a web page, Wi-Fi detection logs within an attribution window determined by industry are considered candidates. For example, within 72 hours for the food and beverage industry, within 7 days for retail, and within 14 days for beauty and medical care. At this point, logs outside of this period are excluded from matching.

[0036] The store visit determination system 11 compares the second attribute information acquired from the user terminal device 13 in the store with the second attribute information included in the device location information. Then, the store visit determination system 11 compares the first attribute information of the device location information that satisfies the specified first condition through the matching with the first attribute information of the association information described above, and identifies the user identification information associated with the first attribute information that satisfies the specified second condition from the association information. Then, it is determined that the user corresponding to the specified user identification information has visited the store.

[0037] The store visit determination system 11 performs the check against the first condition as follows. Based on the first attribute information and device location information, the store visit determination system 11 generates the time difference between the time when the user viewed the web page (the time when the user terminal device 13 accessed it) and the time detected by the user terminal device 13 via wireless communication using the wireless communication router. The store visit determination system 11 identifies the radio wave intensity of the wireless communication router detected by the user terminal device 13 based on the device location information. The store visit determination system 11 generates a degree of match between the user terminal attribute information in the second attribute information acquired from the user terminal device 13 at the store and each index of the user terminal attribute information in the device location information. For example, the degree of match is increased if there are many matching indexes.

[0038] The store visit measurement logic adopted by the store visit determination system 11 will be described below. The store visit determination system 11 performs matching using a plurality of matching keys, and performs partial matching and similarity scoring using UUID, MAC hash, etc. Additionally, user agent, operating system, screen size, font, time zone, acoustic fingerprint, etc. are used as supplementary information.

[0039] The store visit determination system 11 uses, for example, an RSSI distance estimation model to sessionize the stay time, and applies a two-way time difference model between access and store visit. In addition, weighted scoring is performed and optimization is achieved through machine learning. The system is designed to dynamically change conditions and thresholds based on external factors such as season, time of day, day of the week, weather, and surrounding events, depending on the industry, business type, and situation.

[0040] Regarding the criteria for determining scores by industry, for restaurants, a visit within a few hours will be considered a high score, while for beauty salons, a visit within a few days will be considered a high score. In retail stores, conditions can be set according to the characteristics of each industry, such as a customer visiting within the last few days and spending less than five minutes in the store being considered to have a high score.

[0041] The store visit determination system 11 generates matching score information by assigning predetermined weights to the generated time difference, radio wave intensity, and degree of match. Then, it is determined whether a predetermined first condition is satisfied based on the matching score information. For example, it is determined that the first condition is satisfied when the value of the matching score information exceeds a predetermined value.

[0042] An example of a method for calculating the inquiry score information will be described below. The following features are calculated for each candidate log. ·Time consistency (S_time) The time difference between web access and store visit detection is exponentially decayed and evaluated. S_time = exp(-(t - T0) / τ) (τ is a parameter that represents the time characteristics of each industry)

[0043] ·Position matching (S_loc) Specify the BSSID of the Wi-Fi router installed in the store as a fixed key. Only the logs (purchases) of devices that detect the BSSID in question are obtained from the data provider. Therefore, S_loc simply determines "detected = 1, not detected = 0". Furthermore, this matching determination can be combined with RSSI and length of stay to further strengthen the probability of store visits.

[0044] The reasons for using BSSID as a key are as follows: (1) Guaranteed accuracy: BSSID has a unique value for each store, allowing stores to be identified at the building level more accurately than GPS. (2) Feasibility of data provision: Data providers can extract and provide only the logs of devices that detect the specified BSSID, which is efficient and does not involve handling unnecessary nearby Wi-Fi data. (3) Versatility and scalability: Since it is provided as standard with general Wi-Fi devices, no special installation is required. Multiple BSSIDs can be set up even on multiple floors or in large facilities.

[0045] ·Signal strength / stay (S_rssi, S_dwell) Based on the RSSI value and continuous detection time, it is evaluated whether the person has stayed in the store for a certain period of time. Example: Evaluating RSSI values ​​by normalizing them with a threshold S_rssi = min(1, max(0, (RSSI - RSSI_min) / (RSSI_max - RSSI_min))) In addition, if the stay time is long, S_dwell is calculated as an additional point factor.

[0046] Device attribute matching rate (S_dev) The device attributes (OS type and version, language settings, carrier information, etc.) obtained from both the web log and the Wi-Fi log are compared to calculate the degree of match. Example: S_dev= (siguma_i match(attribute_i)) / N Here, match(attribute_i) gives a score of 1 for a match, 0 for a mismatch, and 0.5 for a partial match, and N is the total number of attributes being compared.

[0047] The store visit determination system 11 assigns a weight to each feature and calculates the matching score information (Score: overall score). Score=w1*S_time+w2*S_loc+w3*S_rssi+w4*S_dev+w5*S_intent Here, w1 to w5 can be adjusted based on the characteristics of each industry and product. Finally, if the score exceeds the set threshold, it is determined that the visit was made by the same person.

[0048] In this embodiment, a UUID (Universal Unique Identifier) ​​is used as the user identification information. A UUID is an identifier used to uniquely identify a specific device or user. Using UUIDs, it is possible to aggregate and analyze data related to a particular device or user.

[0049] The UUID is obtained through a specific process when a terminal device browses a web page, which is explained below. Loading a web page: When a user visits a web page through a browser, the server receives the request.

[0050] Using FingerprintJS: UUIDs are typically generated using libraries like FingerprintJS, which is a technology that generates a unique ID based on the characteristics of a user's device and browser. The library generates a UUID by collecting the following information: Browser User-Agent information, OS version, language settings, screen resolution, touch support, font list, and other device characteristics

[0051] Generate a UUID: Based on the collected information, FingerprintJS generates a UUID that is associated with a specific device and browser and distinguishes it from other users.

[0052] UUIDs are a key element in tracking user behavior while maintaining anonymity, allowing data to be collected without identifying the user.

[0053] The flow and operation of the store visit determination service using the store visit determination system 11 will be described below. <Service start process> FIG. 4 is a flowchart illustrating the service start process. Step ST11: The operator of the store visit determination system 11 concludes a contract for the service with the store operator who will receive the store visit determination service.

[0054] Step ST12: The store visit determination system 11 registers necessary information related to the store operator, such as the web page, the store address, and the store's wireless communication router information.

[0055] Step ST13: This involves tasks such as embedding tags for acquiring information on the web pages of stores provided by store operators.

[0056] Step ST14: Information about the wireless communication router in the store is registered to be used to determine whether or not a user should visit the store.

[0057] <Getting access information to web pages> FIG. 5 is a flowchart for explaining the operation of the store visit determination system 11 to acquire user identification information when the user terminal device 13 accesses a store's web page. Each step will be explained.

[0058] Step ST21: When the user operates the user terminal device 13 to access a web page provided by the store and browses the web page, the process proceeds to step ST12.

[0059] Step ST22: The store visit determination system 11 acquires the above-mentioned first attribute information from the user terminal device 13 that has accessed the web page provided by the store operator, and stores it in memory.

[0060] Step ST23: The store visit determination system 11 generates user identification information based on the first attribute information acquired in step ST22. The store visit determination system 11 associates access time information indicating the timing at which the user terminal device 13 accessed the web page with the first attribute information.

[0061] Step ST24: The store visit determination system 11 stores association information indicating the association between the first attribute information acquired in step ST22 and the user identification information generated in step ST23 in a structured data format such as JSON on the VPS or in an external data storage service via a lightweight API or a data receiving module. That is, the association information is stored in a database or storage structure that allows for chronological management.

[0062] <Store visit determination operation> FIG. 6 is a flowchart for explaining the operation of the store visit determination system 11 to determine whether a user has visited a store based on the user identification information of the user terminal device 13 acquired at the store. Each step will be explained.

[0063] Step ST31: A wireless communication router installed in a store transmits wireless signals in a specific frequency band that is designed to cover the entire store.

[0064] Step ST32: When a user carrying user terminal device 13 enters a store, store terminal device 15 acquires the second attribute information from user terminal device 13 via a wireless communication router. Store visit determination system 11 associates detection time information indicating the timing at which the second attribute information was acquired from user terminal device 13 with the second attribute information.

[0065] Step ST33: The store visit determination system 11 requests a data provider to acquire the device location data shown in FIG. 2 described above. At this time, the store visit determination system 11 specifies a range that satisfies a third condition and acquires device location data within that range. The third condition is, for example, identification information of a wireless communication router and a condition that the device is a communication terminal that has detected radio waves within a predetermined period.

[0066] Step ST34: The store visit determination system 11 compares the second attribute information acquired from the user terminal device 13 in the store with the second attribute information included in the device location information, and determines whether the first condition is satisfied. The matching is performed by generating matching score information, for example, as described above.

[0067] Step ST35: If the store visit determination system 11 determines in step ST34 that the first condition is satisfied, the process proceeds to step ST35.

[0068] Step ST36: The store visit determination system 11 collates (compares) the first attribute information of the device location information with the first attribute information of the association information described above, and determines whether or not there is a user terminal device 13 that satisfies the second condition. The second condition is, for example, a condition that a degree of coincidence between a plurality of indices is generated and the degree of coincidence exceeds a certain standard.

[0069] Step ST37: If the store visit determination system 11 determines in step ST36 that the second condition is satisfied, the process proceeds to step ST38.

[0070] Step ST38: The store visit determination system 11 identifies, from the association information, the user identification information associated with the first attribute information used in the matching that satisfies the second condition. Then, it is determined that the user corresponding to the specified user identification information has visited the store.

[0071] As described above, the store visit determination system 11 can measure store visits anonymously without relying on GPS. This process protects user privacy while collecting store visit data, which can be used to improve marketing and store operations. Stores can also obtain more accurate store visit data, allowing them to collect valuable information for analyzing customer trends.

[0072] In other words, the store visit determination system 11 can provide a user trajectory scoring method between the web and physical stores that does not rely on GPS, a hybrid terminal tracking analysis method that combines fingerprint and Wi-Fi logs, store visit measurement logic using anonymous UUIDs, a dynamic scoring condition setting method according to industry and external factors, and a multi-scoring model that integrates complex determination elements. In addition, the store visit determination system 11 compares the web access log (UUID + access information) with the store Wi-Fi detection log (MAC hash, etc.) on the server side based on the degree of agreement between time, location, and auxiliary information, and estimates the possibility that a terminal linked to the UUID has visited the store. This design allows for the UUID to be reset or lost in combination with other auxiliary information. This allows for a certain level of accuracy to be maintained, while also providing the flexibility to dynamically adjust weighting and thresholds depending on the industry and external factors.

[0073] In this way, the store visit determination system 11 combines different indicators rather than a single key such as a UUID, and scores partial matches probabilistically. Furthermore, the scoring model can dynamically change thresholds and weights depending on the industry and situation.

[0074] The store visit determination system 11 does not completely rely on a single key such as a UUID or advertising ID, but estimates the "probability that it is the same person" by combining multiple features. In its current implementation, it uses the UUID, terminal attributes, and access route on the web side, and the BSSID detection log on the store side (hashed terminal identifier + RSSI + detection time), and does not handle raw data that can directly identify an individual (such as name or phone number). This makes it possible to estimate store visits while ensuring anonymity, and complies with regulations such as the Personal Information Protection Act and GDPR.

[0075] The store visit determination system 11 can be implemented using web technology alone, without relying on GPS or dedicated apps. It works on major browsers and achieves highly accurate store visit measurement while maintaining anonymity and preventing users from being identified. It can also measure store visits using UUIDs and auxiliary identifiers even in environments where advertising IDs cannot be obtained, improving reliability through scoring based on multiple indicators. Furthermore, it has the flexibility to dynamically change conditions depending on the industry, season, weather, and time of day, and is scalable to accommodate future external data integration and the application of machine learning models.

[0076] As mentioned above, the store visit determination system 11 employs multiple determination elements in its scoring logic. It may also employ a dynamic condition change method based on the type of business or external factors. It can be implemented using JavaScript on a web browser without relying on GPS or a dedicated app, and can flexibly accommodate future expansion.

[0077] <User terminal device 13> FIG. 7 is a functional block diagram of the user terminal device 13 shown in FIG. As shown in FIG. 7, the user terminal device 13 includes, for example, a display 51, an operation unit / input unit 54, a communication unit 55, a memory 59, and a processing unit 61.

[0078] The display 51 displays an image based on a signal from the processing unit 61 . The communication unit 55 communicates with the store visit determination system 11. The operation unit / input unit 54 is an operation means such as a touch panel, a keyboard, or a mouse. The memory 59 stores the program executed by the processing unit 61 . The processing unit 61 executes the program PRG stored in the memory 59 to perform the above-described processing of the user terminal device 13 defined in this embodiment.

[0079] <Store Visit Judgment System 11> FIG. 8 is a functional block diagram of the store visit determination system 11 shown in FIG. As shown in FIG. 8, the store visit determination system 11 includes, for example, a communication unit 75, an input unit 77, a memory 79, and a processing unit 81.

[0080] The communication unit 75 communicates with the user terminal device 13 and the store terminal device 15 . The input unit 77 is a terminal or the like for inputting data from the outside. The memory 79 stores the program executed by the processing unit 81 . The processing unit 81 executes the program PRG stored in the memory 79 to perform the processing of the store visit determination system 11 defined in this embodiment.

[0081] Modifications of the above-described embodiment will now be described. For example, in this embodiment, the user identification information of a user who is determined to have visited the store through the above-mentioned process may be saved, and the user's re-visit or number of visits may be determined based on the user identification information identified in subsequent determinations.

[0082] At this time, the following determination may be made together with the above-mentioned matching score information. If there is a UUID match, the probability of visit is "high" Even if the UUID does not match, if multiple other keys are available, the probability of the visit is "medium" If only a partial match occurs, the probability of a store visit is "low" An example of a scoring formula: Matching score information = (UUID match x 0.5) +(MAC hash match x 0.3) +(UA similarity x 0.1) +(IP match x 0.1) Example of judgment criteria: · Score ≧ 0.7 → High probability of visit ·0.4≦score<0.7→medium probability of visit Score < 0.4 → Low probability of visit

[0083] In addition, the target for obtaining user identification information (UUID) may be expanded as follows: In the above-described embodiment, the case where the measurement tag is directly embedded in the target page of the store and the access information and the UUID are acquired has been exemplified. Other methods include measuring access from official social media links (YouTube, TikTok, Instagram, X, LINE official accounts, etc.), measuring access via email distribution or push notifications, measuring access via QR codes (registered trademark) displayed on in-store posters, flyers, electronic signage, etc., measuring access by providing tags to other companies' media or external sites, and obtaining UUIDs from advertising surfaces (display ads or native ads).

[0084] In addition, in generating the score data mentioned above, examples of scoring based on UUID and auxiliary identifiers, RSSI (radio signal strength) and length of stay, time difference from access to store visit, and weighting by industry / business type were given. In addition, the determination factors can be enhanced by high-precision distance estimation based on a combination of RSSI values ​​from multiple Wi-Fi access points and duration of stay, and by diversifying auxiliary identifiers (font set, time zone, acoustic fingerprint, sensor data, etc.).

[0085] Furthermore, the model may be expanded to include analysis of repeat and withdrawal patterns in addition to store visits. In addition, for optimization linked to external factors, thresholds and weights for determining whether a customer will visit the store may be dynamically adjusted depending on the season, time of day, day of the week, weather, surrounding events, traffic conditions, etc. Furthermore, the weights and thresholds of the determination factors may be automatically adjusted by machine learning or continuous learning. AI can also be used to build models for predicting the probability of store visits, the average cost per visit, and repeat visits. It is also possible to simultaneously calculate multiple KPIs, such as a store visit probability score, a repeat visit probability score, and a store visit unit price prediction score, and use a composite score index.

[0086] Regarding visualization and analysis functions, data sets including basic indicators are automatically output from the system. No tools are provided to customers; instead, analysts analyze and interpret the data and provide it as a report. Visualization and analysis may be performed as follows: Create a dashboard (daily, weekly, monthly). Generate KPIs such as estimated number of store visits, store visit rate, repeat visit rate, median time to store visit, and contribution by campaign. Also, filter by period, store, industry, business type, campaign, channel (SNS, email, QR, advertising), etc. Generate role-specific views such as management (summary), marketing (measure comparison), and field (progress by store).

[0087] Visualize visitor density using a map and time period (switch between weekdays / weekends and time periods), and generate spatiotemporal heat maps with overlays of nearby events and weather (visual confirmation of correlation).

[0088] "Access → Visit" funnel / contribution by touchpoint (SNS / advertising / QR code, etc.), linking and comparing with MEO measures / advertising initiatives (campaign A vs. B), etc. (funnel & attribution). Generate benchmarks for the repeat visit rate of first-time visit week cohorts, the distribution of visit intervals, and re-visit curves by industry and business type (cohort and repeat analysis).

[0089] Sudden rises / falls in visitor rates, deviations in RSSI distribution, and abnormalities in length of stay are automatically detected and Slack / email notifications are sent, and thresholds are automatically adjusted taking into account season, weather, and events (anomaly detection and alerts). Automatically output weekly and monthly PDFs / slides (insight summary + notes on significant changes), attach "summary by industry / business type" and "drill-down by store" with one click (automatic report generation).

[0090] Export to CSV / BigQuery / Redshift, integrate with BI (Looker / Power BI / Tableau), and use API to integrate and export data such as feeds to in-house DWH and advertising platforms. Role-based permissions, store-level permissions, display only anonymous IDs / sampling / minimization of aggregation units (permissions / anonymization governance)

[0091] Furthermore, the analysis results of the store visit determination system 11 may be utilized in combination with external data as follows (linkage). For example, POS and sales data integration such as linking purchase amounts and purchased product categories on a UUID or store visit measurement basis, and automatically calculating store visit conversion rates. In addition, CRM and membership data integration is also available, including matching store visit data with customer ID and customer rank, and analyzing the frequency of visits by loyal customers.

[0092] This involves linking data within a facility and with other stores, such as statistics on visits to commercial facilities, analysis of traffic patterns between multiple tenants, and collating data from other stores to analyze competitive and synergistic effects. It also integrates external location and traffic data such as trends in the use of transportation IC cards, surrounding traffic volume sensors, and correlation analysis between peak visits and peak traffic. It also works with weather APIs and event calendars to incorporate weather and event data into store visitor prediction models and automatically calculate the contribution of external factors.

[0093] This is an advertising distribution data integration that obtains impression and click data from DSPs / ad networks and scores store visit effects by advertising campaign. This is a two-way API integration that allows for real-time store visit determination requests from external systems and feeds analysis results to external systems and BI tools.

[0094] In this embodiment, as described above, FingerprintJS generates UUIDs, acquires browser information, compares it with Wi-Fi logs, and performs scoring using APIs and databases on the VPS. A wide variety of identification indices are also acquired.

[0095] The present invention may be as follows. In other words, the UUID generation logic may be strengthened by avoiding collisions through weighted hashing of multiple indicators, or by implementing update and reissue rules that take into account changes over time. In addition, the distance measurement algorithm may be improved by integrating and weighting the RSSI of multiple Wi-Fi access points, or by using a machine learning RSSI-to-distance conversion model.

[0096] Scoring processing can be made more efficient by using in-memory processing for real-time judgments or by introducing a distributed processing platform to speed up large-volume data processing. Privacy protection techniques such as data processing using differential privacy, dynamic change of aggregation units, and sampling rate adjustment may also be implemented.

[0097] Additionally, you can perform browser-resident measurements by converting to a Progressive Web App, or you can perform PWA / resident measurements by sending measurement events in the background. In addition, automatic tuning of the score determination logic using a machine learning model or model re-learning incorporating external factors (weather, events, seasonal fluctuations) may be performed.

[0098] The present invention is not limited to the above-described embodiments. That is, those skilled in the art may make various modifications, combinations, subcombinations, and substitutions of the components of the above-described embodiments within the technical scope of the present invention or its equivalents. [Industrial Applicability]

[0099] The present invention is applicable to a store visit determination system. [Explanation of symbols]

[0100] 11...Store visit determination system 13...User terminal device 15...Store terminal equipment

Claims

1. a first step of acquiring first attribute information of a user terminal from the user terminal that has viewed a web page; a second step of generating user identification information that uniquely identifies a user or a user terminal device based on the first attribute information; a third step of storing, in a storage means, association information that associates the user identification information generated in the second step with the first attribute information acquired in the first step; a fourth step of acquiring second attribute information of the user terminal device from the user terminal device that has wirelessly communicated with the user terminal device via a wireless communication router installed in a predetermined store; a fifth step of acquiring device location information including the second attribute information of a communication device located within a wireless communication range of the wireless communication router; a sixth step of comparing the second attribute information acquired in the fourth step with the second attribute information included in the device location information acquired in the fifth step; a seventh step of comparing the first attribute information included in the device location information that satisfies a predetermined first condition through the comparison in the sixth step with the first attribute information indicated by the association information to identify user identification information that satisfies a predetermined second condition; an eighth step of determining that the user corresponding to the user identification information identified in the seventh step has visited the store; The computer executes Visit determination method.

2. The first attribute information, the device location information, and the second attribute information are User terminal attribute information indicating attributes of the user terminal device consisting of a plurality of common indicators. The store visit determination method according to claim 1 .

3. The sixth step comprises: The device location information is used, and time interval information indicating the time difference between the time when the user terminal device accessed the web page and the time when the user terminal device was detected by wireless communication using the wireless communication router satisfies a certain condition. The store visit determination method according to claim 2 .

4. The sixth step comprises: the time interval information; The radio wave intensity of the wireless communication router received by the user terminal device; The degree of match of the index of the user terminal attribute information; Generate matching score information based on The seventh step comprises: It is determined whether the first condition is satisfied based on the matching score information. The store visit determination system according to claim 3 .

5. The user terminal attribute information is information indicating at least one of the OS type, OS version, language setting, carrier, and access route information of the user terminal device. The store visit determination method according to claim 4.

6. the first attribute information is associated with access time information when the user terminal device accesses the web page; the second attribute information acquired in the fourth step is associated with detection time information indicating a timing at which the second attribute information was acquired from the user terminal device; The fifth step acquires the device location information based on the access time information and the detection time information. The store visit determination method according to claim 5 .

7. The device location information includes at least one piece of information about the router identification information of the wireless communication router and the strength of the wireless communication reception. The store visit determination method according to claim 6.

8. a ninth step of determining whether the user has visited the store again or the number of times the user has visited the store based on the user identification information of the user determined to have visited the store in the eighth step; The store visit determination method according to claim 7 , which is executed by the computer.

9. a first step of acquiring first attribute information of a user terminal from the user terminal that has viewed a web page; a second step of generating user identification information that uniquely identifies a user or a user terminal device based on the first attribute information; a third step of storing, in a storage means, association information that associates the user identification information generated in the second step with the first attribute information acquired in the first step; a fourth step of acquiring second attribute information of the user terminal device from the user terminal device that has wirelessly communicated with the user terminal device via a wireless communication router installed in a predetermined store; a fifth step of acquiring device location information including the second attribute information of a communication device located within a wireless communication range of the wireless communication router; a sixth step of comparing the second attribute information acquired in the fourth step with the second attribute information included in the device location information acquired in the fifth step; a seventh step of comparing the first attribute information included in the device location information that satisfies a predetermined first condition through the comparison in the sixth step with the first attribute information indicated by the association information to identify user identification information that satisfies a predetermined second condition; an eighth step of determining that the user corresponding to the user identification information identified in the seventh step has visited the store; A program that causes a computer to execute the following.

10. a first step of acquiring first attribute information of a user terminal from the user terminal that has viewed a web page; a second step of generating user identification information that uniquely identifies a user or a user terminal device based on the first attribute information; a third step of storing, in a storage means, association information that associates the user identification information generated in the second step with the first attribute information acquired in the first step; a fourth step of acquiring second attribute information of the user terminal device from the user terminal device that has wirelessly communicated with the user terminal device via a wireless communication router installed in a predetermined store; a fifth step of acquiring device location information including the second attribute information of a communication device located within a wireless communication range of the wireless communication router; a sixth step of comparing the second attribute information acquired in the fourth step with the second attribute information included in the device location information acquired in the fifth step; a seventh step of comparing the first attribute information included in the device location information that satisfies a predetermined first condition through the comparison in the sixth step with the first attribute information indicated by the association information to identify user identification information that satisfies a predetermined second condition; an eighth step of determining that the user corresponding to the user identification information identified in the seventh step has visited the store; A store visit determination system that executes the above.

Images