RU2767958C2 - Method and system for determining fact of user visiting a point of interest - Google Patents

Abstract

FIELD: computer technology.SUBSTANCE: identifier of a user’s wireless device is received, and, based on it, a set of features of local communication of a local communication module of the wireless device is determined. These features are entered into a machine learning algorithm (hereinafter – MLA) trained using a set of heuristic rules applied to training sets of features of local communication of training wireless devices. The MLA algorithm, based on the set of features of local communication of the local communication module of the wireless device, issues an indication of the fact of the user visiting a POI point.EFFECT: increase in the accuracy of determining the fact of the user visiting a point of interest.20 cl, 8 dwg

Description

The technical field to which the invention belongs

[1] The present technology relates to computer methods and systems for determining whether users have visited points of interest.

State of the art

[2] Wireless devices such as Wi-Fi™-enabled smartphones and tablets are widely used by users. Many such wireless devices contain one or more geolocation systems, such as a global positioning system (GPS, Global Positioning System), to determine their geographic location (hereinafter referred to as geolocation).

[3] For various purposes and applications, it may be desirable to determine users' visits to different points of interest (POIs) based on users' geolocation, either in real time or by processing available geolocation tracking data provided by users' wireless devices. Examples of POIs include parks or other places of interest, businesses, private homes, and so on. In some cases, it may be desirable to determine the name and/or other information associated with each particular POI visited by users based on users' geolocation, either in real time or by processing available geolocation tracking data provided by users' wireless devices.

[4] In at least some cases, known geolocation systems, including geolocation systems of the wireless devices themselves, are suitable for these purposes. For example, many well-known geolocation systems are generally suitable for determining a user's visit to a large park that has few or no other POIs nearby. In this case, the wireless device can determine its geolocation with a relatively high level of confidence compared to the situation when the same device is located in a dense urban area with a large number of POIs nearby.

[5] However, in other situations known geolocation systems may be inaccurate and in some cases may simply be unusable. For example, the denser the area visited by the user, in terms of the number of POIs in that area, the less reliable known geolocation systems are due, at least in part, to the lower accuracy of geolocation provided by such geolocation systems and due to the greater number of possible POIs that the user can visit in this location.

[6] Some geolocation systems have been developed to overcome at least some of the shortcomings described above.

[7] For example, US9639858B2 (Systems and methods to attribute real-world visits of physical business locations by a user of a wireless device to targeted digital content or publicly displayed physical content previously viewable by the user, Facebook Inc.) describes methods and systems for registering the user's location and sending targeted content to the user based on the user's current and past location information. The network may include a programmed server with a database of targeted content, a database of location information, a database of user information, a database search algorithm, and a wireless communication system capable of communicating with a user's mobile device. The location of the mobile device is determined and registered. The location information is analyzed to determine the routes the user has taken, the businesses visited by the user, and other user behavior. Targeted content is sent to the user's mobile device or the visibility of physical content is tracked. Monitors user visits to physical locations associated with the content. Detailed conversion tracking data is provided to targeted content providers and business owners.

[8] Another example is the patent US9135655B2 (Systems and methods for using server side cookies by a demand side platform, MediaMath Inc.), which describes how a user is identified by a demand side platform (DSP, Demand Side Platform) on advertising exchanges. The method includes allowing the DSP to associate cookies with the user. Cookie matching involves matching user IDs for ad exchanges with a user ID assigned by the DSP for that user. The DSP platform stores in the cookie relationship the first relationship with the user ID of the DSP platform, containing the first user ID received by the bidder from the first exchange and the first exchange ID for the first exchange. The bidder inserts the so-called. "pixel" in the application for the possibility of display on the second exchange. The "pixel" contains a key for the cookie ratio, a second user ID, and a second exchange ID. The second user ID is obtained by the bidder from the second exchange.

[9] Another example is patent application US20090234745A1 (Methods and systems for mobile coupon tracking, Millennial Media LLC), which describes a method and system for providing a sponsored mobile coupon to a mobile communication device, at least in part, based on relevance , which, at least in part, is based on the characteristics of the mobile subscriber; redemption of the coupon at the location of the sponsor without connecting to the Internet using a mobile phone; registering the coupon conversion in the conversion data repository; sending the conversion data repository to the mobile operator; and parsing the conversion data repository to determine the action.

[10] Although the systems and methods described above may be suitable for at least some of the intended applications, they have certain disadvantages.

Disclosure of invention

[11] The present technology is designed to address at least some of the shortcomings of known geolocation systems for at least some specific applications.

[12] Briefly, the present technology provides methods and systems for determining whether identified or identifiable users have visited identified or identifiable POIs. Methods and systems use a specially configured machine learning algorithm (MLA, Machine Learning Algorithm), which determines the facts of a visit by one or more users based on Wi-Fi networks and / or other wireless networks detected by wireless devices of one or more users. In some specific non-limiting embodiments of the present technology, the systems comprise a server used to create a specific training dataset and train an MLA algorithm using that training dataset.

[13] In some such embodiments, the server executes a metrics module, a POI signature module, a feature determination module, a training data generation module, and an MLA algorithm learning module. Each of these modules is detailed below. At least one non-limiting embodiment of each module is presented below. It should be understood that despite the description here of a particular set of modules, the functions they perform may be implemented using a different set of modules and/or other server configurations.

[14] Briefly, the metrics module accesses wireless user devices in a specific geographic area of interest and receives from each of these wireless devices instantaneous states of local networks (LAN, Local Area Network) with timestamps for a certain period of time. These instantaneous states represent Wi-Fi networks discovered by wireless devices at specific times in specific geolocations. From each received instantaneous LAN state, the metrics module excludes dynamic and mobile LANs and leaves only information about static LANs. Thus, the metrics module removes less reliable indications, indicators, and data from the instantaneous states of LANs.

[15] Briefly, the POI signature module uses a subset of the instantaneous LAN states defined by the metrics module to generate at least one POI signature for each POI in a geographic area of interest visited at least once during a predetermined time period. interval. To this end, the POI Signature Module obtains geolocation tracking data with timestamps from users' wireless devices and applies a set of heuristics to them that determine the facts and times of user visits to POIs in that geographic area. For each specific visit by a specific user having a specific wireless device, the POI signature module obtains the instantaneous state of the LANs defined by that wireless device at the time corresponding to the specific visit.

[16] Thus, the POI signature module for each at least once visited POI in a geographic area generates at least one, or in many cases, a plurality of instantaneous LAN states captured at the POI visited at least once. The POI signature module filters noise from the received POI instantaneous state sets using known statistical techniques and combines each LAN instantaneous state set for the POI into a single POI signature. Thus, a POI's signature is, in a sense, among other things, a typical set of Wi-Fi networks and their characteristics that are expected to be discovered by a wireless device "visiting" that POI. In some embodiments of the invention, the statistics module also creates additional metrics and features associated with each POI signature.

[17] The feature determination module then performs various comparisons on the data obtained by the above modules, and determines features of the POI signature distinguishing visits to the first POI from visits to the second POI. The MLA training module then injects a training dataset into the MLA algorithm to train that MLA algorithm.

[18] In the use phase, the trained MLA algorithm may receive as input the instantaneous state of the LANs determined by the user's wireless device at some point in time. In some cases, the MLA algorithm may receive a sequence of such input data corresponding to different times at which the wireless device reports its geolocation. In some cases, the sequence of input data may correspond to the movement of the user of the wireless device in the geographic area of interest. For each such input, the MLA algorithm determines if the user is visiting one of the POIs in a given geographic area at the time associated with that input.

[19] In view of the above overview and according to the first aspect of the present technology, a method for determining whether a user using a wireless device with a local communication module has visited a POI located at a location associated with a LAN network containing at least one wireless access point , which broadcasts a wireless signal. The method is executed by a server containing a processor and a long-term memory associated with the processor, which stores instructions executed by the processor to execute the method.

[20] The method includes: receiving at the server at the stage of using the identifier of the wireless device; determining at the server at the stage of use corresponding to the mentioned point in time a set of local communication features associated with the local communication module of the wireless device of the user, based on the identifier of the wireless device; input by the server in the use step of the local communication feature set of the local communication module of the wireless device as input to the use step into the MLA algorithm trained on the training data set associated with the POI in the training step preceding the use step.

[21] At the same time, the training object from the training data set is formed by: determining at least one geolocation of the training wireless device of the training user, with which the time point preceding the mentioned time point is associated, while the POI point is located in a predetermined neighborhood of at least one geolocation; determining a set of local communication training features that corresponded to the learning wireless device at a point in time associated with at least one geolocation and contains an indication of the learning wireless device that detected a wireless signal of at least one wireless access point of the LAN network in at least one geolocation having a wireless device learning ID and containing a local communication learning module; applying a set of heuristics to a user profile associated with the learning user during the training step to generate a training label indicating whether the training user visited a POI at a point in time associated with at least one geolocation, generating a training object during the training phase , which contains a set of training features of a local connection and a training label associated with it.

[22] The method includes obtaining by the server at the use step whether the user has visited the POI at said point in time from the MLA algorithm as the output of the use step based on the local communication feature set of the local communication module of the wireless device.

[23] In some embodiments of the invention, the user profile contains a user-specific part and a device-specific part.

[24] In some embodiments of the invention, the heuristic rule set comprises a first subset of heuristic rules applicable to the user-specific part and allowing to determine that the learning user interacted with a resource associated with a POI and having a telephone number associated with a POI, and/ or the website associated with the POI.

[25] In some embodiments of the invention, the first subset of heuristics further allows to determine that the learning user interacted with the web resource and that the interaction included the learning user performing a web search associated with the POI.

[26] In some embodiments of the invention, the web resource is a map service, and the interaction was a search for a POI performed by a learning user in the map service.

[27] In some embodiments of the invention, the method further includes determining, based on the learning wireless device ID, a set of tracked geolocations of the learning wireless device with timestamps that indicates locations where the learning wireless device has been for a predetermined time period, wherein the set heuristic rules contains a second subset of heuristic rules, applicable to the device-dependent part, and allowing to determine, based on a set of tracked geolocations with timestamps, that the training user has moved to a predetermined neighborhood of the POI.

[28] In some embodiments of the invention, the second subset of heuristics further allows you to determine that the training user has been shown a targeted message associated with the POI, before moving to a predetermined neighborhood of the POI.

[29] In some embodiments, the targeted message is a targeted online message.

[30] In some embodiments, the targeted message is a targeted offline message.

[31] In some embodiments of the invention, the second subset of heuristics further allows to determine that the training user has moved to a predetermined neighborhood of the POI within a predetermined deadline after being shown the targeted message.

[32] In some embodiments, the predetermined deadline is less than five hours.

[33] In some embodiments, the predetermined neighborhood is defined by a radius around the POI of five hundred meters.

[34] In some embodiments, the predetermined neighborhood is defined by a radius around the POI of one hundred and fifty meters.

[35] In some embodiments of the invention, the second subset of heuristics further allows to determine the connection of the training wireless device through the training communication module to the LAN associated with the POI.

[36] In some embodiments of the invention, said location is associated with a plurality of LANs, wherein the local communication training features include an SSID (Service Set ID) of each LAN of the plurality of LANs and/or a Medium Access Control (MAC) address. , Media Access Control) of the wireless access point associated with each LAN of the plurality of LANs, and/or the signal strength of the wireless access point associated with each LAN of the plurality of LANs.

[37] In some embodiments, the method further includes performing, prior to the training step, a feature determination procedure that determines a set of features that are part of the local link training feature set and to which greater weights should be assigned by the MLA during the training step than for the rest of the set of training features of the local connection.

[38] According to another aspect of the present technology, a method for determining a conversion rate of a targeted message associated with a POI is implemented, executing on a server containing a processor and a non-volatile memory associated with the processor that stores instructions executed by the processor to execute the method.

[39] The method includes: determining by the server at the stage of using a plurality of users to whom, within a first predetermined time period, a targeted message associated with a POI was shown, each user of the plurality of users using a wireless device containing a local communication module and having a wireless device ID associated with the local communication module; obtaining at the server at the stage of using a plurality of sets of signs of local communication of modules of local communication of wireless devices based on the identifiers of wireless devices of a plurality of users; input by the server into the MLA algorithm as input of the stage of using a plurality of sets of features of local communication of wireless devices; receiving by the server from the MLA algorithm as the output of the stage of using a set of indications based on a set of local communication flags of local communication modules of wireless devices, wherein the indication from the set of indications is associated with a user from the set of users and indicates whether this user has visited the POI in within a second predetermined time period after the targeted message has been shown to him; determining by the server at the stage of use, based on the set of indications, a conversion rate parameter indicating the conversion rate of the targeted message, which is the percentage of the set of users who visited the POI within the second predetermined time period after they were shown the targeted message.

[40] According to another aspect of the present technology, a method for determining a Place Visit Lift (PVL) metric is implemented, executed on a server containing a processor and a long-term memory associated with the processor, which stores instructions executed by the processor to execute the method.

[41] The method includes: determining, by the server, a visitors_site parameter equal to the number of users who (a) were shown a targeted message directing users to the POI and (b) visited the POI after showing them the targeted message; determination by the server of the bypassers_site parameter equal to the number of users who did not visit the POI after showing them a targeted message; determination by the server of the visitors_non-site parameter, equal to the number of users who visited the POI without first showing them a targeted message; determination by the server of the bypassers_non-site parameter, equal to the number of users who were not shown the targeted message and who did not visit the POI; and the server determines the PVL metric using the formula

.

.

[42] In some embodiments, at least one of the parameters visitors_site, bypassers_site, visitors_non-site, and bypassers_non-site is determined using the visit determination method described above.

[43] In some embodiments, the targeted message is a targeted online message.

[44] In some embodiments, the targeted message is a targeted offline message.

[45] The above examples are not limiting.

Brief description of the drawings

[46] FIG. 1 is a diagram of a system suitable for non-limiting embodiments of the present technology.

[47] FIG. 2 shows a log containing geolocation coordinates generated or hosted on a wireless device in the system shown in FIG. one.

[48] FIG. 3 shows a database of digital content items in the system shown in FIG. one.

[49] FIG. 4 shows the user profile database in the system shown in FIG. one.

[50] FIG. 5 is a non-limiting example of a geographic area, POIs in a geographic area, and users in a geographic area, each with at least one wireless device.

[51] FIG. 6 is a flow diagram of a non-limiting embodiment of a method according to the present technology.

[52] FIG. 7 is a flow diagram of a non-limiting embodiment of another method according to the present technology.

[53] FIG. 8 is a flow diagram of a non-limiting embodiment of yet another method according to the present technology.

Implementation of the invention

[54] FIG. 1 is a diagram of a system 100 suitable for non-limiting embodiments of the present technology. It is obvious that the system 100 is shown only to illustrate an embodiment of the present technology. Thus, the following description of the system is a description of examples illustrating the present technology. This description is not intended to define the scope or boundaries of the present technology.

[55] In some instances, useful examples of modifications to system 100 are provided. They aid understanding but also do not define the scope or boundaries of the present technology. These modifications do not constitute an exhaustive list. As one of skill in the art would appreciate, other modifications are likely to be possible. In addition, if modifications are not described in some cases (i.e., there are no examples of modifications), this does not mean that they are impossible and / or that this description contains the only possible implementation of one or another element of the present technology.

[56] One skilled in the art would appreciate that this may not be the case. In addition, it should be understood that system 100 may, in some cases, be a simplified implementation of the present technology, and that such variations are presented to facilitate a better understanding of it. Those skilled in the art will appreciate that other embodiments of the present technology may be significantly more complex.

[57] The examples and conventions presented herein are intended to provide a better understanding of the principles of the present technology, and not to limit its scope to such specifically given examples and terms. It is obvious that specialists in the art are able to develop various methods and devices that are not explicitly described or shown, but implement the principles of the present technology within its essence and scope. In addition, to facilitate a better understanding, the following description may contain simplified implementations of the present technology. A person skilled in the art should be clear that other embodiments of this technology can be significantly more complex.

[58] Moreover, the description of principles, aspects, and embodiments of the present technology, as well as their specific examples, is intended to cover their structural and functional equivalents, whether they are currently known or will be developed in the future. For example, it should be apparent to those skilled in the art that any block diagrams described herein correspond to conceptual representations of illustrative circuit diagrams embodying the foundations of the present technology. It should also be obvious that any flowcharts, process diagrams, state transition diagrams, pseudocodes, etc. correspond to various processes that may be represented on a computer-readable physical storage medium and may be performed by a computer or processor, whether such a computer or processor is explicitly shown or not.

[59] The functions of the various elements shown in the drawings, including any functional block designated as "processor", can be implemented using specialized hardware, as well as hardware capable of executing the corresponding software. If a processor is used, these functions may be performed by a single dedicated processor, a single shared processor, or multiple individual processors, some of which may be shared.

[60] In some embodiments of the present technology, the processor may be a general purpose processor such as a central processing unit (CPU) or a specialized processor such as a graphics processing unit (GPU). In addition, explicit use of the term "processor" or "controller" should not be construed as referring solely to the hardware capable of executing the software and may refer to, but is not limited to, digital signal processor (DSP) hardware, network processor, ASIC (ASIC), Field Programmable Gate Array (FPGA), Read Only Memory (ROM) for storing software, Random Access Memory (RAM), and Non-Volatile Memory. Other hardware, general purpose and/or custom, may also be contemplated.

[61] In the following, in view of the foregoing principles, some non-limiting examples are provided to illustrate various embodiments of aspects of the present technology.

[62] As shown in FIG. 1, system 100 has access to a plurality of wireless devices 101. Some wireless devices 101 are shown in FIG. 1. Wireless devices 101 are similar to each other, at least to the extent that they are suitable for use within the present technology. For this reason, only one exemplary wireless device 101 is described in detail here, namely wireless device 102. Wireless device 102 is typically a portable wireless device that interacts with a user (not shown) and may sometimes be referred to as a "client device". It should be noted that the interaction of the wireless device 102 with the user is not meant to suggest or imply any mode of operation, such as logging in, registering, or the like.

[63] As used herein, unless expressly stated otherwise, the term "wireless device" means any portable wireless device equipped with at least one form of geolocation capability to determine its geolocation and capable of using a Wi-Fi connection to connect to LANs. and/or other devices. Non-limiting examples of wireless devices include Wi-Fi and GPS-enabled smartphones and tablets that users typically carry with them at all times. It should be noted that, in this context, a device functioning as a wireless device may also function as a server with respect to other wireless devices. The use of the term "wireless device" does not preclude the use of multiple client devices to receive, send, perform or initiate any task or request, or the results of any tasks or requests or steps of any method described herein.

[64] The wireless device 102 includes a nonvolatile memory 104 (memory) that stores instructions executable by the processor, a processor 106 associated with the nonvolatile memory 104, and a local communication module 109 associated with the processor 106. The nonvolatile memory 104 may include one or more storage media and provides storage space for computer instructions executed by processor 106. For example, non-volatile memory 104 may be implemented as a computer-readable physical storage medium, including ROM, hard disk drives (HDDs), solid state drives (SSDs), and flash memory cards. In this embodiment, the local communication module 109 includes a Wi-Fi module capable of connecting to a LAN that can be shared via a Wi-Fi access point such as a wireless router. In some embodiments, the local communication module 109 also includes a Bluetooth™ module for communication using a Bluetooth connection.

[65] The wireless device 102 also includes hardware and/or application software and/or firmware (or a combination thereof) for running a plurality of applications 107, including a navigation application 108, which is at least partially stored in non-volatile memory 104 One of the purposes of the navigation application 108 is to provide the user with a route to navigate from one location to another location. The method for implementing the navigation application 108 is known in the art and is not described here. For example, the navigation application 108 may be Yandex.Maps™, Yandex.Navigator™, or another commercially available or proprietary mapping application.

[66] The wireless device 102 also has access to a GPS receiver capable of receiving GPS satellite signals and determining the geolocation of the wireless device 102, or other means of determining the geolocation of the wireless device 102 at certain times and/or during certain time intervals. In the present embodiment, the wireless device 102 includes an embedded GPS receiver. In at least some embodiments, the wireless device 102 does not need to be connected to the Internet or running a navigation application 108 to receive GPS satellite signals. Therefore, the wireless device 102 can track a user's movement in the form of geolocations, which can be recorded in terms of latitude and longitude. regardless of the availability of an Internet connection. In this embodiment of the invention, the wireless device 102 determines the GPS coordinates with a frequency, in particular depending on the particular implementation of the wireless device 102 and its operating system, and logs them in the log 110.

[67] In some embodiments, the wireless device 102 stores the log 110 in the onboard non-volatile memory 104. In some embodiments, the log 110 is stored remotely from the wireless device 102, such as on a remote server. In some embodiments, the wireless device 102 provides one or more remote servers with access to its geolocation. In some embodiments, wireless device 102 provides real-time or near real-time access to its geolocation to one or more remote servers.

[68] In some non-limiting embodiments of the invention, the log 110 stores the geolocations of the wireless device 102 as geolocation coordinates of the wireless device 102 with timestamps collected or provided by the wireless device 102 over time. In some embodiments of the invention, the wireless device 102 collects and in some cases stores a log 110 and/or corresponding GPS coordinates with timestamps, for example, by combining the use of applications 107 and/or the operating system of the wireless device 102 without the user opening the specific application stored in wireless device 102.

[69] FIG. 2 shows a non-limiting embodiment of the log 110. No restrictions are placed on the method and format for populating and storing the log 110. For example, the wireless device 102 may receive GPS satellite signals at predetermined time intervals and convert the GPS satellite signals into geolocation coordinates with timestamps. In some embodiments, wireless device 102 includes and/or uses cell tower triangulation or other geolocation tools to obtain its geolocations instead of or in addition to using GPS.

[70] In FIG. 2, log 110 contains first geolocation coordinates 202, which in a non-limiting example correspond to MacDonald's™ building 500 in geographic area 136. First geolocation coordinates 202 are associated with first timestamp 204. The first timestamp 204 corresponds to the time when the wireless device 102 was located according to the first geolocation coordinates 202. The log 110 also contains additional geolocation coordinates and associated timestamps (not shown separately). For example, based on the timestamps and geolocation coordinates in the log 110, it should be clear that the wireless device 102 has moved 0.1 arc seconds north and 0.4 arc seconds west over the course of 6 minutes. For clarity, log 110 shows only four geolocations. Log 110 may contain more or less geolocations or coordinates.

[71] In some embodiments, wireless device 102 is capable of assigning a user device identifier 206 to log 110. For example, user device identifier 206 may be the MAC address of wireless device 102 or a member thereof and/or may be an indication of a user name ( such as an email address) associated with the user if the wireless device 102 requires user authorization, and the like. In another example, user device identifier 206 may correspond to a proprietary identification number assigned, for example, by navigation application 108 and/or other service application(s) 112 (described below). Thus, the user device identifier 206 is one example of the identifier 206 of the wireless device 102.

[72] Wireless device 102 includes hardware and/or application software and/or firmware (or a combination thereof) for executing one or more service applications 112 (see FIG. 1). In some embodiments of the invention, one or more service applications 112 include at least one service application (not indicated) managed by the organization providing the navigation application 108 described above. For example, if the navigation application 108 is a Yandex.Navigator application, then one or several service applications 112 may include a Yandex.Browser™ web browser application, a Yandex.News™ news application, a Yandex.Market™ product search and selection application, and the like. Obviously, one or more service applications 112 may also include service applications that are not operated by the organization providing the navigation application 108 described above, such as social media applications such as the Vkontakte™ application and a music streaming application such as Spotify. ™.

[73] In some embodiments, user actions performed using wireless device 102 are collected by one or more servers and used to generate a user profile associated with the user and/or wireless device (or devices) 102 of the user. In some embodiments of the invention, the user profile contains a user-specific part containing information regarding the user, and a device-specific part containing information regarding the wireless device (or devices) 102 of the user. In some embodiments of the invention, one or more service applications 112 managed by the organization providing the navigation application 108 are capable of reporting and/or storing user actions and/or geolocations with an accompanying reference to the identifier 206 of the user device or to other means that allow the subsequent determination of the history actions or geolocations of the user with the wireless device (or devices) 102 of the user, with the help of which the user performed the actions that make up the history of the user's actions.

[74] The wireless device 102 includes a communication interface (not shown) for providing two-way communication with the communication network 114 over the communication line 116 . In some non-limiting embodiments of the present technology, the Internet may be used as communication network 114 . In other embodiments of the present technology, communications network 114 may be implemented in other ways, such as any wide area network, local area network, private network, or the like. In some embodiments of the invention, the communication interface contains the local communication module 109 of the wireless device 102 and/or works with it.

[75] The implementation of the link 116 is not particularly limited, but depends on the implementation of the wireless device 102. By way of non-limiting example only, in those embodiments of the present technology in which the wireless device 102 is implemented as a wireless device (such as a smartphone), link 116 may be implemented as a wireless link (such as a 3G network link, a 4G network link, Wireless Fidelity or Wi- ^Fi® , ^Bluetooth® , etc. for short) or a wired communication lines (such as an Ethernet-based connection).

[76] It should be obvious that the embodiments of wireless device 102, link 116, and network 114 are for illustration purposes only. A person skilled in the art should be clear and other specific details of the implementation of the wireless device 102, communication line 116 and communication network 114. The above examples do not limit the scope of the present technology in any way.

[77] System 100 also includes a server 118 connected to a communications network 114 (see FIG. 1). Server 118 may be implemented as a computer server. In an exemplary embodiment of the present technology, server 118 may be implemented as a Dell™ PowerEdge™ server running a Microsoft™ Windows Server™ operating system. Obviously, the server 118 may be implemented using any other suitable hardware and/or application software and/or firmware, or a combination thereof. In the present non-limiting embodiment of the present technology, the server 118 is implemented as a single server. In other non-limiting embodiments of the present technology, the functions of the server 118 may be distributed among multiple servers.

[78] The server 118 includes a communications interface 119 whose structure and settings allow communication with various entities (such as wireless devices 101 and other devices that may be associated with the communications network 114) via the communications network 114. Server 118 includes non-volatile memory 120 that includes one or more storage media and generally provides storage space for computer program instructions executable by processor 122. For example, memory 120 may be implemented as a machine-readable physical storage medium, including ROM and/or or RAM. Memory 120 may also include one or more storage devices such as hard disk drives (HDDs), solid state drives (SSDs), and flash memory cards.

[79] In some embodiments of the invention, the server 118 may be managed by an organization that provides the browser application 108 described above. Moscow, 119021, Russia). In other embodiments of the invention, the server 118 may be managed by an organization other than that provided by the navigation application 108 described above. functions, in alternative non-limiting embodiments of the invention, the server 118 may be configured with a different set of modules and/or in an at least partially distributed manner, provided that the functionality described herein is provided.

[80] According to some non-limiting embodiments of the present technology, server 118 is capable of executing a digital content item selection application 124 (selection application 124). The method for implementing the selection application 124 is described in detail below.

Database of digital content items

[81] The server 118 is connected to the database 126 of the digital content items. In other embodiments, the digital content item database 126 may be connected to the server 118 via a communications network 114 . While the digital content item database 126 is schematically shown here as a single item, it is contemplated that the digital content item database 126 may be distributed.

[82] The digital content item database 126 is populated with a plurality of digital content items (not separately indicated). No particular limitation is imposed on the nature of each item of the plurality of digital content items. In general, a digital content item may correspond to a targeted message associated with one or more POIs and containing one or more offers, images, videos, etc.

[83] FIG. 3 shows a non-limiting embodiment of a digital content item database 126 populated with a plurality of digital content items.

[84] There are no restrictions on how to populate the digital content item database 126 . For example, the digital content item database 126 may receive digital content items from one or more content providers.

[85] In at least some non-limiting embodiments of the present technology, the digital content item database 126 may store a plurality of digital content items grouped according to one or more topics. Thus, the digital content item database 126 can perform a topic clustering procedure (not shown). There are no restrictions on how to cluster a plurality of digital content items according to one or more topics or events. For example, traditional clustering techniques such as topic modeling or keyword based approaches can be used.

[86] A plurality of digital content items are grouped into one or more topic clusters 302. For example, the digital content item database 126 stores a first topic cluster 304, a second topic cluster 306, a third topic cluster (not labeled), and a fourth topic cluster (not labeled ). The first topic cluster 304 contains the first digital content item 308 and the second digital content item 312, and the second topic cluster 306 contains the third digital content item 310.

[87] In some non-limiting embodiments of the present technology, the digital content item database 126 contains additional information regarding each item of a plurality of digital content items, such as the duration of the digital content item, the minimum bid, the target parameters selected by the source of the digital content item, an indication of the static or dynamic nature of a digital content element, etc.

User profile database

[88] The server 118 is also connected to a user profile database 128 which, among other things, stores user profiles 129 (described below) of the wireless devices 101 (see FIG. 1). In other non-limiting embodiments of the present technology, the user profile database 128 may be connected to the server 118 via the communications network 114 . Although the user profile database 128 is schematically shown here as a single element, it is contemplated that the user profile database 128 may be distributed or have any other configuration.

[89] The user profile database 128 is a repository of user profiles 129, each of which is associated with one or more wireless devices 101. There are no restrictions on how one or more user profiles 129 are stored. For example, they may be a set of vectors representing the interests of a user of one or more wireless devices 101.

[90] FIG. 4 is a diagram of the association of various user profiles associated with a user of a wireless device 102, which is an example of one of the wireless devices 101.

[91] The first profile 402 is received from the first service server 404. For example, the first service server 404 may be associated with the first service application 401, which is a Yandex.Browser application that is managed by the organization providing the navigation application 108 of the wireless device 102 described above. For example, , the first profile 402 may be generated by the first service server 404 based on browsing logs 403 associated with the wireless device 102.

[92] The first profile 402 is associated with the first unique identifier 406, which is a proprietary user identifier associated with the wireless device 102. In particular, given that the first service application 401 is managed by the organization providing the aforementioned navigation application 108 of the wireless device 102, the first the unique identifier 406 of the first user profile 402 corresponds to the user device identifier 206 of the wireless device 102.

[93] The second profile 408 for the user of the wireless device 102 is received from the second service server 410. For example, the second service server 410 may be associated with the second service application 409 of the wireless device 102, which is a Yandex.Market application that is managed by an organization that provides the above described navigation application 108. For example, the second profile 408 may be generated by the second service server 410 based on search logs 411 associated with searches performed by the user of the wireless device 102 using the Yandex.Market application on the wireless device 102.

[94] The second profile 408 is also associated with the second unique identifier 412. Since the Yandex.Market application requires the user to provide an email address to access the digital service, the second unique identifier 412 contains both the user's email address and the user's proprietary identifier. , corresponding to the identifier 206 of the user device.

[95] In some non-limiting embodiments of the invention, the user profile database 128 is capable of performing a profile merging procedure (not shown). The profile merging procedure makes it possible to determine whether the first profile 402 and the second profile 408 correspond to the same user. For example, the profile merging procedure may determine if, at least in part, the first unique identifier 406 matches the second unique identifier 412.

[96] If it is determined that the first unique identifier 406 at least partially matches the second unique identifier 412, then the profile merging procedure may combine the first profile 402 and the second profile 408 to form a combined profile 414.

[97] As a result of the profile merging procedure, the merged profile 414 is stored in the user profile database 128 along with a third unique identifier 416 (which corresponds to the user device identifier 206 and email address).

[98] On the other hand, if the profile merging procedure determines that the first unique identifier 406 does not even partially match the second unique identifier 412, then the first profile 402 and the second profile 408 are considered to be associated with different users. The first profile 402 (and the first unique identifier 406) and the second profile 408 (and the second unique identifier 412) are then stored separately in the user profile database 128 .

[99] Although only two user profiles (first profile 402 and second profile 408) are shown here to form the combined profile 414, it should be understood that the combined profile 414 for the wireless device 101 may be formed based on more than two profiles. user.

[100] In addition, despite the fact that the combined profile 414 is formed solely on the basis of service applications managed by the same organization, the scope of the invention is not limited to this. Because the second user profile 408 contains an email address associated with the user, it is possible to further merge the user profile with a third user profile (not shown) received from another organization if the third user profile is associated with a third unique identifier corresponding to the same email address.

[101] Briefly, the user profile database 128 stores a plurality of user profiles 129, each of which is associated with at least one wireless device 101. In some non-limiting embodiments of the present technology, user profiles 129 contain additional information about the users that they belong. A non-limiting example of such additional information is the gender of users. Another non-limiting example is the purchase history of users.

Metrics Module

[102] In some non-limiting embodiments of the invention, the server 118 stores in its memory 120 and executes the metrics module 139 (see FIG. 1). The metrics module 139 receives timestamped geolocation data of the wireless devices 101 available to the server 118 over a period of time. device 101. The metrics module 139 stores time-stamped geolocation data of each wireless device 101 along with a corresponding unique identifier of the wireless device 101 in a user geolocation database 141 . In some embodiments, the metrics module 139 further configures the user geolocation database 141 by matching the geolocation data of each wireless device 101 to the corresponding user profile 402 and/or user device ID 206 or other appropriate unique identifier.

[103] The metrics module 139 also receives from each wireless device 101 information about the LANs 135 LANs detected by each wireless device 101 in different geolocations, at least at a data collection rate (see FIG. 5). The frequency of data collection may be different for different wireless devices 101 and may depend on the particular implementation of each wireless device 101 and/or on its operating system and/or on the configuration of each wireless device 101. In a non-limiting example, wireless device 101 may detect available networks 135 LANs at 1/3 Hz (once every three seconds).

[104] As shown schematically in FIG. 5, LAN network 135 may include a wireless access point 137 through which LAN network 135 can communicate with one or more wireless devices 101. In a non-limiting example, wireless access point 137 can be a router. The wireless device 101 can detect the wireless signal emitted by the wireless access point 137 of the LAN 135 . Based on the wireless signal, each wireless device 101 can determine: 1) the MAC address of the wireless access point 137, 2) the SSID of the LAN 135, and 3) the strength of the wireless signal. In some embodiments, the wireless device 101 may also derive additional metrics from the detected wireless signal that may indicate at least some characteristics of the wireless access point 137.

[105] At any geolocation, the wireless device 101 can detect multiple wireless signals emitted by different wireless access points 137 that the wireless device 101 can be in range of. The wireless device 101 can obtain the MAC address, SSID, and signal strength for each detected wireless signal. In a non-limiting example, the user 103 of the wireless device 102 may visit MacDonald's job 500 in geographic area 136 at some point in time (see FIG. 5). When the user 103 is in the building, the wireless device 102 of the user 103 can discover the following set of Wi-Fi networks: 1) a LAN 135 owned by the MacDonald's restaurant 500 itself, 2) a LAN 135 LAN of an apartment building 501 located next to the MacDonald's building 500, and 3) a LAN 135 shared via a Wi-Fi hotspot in a portable wireless device 502 of another user 504 who is in the MacDonald's building 500 at the same time as the user 103.

[106] In another non-limiting example, after visiting MacDonald's building 500, user 103 may decide to navigate, as indicated by arrow 506, to a nearby Petro-Canada™ gas station 508. At an appropriate subsequent point in time when the user 103 is at the gas station 508, the wireless device 102 of the user 103 may discover a different set of Wi-Fi networks: 1) a LAN 135 owned by the gas station 508, and 2) a LAN 135 LAN owned by the restaurant 500 MacDonald's because the user is still within range.

[107] As shown in FIG. 5, when the user 103 is at the gas station 508, the wireless device 102 of the user 103 is out of range of residential LANs in geographic area 136 and out of range of another user's Wi-Fi access point 504 because the other user's device 502 emits more weaker signal than wireless access point 137 belonging to 500 MacDonald's restaurant.

[108] It is contemplated that during user 103's movement from MacDonald's building 500 to gas station 508, wireless device 102 of user 103 may have detected additional distinct sets of LANs at the frequency associated with wireless device 102.

[109] The MAC address(es), SSID(s), and signal strength(s) detected by each wireless device 101 at each point in time, as well as the corresponding discovery time and additional metric(s) that may be determined by the wireless device 101 may be collectively referred to as the LAN instantaneous state 123 with a timestamp (FIG. 1). The metrics module 139 can receive from each wireless device 101 a corresponding instantaneous LAN status 123 with a timestamp, for example, at a frequency associated with each wireless device 101, and store the instantaneous LAN states 123 in the instantaneous LAN state database 121 in a suitable structure. data.

[110] For example, in some non-limiting embodiments of the invention, the metrics module 139 can receive instantaneous LAN states 123 from all wireless devices 101 available to it in bursts at predetermined time intervals. Thus, the server 118 can use the collected LAN snapshots 123 from all wireless devices 101 to form a LAN snapshot database 121 .

[111] In some non-limiting embodiments of the invention, the metrics module 139 may filter at least some LANs from the collected or stored instantaneous states 123 of the LANs. For example, in some non-limiting embodiments of the invention, the server 118 and/or the metrics module 139 can determine or discover, for each wireless device 101, which LANs are in the vicinity of that wireless device 101 at any given time or at least some number of consecutive instantaneous states 123 of LANs are static LANs and which are mobile or dynamic LANs.

[112] There are no restrictions on the specific methods or algorithms used to distinguish and filter mobile or dynamic LANs from static LANs. Any suitable methods or algorithms may be used for this purpose. In the filtered successive instantaneous states of the 123 LANs, only static LANs are represented. In at least some non-limiting embodiments of the invention, the instantaneous states 123 of static LANs contribute to the determination of more reliable signatures of POIs 134, as described in more detail below.

[113] In some non-limiting embodiments of the invention, the module 139 metrics can remove from the database 121 instantaneous states of LANs or simply not initially store in the database 121 instantaneous states of LANs MAC address (or addresses), identifier (or identifiers ) SSID, signal strength(s), and any additional mobile or dynamic LAN information found in any instantaneous state(s) of the 123 LANs.

[114] Accordingly, in such non-limiting embodiments of the invention, the instantaneous LAN state database 121 may only contain information about static LANs in the vicinity of the wireless devices 101. As a non-limiting example of a static LAN, LAN 135 whose SSID and geolocation do not change for at least a predetermined time period, such as a month. A non-limiting example of a mobile or dynamic LAN is a LAN whose geolocation changes over a predetermined time period, such as a day. As a non-limiting example of a mobile or dynamic LAN, a LAN shared via a Wi-Fi access point in the portable wireless device 502 of the user 504 can be given (see FIG. 5).

[115] It is contemplated that additional and/or other definitions for static LANs and/or mobile or dynamic LANs may be used instead of or in addition to the above non-limiting examples.

POI databases

[116] The server 118 may further be connected to a remote POI point profile database 130 and to a POI point signature database 131 (see FIG. 1). In some non-limiting embodiments of the present technology, the remote POI profile database 130 may be connected to the server 118 using a local connection or other direct connection. In some non-limiting embodiments of the present technology, the POI signature database 131 may be a remote database associated with the server 118 via the communications network 114 . While the POI profile database 130 and POI signature database 131 are schematically shown here as respective single entities, it is contemplated that the database 130 and/or the database 131 may be distributed.

[117] In some non-limiting embodiments of the invention, a plurality of profiles 132 of POI points 134 located in a geographic area 136 may be stored in the POI point profile database 130. A non-limiting example of a geographic area 136 and POI points 134 located therein schematically shown in FIG. 5. For example, geographic area 136 may represent a region within a country, a city, a specific city block, a specific street, and the like. For example, POIs 134 may represent businesses, institutions, points of interest, and the like. Specific examples of POIs 134 include MacDonald's building 500 and Petro-Canada gas station 508 (see FIG. 5). Specifically, POI 134 profile 132 may include POI 134 schedule, POI 134 category, POI 134 geolocation, POI 134 rating, and the like.

POI signature module

[118] In some non-limiting embodiments of the invention, the server 118 may additionally store in its memory 120 and execute the POI signature module 140 (see FIG. 1). Briefly, POI signature module 140 generates a POI signature database 131 for POIs 134 in geographic region 136 based on at least a first subset of instantaneous LAN states 123 captured or acquired by metrics module 139 during at least a first time period and stored in the database 121 instantaneous state data of LANs, as described in detail below.

[119] The generated POI signature database 131 stores at least one POI signature 157 for each POI 134 in geographic area 136. The POI signature 157 indicates a typical set of LANs 135 (if any) and their respective characteristics. , since they are typically detected by the user's wireless device 101 when the user visits that POI 134 . An example of the process implemented by the POI signature module 140 is described in detail below.

[120] In some non-limiting embodiments of the invention, the POI signature module 140 generates or defines a POI signature database 131 such that, for each user of each wireless device 101, it contains a set of data indicating the POIs 134 visited by that wireless device. by the user during some time interval, and at the time points of each visit. As described in more detail below, the POI signature module 140 then determines one or more instantaneous LAN states 123 corresponding to each detected visit to each POI 134 . Thus, the POI signature module 140 defines a set of instantaneous LAN states 123 corresponding to different times and different wireless devices 101 for each POI 134 . Based on each such set of LAN instantaneous states 123 for each POI 134, the POI signature module 140 determines a signature representing the POI 134 . This process is detailed below.

[121] The POI signature module 140 accesses the user profile database 128 to obtain user device IDs 206, user profiles 129, and corresponding wireless device IDs 101. Based on at least some of this information, the POI signature module 140 can then identify the logs. 110 of the respective wireless devices 101 and obtain from them the geolocation data associated with the wireless devices 101 in the geographical area 136 for at least a predetermined time interval, for example, a month. The POI point signature module 140 may then access the POI point profile database 130 and obtain from it the POI point profiles 134 in geographic area 136.

[122] Then, the POI signature module 140 may apply a set of heuristics to the above received or determined data and thus determine, for each combination of user and user device, specific POIs 134 (if any) that the user visited during a predetermined time interval, and time points of each visit.

[123] In a non-limiting example, the POI signature module 140 may determine that a user 103 having a wireless device 102 visited MacDonald's building 500 on November 29, 2018 from 12:00 to 12:10, November 31, 2018 from 13:00 to 14:00 and December 1, 2018 from 18:25 to 18:45 (see Fig. 5). The POI signature module 140 can also determine that another user 504 having a wireless device 502 visited MacDonald's building 500 on Nov. 29, 2018 from 9:00 am to 10:00 am and Nov. . The POI signature module 140 can also determine many other visits to this building by other users.

[124] POI signature module 140 can similarly determine many different visits by many different users to different POIs 134 in geographic area 136. Suffice it to say that POI signature module 140 performs similar steps for visits by other users. For simplicity, in the present example, only the user 103 and the corresponding steps of the POI signature module 140 are described below.

[125] Thus, if the user 103 is used as an example of a user among a large number of users for which the POI signature module 140 determines the facts of the visit, the POI signature module 140 can determine in the log 110 of the wireless device 102 of the user 103 a plurality of geolocations with timestamps representing three visits to the 500 MacDonald's building. For example, for a first visit, POI signature module 140 can determine two hundred timestamped geolocations provided by wireless device 102 of user 103 associated with the first visit if wireless device 102 was reporting its geolocations at 1/3 Hz at the time.

Set of heuristic rules

[126] In order to enable the aforementioned determination of the facts and times of visits by different users to different POIs 134, the set of heuristics contains a first subset of heuristics applicable to the user-specific part of each user profile 129. In some embodiments of the invention, the first subset of heuristics allows you to determine the interaction of the relevant user with the resource associated with the point 134 POI.

[127] In some non-limiting embodiments of the invention, the resource associated with POI 134 may have a phone line associated with POI 134 and/or a website associated with POI 134. The website is one example of a web resource associated with POI 134 . Another example of a web resource is a map service. For the example of visiting MacDonald's building 500 described above, the telephone line may be the telephone line of a specific MacDonald's building 500, the website may be the general MacDonald's website, and the map service may be the navigation application 108 of the wireless device 102 of the user 103.

[128] In some non-limiting embodiments of the invention, a set of heuristics further allows you to determine the interaction of the user of one of the wireless devices 101 with the web resource of the POI 134 and the user performing a web search associated with the POI 134 in this interaction. For example, in the case of the MacDonald's building 500, the first subset of heuristics determines that the user 103 searched for the location of the MacDonald's task 500 using the navigation application 108 of the wireless device 102 and that the user 103 then arrived at a predetermined neighborhood of the MacDonald's building 500.

[129] In the example of user 103 visiting gas station 508, a first subset of heuristics determines that user 103 has searched for a product, such as an energy drink, using a search engine running on wireless device 102, that the search engine has indicated that the product is available. at the gas station 508, and that the geolocation tracking data of the user 103 in the corresponding time interval corresponded to the direction to the gas station 508. It is assumed that the first subset of heuristics allows the generation of appropriate visit determination results, additionally based on other combinations of conditions met.

[130] In some non-limiting embodiments of the invention, the set of heuristics may further comprise a second subset of heuristics applicable to the device-specific portion of the user profile 129 associated with each wireless device 101 of each user. A second subset of heuristics may be applied to a set of tracked geolocations with timestamps of each wireless device 101 indicating locations where that wireless device 101 has been during a predetermined time period. In particular, the POI signature module 140 may apply a second subset of heuristics to this set of tracked geolocations with timestamps of the wireless device 101 and thereby determine that the user of the wireless device 101 has moved to a predetermined neighborhood of the POI 134.

[131] In the event that the user 103 visits the MacDonald's building 500, the POI signature module 140 can thus determine that the user 103 of the wireless device 102 has moved to a predetermined neighborhood of the MacDonald's building 500, such as 150 meters, and has remained in that neighborhood for a period of time. associated with each visit, in other words, November 29, 2018 from 12:00 to 12:10, November 31, 2018 from 13:00 to 14:00 and December 1, 2018 from 18:25 to 18:45 .

[132] In addition, in some non-limiting embodiments of the invention, the second subset of heuristics allows you to determine that the user was shown a targeted message associated with the point 134 POI, before moving to a predefined neighborhood of the point 134 POI. In some non-limiting embodiments of the invention, the targeted message may be an online message. For example, in the case of the MacDonald's building 500, the targeted online message may be an indication delivered to the wireless device 102 of the user 103 that one of the items offered at the MacDonald's restaurant 500 has a special offer.

[133] In some non-limiting embodiments of the invention, the targeted message may be a physical (offline) message. For example, in the case of MacDonald's building 500, the offline message may be an indication that one of the products offered at MacDonald's restaurant 500 has a special offer shown to user 103 using an advertising panel 150 (FIG. 5) located in geographical area 136.

[134] In some non-limiting embodiments of the invention, the second subset of heuristics further allows to determine that the user has moved to a predetermined neighborhood of the point 134 POI within a predetermined deadline after showing him a targeted message associated with this point 134 POI.

[135] For example, in some non-limiting embodiments of the invention, the predetermined time limit programmed in the second subset of heuristics may be five hours. In some non-limiting embodiments of the invention, the predetermined neighborhood may be defined by a radius around the POI 134 of five hundred meters. In some non-limiting embodiments of the invention, the predetermined neighborhood may be defined by a radius around the POI 134 of one hundred and fifty meters.

[136] In the example of a user 103 visiting a gas station 508, a second subset of heuristics determines that an advertisement (targeted message) for an energy drink available at a gas station 508 has been delivered to the user's wireless device 102. The second subset of heuristics further determines that the user 103 moved to a 150-meter neighborhood of the gas station 508 within 30 minutes of being shown this advertisement. station 508.

[137] In some such non-limiting embodiments of the invention, the display facts and corresponding time points of displaying one or more identified targeted messages, including targeted online messages and targeted offline messages, to identified users can be determined by the server 118 using the technology described in the same applicant's patent application "Method and system for providing a recommended digital content item" with attorney case number 40703-128 (filed concurrently with this application), the contents of which are incorporated herein by reference in their entirety. It is contemplated that other methods of determining when targeted messages are shown to users can also be used.

[138] In some non-limiting embodiments of the invention, the second subset of heuristics further allows you to determine that the wireless device 101 is connected through its local communication module 109 to the LAN network 135 associated with the location of the POI point 134 and/or with the point 134 itself. POI. In the example of gas station 508 visited by user 103, the second subset of heuristics determines that after arriving in the 50-meter neighborhood of gas station 508, user 103 connected to the gas station 508 LAN network 135 using his wireless device 102.

[139] Briefly, a set of heuristics allows the POI signature module 140 to determine the facts and times of visits by identified users to identified POIs 134 . In at least some non-limiting embodiments of the invention, the POI signature module 140 stores indications of specific visits and/or times in the POI signature database 131 .

[140] After determining the first set of facts of visits by various users of each POI 134 in the geographic area 136, the POI signature module 140 obtains instantaneous LAN states 123 corresponding to each fact and time of visit from the first set of facts and times of visits.

[141] In particular, in some non-limiting embodiments of the invention, the POI signature module 140 obtains from the LAN instantaneous state database 121 with respect to a predetermined time interval the instantaneous LAN state 123 with a time stamp for the wireless device 101 for each point 134 POI and for each specific pair of (a) the fact of visiting and (b) the time of visiting this point 134 POI by the user of the wireless device 101 during a predetermined time interval. In other words, the POI signature module 140 obtains a set of data indicative of the LANs 135 and their respective metrics discovered or determined by the wireless user devices 101 at each time the users visit each respective POI 134 .

[142] Thus, for each POI 134, the POI signature module 140 can determine a set of instantaneous states 123 of LANs with timestamps corresponding to different times and/or different visits to that POI 134 . The set of momentary LAN states 123 with timestamps may represent various changes in the LANs 135 (if any) and/or their respective characteristics detected or determined by the wireless devices 101 at the respective times.

[143] In a non-limiting example, during a visit by user 103 to MacDonald's building 500 on November 29 from 12:00 to 12:10, wireless device 102 of user 103 could report two hundred instantaneous states 123 of LANs if wireless device 102 provided instantaneous states of 123 networks LAN with 1/3 Hz timestamps. Obviously, the specific frequency example is for illustration purposes only and this frequency may vary. For example, two hundred timestamped LAN instantaneous states 123 may reflect changes in signal strengths and different sets of SSIDs and MAC addresses as they could be detected by wireless device 102 during a ten minute visiting period.

POI signatures

[144] After determining the sets of instantaneous states 123 LANs for each point 134 POI module 140 signature points POI can determine the signature 157 point POI for each point 134 POI. To this end, the POI signature module 140 may perform, for each POI point 134, a statistics module that is capable of averaging changes in various metrics represented by LAN instantaneous states 123 with timestamps corresponding to the POI point 134. This statistics module can be implemented using any suitable known technology. Thus, the POI signature module 140 may generate a POI signature 157 for each POI 134 .

[145] The POI signature 157 of a POI 134 may indicate a typical LAN network (or networks) 135 (if any), which should be determined by the wireless device 101 at any time at the location of this POI 134 . In at least some non-limiting embodiments of the invention, the POI signature 157 for a certain POI 134 is a statistical average of all instantaneous LAN states 123 that were determined to be received during a user's visit to that POI 134.

[146] In some non-limiting embodiments of the invention, the statistics module of the POI signature module 140 may also determine additional statistics and/or metrics that can be expected to be detected or determined by the wireless device 101 at any time at the location of the POI 134. In some such embodiments, additional statistics and/or metrics may form part of the POI signature 157 for some POI 134 . For example, in some non-limiting embodiments of the invention, additional statistics and/or metrics may include weighted percentages of visitors or users whose wireless device(s) 101 has discovered specific LANs at each POI 134. In another example, additional statistics and/or metrics may include indications of signal strength spread at the location of each POI 134 .

[147] In some non-limiting embodiments of the invention, the POI signature module 140 generates multiple POI signatures 157 for each POI 134 in geographic area 136. In some such embodiments, each POI signature 157 corresponds to a specific time interval of a certain day. . In some such embodiments, each POI signature 157 corresponds to a specific time slot of a specific day of any week. In some such embodiments, each POI signature 157 corresponds to a particular time slot of a particular day of a particular week of the year. In other words, in at least some such non-limiting embodiments of the invention, POI signatures 157 provide information about the network environment of various POIs as a function of time. In some such embodiments, the POI signatures 157 may be relatively more accurate.

[148] After generating POI signatures 157 for POIs 134 in geographic area 136, server 118 may perform a feature determination procedure. The feature definition aspect of the present technology is described below.

Feature detection module

[149] In some non-limiting embodiments of the invention, server 118 further stores in its memory 120 and executes feature determination module 152 (see FIG. 1). Feature determination module 152 is configured to perform a feature determination procedure to determine those metrics of POI signatures 157 that distinguish visits to the first POI from visits to the second POI. These metrics can be referred to as more meaningful features that are more indicative of the fact that a POI has been visited.

[150] In particular, in some non-limiting embodiments of the present technology, feature determination module 152 determines more meaningful features by comparing POI signatures 157 with each other signature to determine differences between them. The feature determination module 152 then applies a set of statistical criteria to certain differences between the POI signatures 157 in order to determine the corresponding trends and thus obtains more meaningful features related to each location in the geographic area 136. It is assumed that any suitable one can be used for this purpose. a set of statistical criteria, for example, to detect trends in data. It is also contemplated that, in some embodiments of the invention, feature determination module 152 and related steps may be omitted.

Training data set generation module

[151] The server 118 also stores in its memory 120 and executes the module 154 for generating the training data set. Briefly, the training data generating module 154 generates a training data set, which is then fed into the MLA algorithm 158 to train the MLA algorithm 158 to determine whether a POI 134 has been visited based on the instantaneous LAN state 123 detected by the wireless device 101, as described in detail below.

[152] To generate the training dataset, the training dataset generation module 154 obtains the identifiers of a previously unused set of wireless devices 101 associated with geographic area 136 in order to obtain previously unused geolocation tracking data from them. In some embodiments, training data generation module 154 obtains previously unused geolocation tracking data from at least some wireless devices 101 that may have been used in generating POI signatures 157 . The expression "not previously used" in relation to wireless devices 101 means that the wireless devices 101 were not used by the server 118 when generating signatures 157 points POI. The expression "not previously used" in relation to the geolocation tracking data means that the geolocation tracking data was not used by the server 118 when generating the signatures 157 of the POIs. In at least some cases, the use of previously unused wireless devices 101 and/or their previously unused geolocation tracking data improves accuracy at the stage of using the MLA algorithm 158 trained as detailed below.

[153] This new previously unused geolocation tracking dataset with timestamps is referred to as a training geolocation tracking dataset with timestamps, with each training geolocation tracking dataset with timestamps containing a plurality of timestamped geolocations of the corresponding wireless device 101. Wireless devices 101 from which the geolocation tracking training dataset with timestamps has been received are referred to as the training wireless devices 101.

[154] For each geolocation with a timestamp of the training wireless device 101, the training dataset generator 154 obtains from the LAN instantaneous state database 121 described above an instantaneous LAN state 123 with a timestamp that corresponds to the timestamp of the given geolocation with the timestamp . In other words, the training dataset generating module 154 determines which LANs and corresponding metrics (if any) have been detected by the training wireless device 101 at that timestamp geolocation.

[155] Training data generating module 154 determines which POIs 134 (if any) are in the predefined neighborhood of each geolocation time stamp from each training geolocation tracking data of each training wireless device 101. In some non-limiting embodiments of the invention the predetermined neighborhood is defined by a radius around the corresponding geolocation with a timestamp of 60 meters. In other non-limiting embodiments of the invention, this radius is different. It is contemplated that the radius can be predetermined based on the accuracy of each respective time-stamped geolocation. It is contemplated that a radius can be predetermined for each wireless device 101 in both the learning phase and the usage phase based on the geolocation accuracy of that wireless device 101.

[156] If the training data generating module 154 determines that there are no POIs 134 in the predetermined geolocation timestamp neighborhood, then the training data generating module 154 marks the timestamped LAN instantaneous state 123 corresponding to the timestamped geolocation, a label indicating the absence of visited points 134 POI.

[157] On the other hand, if the training data generating module 154 determines that there is at least one POI 134 in the predefined geolocation neighborhood with the timestamp, then the training data generating module 154 applies a set of heuristics as described above, to determine if the POI 134 has been visited by the user of the corresponding educational wireless device 101 at a time corresponding to the timestamp of this geolocation. Then, the training dataset generating module 154 marks the instantaneous state 123 of the LANs with a timestamp corresponding to this geolocation with a timestamp, a tag indicating the result of the determination using the heuristic set. Such labels may be referred to as learning labels.

[158] Thus, in some non-limiting embodiments of the invention, where the set of heuristics provides a positive result for the POI 134, the training dataset generation module 154 marks the instantaneous state 123 of the LANs with a timestamp corresponding to the geolocation with the timestamp , training label indicating the name of a particular visited point 134 POI. In some non-limiting embodiments of the invention, the training label may contain the name of the visited POI 134 and the probability of visiting the POI 134 . In some non-limiting embodiments of the invention, the training label may further comprise the names of other POIs 134 in the predefined neighborhood and the corresponding probabilities of visiting those other POIs 134 .

[159] The data set resulting from the above detection and labeling steps is called the MLA training data set and is used to train the MLA algorithm 158 to determine or predict when a user visits a wireless device 101 based on the instantaneous LAN state 123 detected by the wireless device. device 101 at some point in time. The learning phase is detailed below.

MLA Algorithm Learning Module

[160] The server 118 also stores in its memory 120 an MLA algorithm learning module 156 and an MLA algorithm 158, such as a neural network type MLA algorithm or a gradient boosting type MLA algorithm. The MLA algorithm learning module 156 inputs the MLA algorithm training data into the MLA algorithm 158 in order to train the MLA algorithm 158.

[161] In particular, during the training phase, MLA learning module 156 may input each training LAN instantaneous state 123 into the MLA algorithm 158, and may assign a training label corresponding to the training LAN instantaneous state 123 as the corresponding output of the MLA algorithm 158. In this sense, each training instantaneous state 123 of the LANs provides a training set of local communication features for the MLA algorithm 158.

[162] In some non-limiting embodiments of the invention, the MLA learning module 156 adapts the MLA algorithm 158 to assign the more significant features of each learning instantaneous state 123 of the LAN networks with a larger weight used in determining the visit than other features of the learning instantaneous state 123 LAN networks. It is contemplated that this adaptation step may be omitted in some embodiments of the invention.

[163] In some embodiments, the MLA algorithm 158 is trained to return a name or other identifier of the visited POI 134 for positive determinations and an indication of no visited POIs 134 for negative determinations.

[164] In some embodiments, the MLA algorithm 158 can be trained to produce a ranked list of specific POIs 134, in which each POI 134 is represented with an indication of the likelihood that the corresponding user will visit the particular POI 134. The list may be ordered from the POI 134 with the highest probability of being visited to the POI 134 with the least probability of being visited.

[165] When the MLA algorithm 158 is sufficiently trained, various methods can be applied to the structure of the server 118 and system 100 described above. These methods are described below.

Method for determining when a POI has been visited

[166] In some non-limiting embodiments of the present technology, a method 600 for determining whether a user, using a wireless device with a local communication module 109, has visited a POI point 134 located at a location associated with a LAN network containing at least one point 137 wireless access, which broadcasts a wireless signal (see Fig. 6).

[167] In some non-limiting embodiments of the invention, method 600 is executed on a server 118 comprising a processor 122 and associated with processor 122 non-volatile memory 120 that stores instructions executed by the processor to execute method 600. One example of a server suitable for execution method 600 is the server 118 described above. Accordingly, the method 600 is described in relation to the server 118 (for simplicity, but not to limit the scope of the invention).

[168] Step 602: receive at the server at the stage of using the identifier of the wireless device.

[169] In one embodiment, method 600 begins at step 602, which includes obtaining, at server 118, a wireless device identifier 206 for wireless device 101 for which method 600 determines that a POI 134 has been visited.

[170] As a non-limiting example, it can be assumed that the server 118 obtains a wireless device identifier 206 for the wireless device 166 used by the user 162 (see FIG. 5). As described above, one example of a wireless device identifier 206 is a user device identifier 206 .

[171] It can be assumed that the point 134 POI, for which the visit of the user 162 should be determined, is the MacDonald's building 500 in the geographical area 136. It should be noted that for simplicity, the user 162 is shown with the same figure as the user 103 described above. However, it is assumed that in the present example, the user 162 is different from the user 103 described above.

[172] Step 604: determining at the server at the stage of use corresponding to a certain point in time a set of local communication features associated with the local communication module of the wireless device of the user, based on the identifier of the wireless device.

[173] In one embodiment of the invention, the method 600 may proceed to determine at the server 118 at the stage of using a corresponding set of local communication features associated with a local communication module 109 of the wireless device 166 of the user 162, based on the identifier 206 of the wireless device.

[174] In some non-limiting embodiments of the invention, the mentioned point in time for each execution of the method 600 corresponds to the point in time for which the method 600 needs to determine whether the user is visiting the point 134 POI.

[175] It can be assumed that user 162, having wireless device 166, enters MacDonald's Building 500 on December 4, 2018 at 10:43 AM ET (US and Canada) (see FIG. 5).

[176] In some non-limiting embodiments of the invention, for this point in time, the server 118, based on the identifier 206 of the wireless device, receives the instantaneous state 123 of the LAN networks detected by the local communication module 109 (in this non-limiting example, the Wi-Fi module) wireless device 166 at that point in time.

[177] Thus, in this example, the instantaneous LAN state 123 received by server 118 includes LANs 135 discovered by wireless device 166 inside MacDonald's Building 500 on December 4, 2018 at 10:43 AM ET. For example, on December 4, 2018 at 10:43 am, wireless device 166 could detect: 1) LAN 135 owned by MacDonald's restaurant 500, 2) LAN 135 of residential building 501 located in the southeast near job 500 MacDonald's, and 3) the 172 Wi-Fi network of the 508 Petro-Canada gas station located nearby to the northeast.

[178] In the present non-limiting embodiment of the invention and in the example, the wireless device 166 for each of the aforementioned Wi-Fi networks has detected at least the SSID, the MAC address of the corresponding wireless access point 137 and/or LAN 135, and the wireless signal strength corresponding point 137 wireless access. This defined data forms at least a portion of the local communication feature set for input to the MLA algorithm 158.

[179] Step 606: Input by the server in the use step of the local communication feature set of the local communication module of the wireless device as the input of the use step to the MLA algorithm trained on the training data set associated with the POI in the training step preceding the use step. The training object from the training data set is generated by: determining at the training stage at least one geolocation of the training wireless device of the training user, which is associated with a time point preceding the said time point, while the POI is in a predetermined neighborhood of at least one geolocation; determining at the training stage a set of training signs of local communication, which corresponded to the training wireless device at the point in time associated with at least one geolocation, and contains an indication of the training wireless device that detected a wireless signal of at least one wireless point in at least one geolocation LAN network access having a wireless device learning ID and containing a local communication learning module.

[180] In one embodiment, method 600 may proceed to input by server 118 in use step into an MLA algorithm, such as MLA algorithm 158, as input to use step a set of local link features (SSIDs, MAC addresses, signal strengths, etc.) .d.) determined by the wireless device 166 of the user 162 at the first time.

[181] Thus, in this part of the example, the local communication feature set input to the MLA algorithm 158 contains at least the SSID, MAC address, and signal strength of each of: 1) the LAN 135 owned by the MacDonald's restaurant 500, 2) the LAN network 135 of the residence 501; and 3) the Wi-Fi network 172 of the Petro-Canada gas station 508.

[182] It is contemplated that in some embodiments of the invention, only the SSID and signal strength may be used. It is also contemplated that in some embodiments of the invention, only the MAC address and signal strength may be used. In some non-limiting embodiments of the invention, POIs 134 in a predetermined geolocation neighborhood associated with each set of local communication features are determined similarly as described above.

[183] In some non-limiting embodiments of the invention, at this stage, the MLA algorithm 158 is already properly trained, as described above, in the training stage based on the training data associated with at least the POI 134 and, in some cases, with all 134 POIs in geographic area 136. The description of the formation of the training data set is given above and its details are not repeated here.

[184] Step 608: Indicate whether the user has visited the POI at said time from the MLA algorithm as an output of the use step based on the local communication feature set of the local communication module of the wireless device at the use stage.

[185] In one embodiment, the method 600 may proceed to receive by the server 118 at the stage of using an indication that the user 162 has visited the POI 134 at the mentioned point in time from the MLA algorithm 158 as the output of the use stage based on the local communication feature set of the local communication module 109 wireless device 166.

[186] Thus, in the present example, the MLA algorithm 158 issues an indication that the user 162 has visited the MacDonald's building 500. In some embodiments of the invention, the indication contains the name and/or address and/or other identifier (or identifiers) of a particular defined visited point 134 POI. Thus, in the example above, the indication could simply be the following information: MacDonald's 500, Address: XYZ.

[187] In some embodiments, the MLA algorithm 158 provides a ranked list of potential visited POIs 134 located in a predetermined geolocation neighborhood of the user 162. In some embodiments, the ranked list contains a "Yes" or "No" indication next to each POI 134 in the list, and the POIs 134 are more likely to be visited closer to the top of the list than the rest of the POIs 134 in the list. In some embodiments, the ranked list in the above example may contain the following information: 1) MacDonald's 500, "Yes", Address: XYZ; 2) Petro-Canada, "No", Address: XYZ.

[188] In other embodiments, the ranked list may indicate the probability of visiting each POI 134 in the ranked list. Thus, in the example above, the ranked list might contain the following information: 1) MacDonald's 500, 86% (probability of visiting), Address: XYZ; 2) Petro-Canada, 14% (probability of visiting), Address: XYZ.

[189] Obviously, steps 602-608 of method 600 can be performed on each geolocation reported by wireless device 166 of user 162. Each new instantaneous state 123 of LANs corresponding to each new geolocation provided can be entered into MLA algorithm 158 as a new set of features of a local connection.

[190] In the above example, visits to POIs by a single user are tracked. However, method 600 can be used to track and determine multiple user visits to any POI 134 in geographic area 136.

[191] Other embodiments, uses, and implementations of method 600 are also contemplated. For example, in FIG. 7 shows method 700 performed using method 600.

How to determine the conversion rate of a targeted message

[192] Specifically, method 700 is a method for determining the conversion rate of a targeted message associated with a POI. As with method 600, method 700 may be executed on a server that includes a processor and a non-volatile memory associated with the processor that stores instructions executed by the processor to execute the method.

[193] As with method 600, one example of a server suitable for performing method 700 is server 118. Therefore, method 700 is described below with respect to server 118 by way of illustration.

[194] Step 702: The server determines, at the use step, a plurality of users to whom, within a first predetermined time period, a targeted message associated with a POI was shown, each user of the plurality of users using a wireless device containing a local communication module and having an identifier wireless device associated with the local communication module.

[195] In one embodiment, the method 700 begins at step 702, including determining by the server 118, at the use step, a plurality of users to whom, within a first predetermined time period, a targeted message associated with a POI has been shown, each user from a plurality of users uses a wireless device 101 containing a local communication module 109 and having a wireless device identifier 206 associated with the local communication module 109 .

[196] It can be assumed that the targeted message, for which the method 700 determines the conversion rate, is presented in the advertising panel 150 and directs viewers to the MacDonald's building 500 (i.e. to the target point 134 POI in this example) (see Fig. 5 ). It can also be assumed that the predetermined time period corresponds to the day of the week.

[197] In addition, it can be assumed that the server 118 has determined that the users U ₁ ...U _n shown in FIG. 5, a targeted message was shown to be displayed on the promotional bar at various points in time during that day of the week.

[198] In some non-limiting embodiments of the invention, this determination step is performed by the server 118 using the technology described in patent application attorney case number 40703-128 filed concurrently with the present application by the same applicant. It is contemplated that other methods of determining which users have been shown targeted messages may also be used.

[199] Step 704: receiving at the server at the stage of using a plurality of local communication feature sets of local communication modules of wireless devices based on the wireless device identifiers of the plurality of users.

[200] After receiving or determining by the server 118 the time (or moments) of showing each of the users U ₁ ...U _n a targeted message on the advertising panel 150, the method 700 may proceed to receive on the server 118 at the stage of using a plurality of sets of local communication features of the modules 109 local communication of wireless devices 101 based on identifiers 206 of wireless devices 101 of multiple users U ₁ ...U _n .

[201] In some embodiments, the plurality of local communication feature sets are instantaneous LAN states 123 generated by wireless devices 101 of users U ₁ ...U _n that server 118 determined were generated by wireless devices 101 at times after showing each to the corresponding user U ₁ ...U _n of the targeted message on the advertising panel 150.

[202] Some U ₁ ...U _n users may ignore the targeted message and navigate within geographic area 136 without visiting the 500 MacDonald's building. Some U ₁ …U _n users might be attracted to a targeted message and actually head to the 500 MacDonald's building. When entered into the MLA algorithm 158, time-stamped LANs 123 from the devices of each of these users generated after they were shown the targeted message reflect these movements and visits.

[203] The steps and methods for determining the instantaneous states of 123 LANs have been described above, so their details are not re-described here.

[204] Step 706: server input to the MLA algorithm as input to the step of using a plurality of wireless device local communication feature sets as described above.

[205] Once the instantaneous states 123 of the LANs are received or determined by the server 118, the method 700 may proceed to the server 118 inputting the instantaneous states 123 of the LANs into the MLA algorithm 158 as input.

[206] Step 708: The server receives from the MLA algorithm as output of the use step, based on a plurality of sets of local communication attributes of wireless devices, a plurality of indications, each of which is associated with a user from a plurality of users and indicates whether this user has visited the point POI within the second predetermined time period after showing him the targeted message.

[207] The method 700 may proceed to have the server 118 receive a plurality of indications from the MLA algorithm 158 as the output of the use phase based on the plurality of instantaneous states 123 of the LANs (local link feature sets) of the wireless devices 101. These indications may be implemented in any suitable form, for example, in the form of one or more suitable data packets for each indication. They can be described as follows.

[208] An indication from a set of indications may be associated with a user from a set of users U ₁ ...U _n and may indicate that the user visited the MacDonald's building 500 within a second predetermined time period after being shown the targeted message on the advertising panel 150.

[209] In the present non-limiting example, it can be assumed that the second predetermined time period is five hours. Thus, at this step, each such indication indicates that one of the users U ₁ ...U _n visits the MacDonald's building 500 within five hours after showing him the targeted message on the advertising panel 150. Obviously, the value of the second predetermined time period may differ from five hours used as an example.

[210] Step 710: Based on the set of indications, the server determines at the use stage a conversion rate parameter indicating the conversion rate of the targeted message, which is the percentage percentage of the set of users who visited the POI within the second predetermined time period after they were shown the targeted message .

[211] In one embodiment of the invention, the method 700 may proceed to determine by the server 118 at the stage of using a conversion rate parameter indicative of the conversion rate of the targeted message, based on the set of mentioned indications. In some non-limiting embodiments of the invention, the conversion rate is the percentage of the set of users who visited the POI 134 targeted by the targeted message within a second predetermined time period after they were shown the targeted message.

[212] Thus, in this non-limiting example, the conversion rate is the percentage of users U ₁ ...U _n who visited the MacDonald's building 500 within five hours of being shown the targeted message on the advertising panel 150.

[213] While method 700 has been described above for one specific example of a time value, a specific POI, and a specific targeted offline message, it is contemplated that method 700 can be used to determine conversion rates for a large number of different POIs over a period of time. different time intervals and for different types of targeted messages. It is contemplated that the targeted messages may be targeted offline messages, targeted online messages, and/or a combination thereof.

Method for determining the growth in attendance of a place

[214] The present technology also provides an additional specific use of the method 600. In particular, according to the present technology, a method 800 for determining the PVL metric is implemented.

[215] The present technology for determining the PVL metric is developed based on the understanding by the developers that when targeted messages are intended for a place without an Internet connection, such as a POI (for example, for the 134 POIs shown in Fig. 5), it is difficult to determine the effectiveness such targeted messages in terms of directing users who have been shown targeted messages to the POIs targeted by those targeted messages. In particular, the developers have found that the difficulty of determining the effectiveness of targeted messages increases in at least some cases when targeted online messages are intended for POIs without an Internet connection (for example, for physical points 134 POI, presented in Fig. 5).

[216] In some non-limiting embodiments of the invention, determining the PVL metric of the present technology quantifies the effectiveness of a targeted message in terms of directing users to the specific POI to which the targeted message refers. In one aspect, it can be said that an increase in the PVL metric indicates how much more users were stimulated by one or more targeted messages to visit the specific POIs to which these targeted messages relate, compared to users who were not shown one or more targeted messages. .

[217] Thus, monitoring the PVL metric for one or more targeted messages according to the present technology allows you to determine which targeted message (or multiple messages) is inefficient, effective, or ultra-efficient. Accordingly, with the present technology, the provider (or providers) of one or more targeted messages can, for example, stop displaying ineffective targeted messages to users and can therefore reduce and/or save resources associated with displaying such targeted messages. Other uses for the PVL metric defined in accordance with the present technology are also contemplated.

[218] A specific, non-limiting embodiment of a method 800 for determining a PVL metric according to the present technology is shown in FIG. 8 described below. The method 800 is executed on a server that includes a processor and, associated with the processor, non-volatile memory that stores instructions executed by the processor to execute the method. An example of a suitable server is server 118, described above. Thus, method 800 is described and illustrated using server 118 as an example (for simplicity, but not to limit the scope of the invention).

[219] Step 802: The server determines a visitors_site parameter equal to the number of users (a) who were shown a targeted message directing users to the POI, and (b) who visited the POI after showing them the targeted message.

[220] In one embodiment of the invention, the method 800 begins with step 802, including determining the server 118 parameter visitors_site.

[221] In some non-limiting embodiments of the invention, the visitors_site parameter is equal to the number of users who (a) were shown a targeted message directing users to the POI, and who (b) visited the POI after showing them this targeted message.

[222] As a non-limiting example, the targeted message can be assumed to be placed on the advertising panel 150 shown in FIG. 5, an indication that a special offer is available for one of the items offered at 500 MacDonald's in Geographic Area 136. In this non-limiting example, the POI or target POI of the targeted message is the MacDonald's building 500 shown in FIG. five.

[223] In this non-limiting example, the facts and times of showing a targeted message to each user can be determined using the technology described in the aforementioned patent application with attorney case number 40703-128, filed simultaneously with the present application by the same applicant.

[224] In addition, in this non-limiting example, the occurrences of visits or simply visits and the corresponding timestamps of each (identified) user's visits can be determined according to the method 600 described above.

[225] Step 804: Server determines bypassers_site equal to the number of users who did not visit the POI after showing them the targeted message.

[226] In some non-limiting embodiments of the invention, method 800 may proceed to determine bypassers_site by the server that is equal to the number of users who did not visit the POI after showing them the targeted message.

[227] In this non-limiting example, for each user of one of the wireless devices 101, for which it is determined that he was shown a targeted message displayed on the advertising panel 150, the method 600 described above can be used to control the actions of the user in for at least a predetermined time period after the display.

[228] In particular, method 600 can be used to determine whether a user has visited MacDonald's building 500 at any point in time (or multiple points in time) within a predetermined time period after the show.

[229] Step 806: The server determines a visitors_non-site parameter equal to the number of users who visited the POI without showing them a targeted message.

[230] In one embodiment of the invention, method 800 may proceed to determine the visitors_non-site parameter. In some non-limiting embodiments of the invention, the visitors_non-site parameter is equal to the number of users who visited the POI without first showing them a targeted message.

[231] As described above, the facts and times of visits can be determined using the method 600, and the facts of the targeted message display can be determined using the technology described in patent application with attorney case number 40703-128, filed simultaneously with this application. same applicant.

[232] Step 808: The server determines the parameter bypassers_non-site equal to the number of users who were not shown the targeted message and who did not visit the POI.

[233] In one embodiment of the invention, method 800 may proceed to determine the bypassers_non-site parameter. In some non-limiting embodiments of the invention, the bypassers_non-site parameter is equal to the number of users who were not shown the targeted message and who did not visit the POI.

[234] As described above, the facts and times of visits can be determined using the method 600, and the facts of the absence of a targeted message can be determined using the technology described in patent application with attorney case number 40703-128, filed simultaneously with this application. by the same applicant.

[235] Step 810: server determines the PVL metric.

[236] Once all of the above parameters have been determined, method 800 may proceed to determine the PVL metric based on the determined parameters.

[237] According to the present technology, the PVL metric is determined by the server 118 based on the parameters described above according to the formula

, (1)

, (one)

[238] where each parameter has the corresponding definition given above in the description of the previous steps of the method 800.

[239] It is contemplated that in some embodiments of the invention, some but not all of the above parameters can be determined according to the present technology. Although the particular targeted message used to illustrate method 800 is a targeted offline message, in some non-limiting embodiments of the invention, method 800 is used to determine a PVL metric for targeted online messages.

[240] For a person skilled in the art, possible changes and improvements to the above-described embodiments of the present technology may be obvious. The foregoing description is for illustrative purposes only and is not intended to limit the scope of the invention.

[241] Although a certain order of steps in the methods is given above, it is contemplated that a different order of steps in the methods and/or additional and/or alternative steps in the methods may be used to perform the described methods.

[242] For a person skilled in the art, possible changes and improvements in the above-described embodiments of the present technology may be obvious. The above description provides specific non-limiting embodiments of the present technology.

Claims (39)

1. A method for determining whether a user using a wireless device with a local communication module has visited a point of interest (POI) located at a location associated with a local area network (LAN) containing at least one wireless access point that broadcasts a wireless signal executable by a server containing a processor and a long-term memory connected to the processor, which stores instructions executed by the processor to execute the method, and includes: - receiving on the server at the stage of using the identifier of the wireless device; - determination on the server at the stage of use of a set of local communication features corresponding to a certain point in time, associated with the local communication module of the user's wireless device, based on the identifier of the wireless device; - input by the server at the use stage of the local communication feature set of the local communication module of the wireless device as the input of the use stage into the machine learning algorithm (MLA) trained on the basis of the training data set associated with the POI in the training stage preceding the use stage, when In this case, the training object from the training dataset is formed by: - determining at least one geolocation of the educational wireless device of the educational user, which is associated with a time point preceding the mentioned point in time, while the POI is in a predetermined neighborhood of at least one geolocation; - determining a set of training signs of local communication that corresponded to the learning wireless device at a point in time associated with at least one geolocation, and containing an indication of the learning wireless device that detected the wireless signal of at least one wireless access point of the LAN network in at least one geolocation having a wireless device learning ID and containing a local communication learning module; - applying in the learning phase a set of heuristics to the user profile associated with the learning user to generate a learning label indicating whether the learning user visited the POI at a point in time associated with at least one geolocation, with the generation in the learning phase of the learning an object containing a set of local link training features and an associated training label; - receiving by the server at the stage of use an indication of whether the user has visited the POI at the mentioned point in time, from the MLA algorithm as the output of the stage of use, based on the local communication feature set of the local communication module of the wireless device. 2. The method of claim 1, wherein the user profile comprises a user-specific part and a device-specific part. 3. The method of claim 2, wherein the set of heuristics comprises a first subset of heuristics applicable to the user-specific part and determining that the learning user has interacted with a resource associated with a POI and having a telephone number associated with the POI. POI and/or website associated with the POI. 4. The method of claim 3, wherein the first subset of heuristics further allows determining that the tutor has interacted with the web resource and that the interaction has included the tutor performing a web search associated with the POI. 5. The method according to claim 4, characterized in that the web resource is a map service, and the interaction was a search for a POI performed by a learning user in the map service. 6. The method according to claim. 2, characterized in that it further includes determining, based on the training wireless device identifier, a set of tracked geolocations of the training wireless device with timestamps, which indicates the places where the training wireless device was during a predetermined time period , wherein the set of heuristics contains a second subset of heuristics applicable to the device-dependent part and allowing to determine, based on the set of tracked geolocations with timestamps, that the training user has moved to a predetermined neighborhood of the POI. 7. The method according to claim 6, characterized in that the second subset of heuristics further allows to determine that the training user was shown a targeted message associated with the POI before moving to a predetermined neighborhood of the POI. 8. The method of claim 7, wherein the targeted message is a targeted online message. 9. The method of claim 7, wherein the targeted message is a targeted offline message. 10. The method of claim. 7, characterized in that the second subset of heuristics further allows you to determine that the learning user has moved to a predetermined neighborhood of the POI within a predetermined deadline after showing him a targeted message. 11. The method according to claim 10, characterized in that the predetermined deadline is less than five hours. 12. The method according to claim 6, characterized in that the predetermined neighborhood is defined by a radius around the POI of five hundred meters. 13. The method according to claim 6, characterized in that the predetermined neighborhood is defined by a radius around the POI of one hundred and fifty meters. 14. The method of claim. 6, characterized in that the second subset of heuristics further allows you to determine the connection of the training wireless device through the training communication module to the LAN associated with the POI. 15. The method according to claim 14, characterized in that said location is associated with a plurality of LANs containing said LAN, wherein the local link training features include an SSID of each LAN of the plurality of LANs and/or a MAC address of a point wireless access point associated with each LAN of the plurality of LANs, and/or signal strength of the wireless access point associated with each LAN of the plurality of LANs. 16. The method according to claim 1, characterized in that, prior to the training stage, it additionally includes performing a feature determination procedure that determines a set of features that are part of the set of local link training features and for which larger weight coefficients are assigned by the MLA algorithm at the training stage than for the rest of the set of training features of the local connection. 17. A method for determining the conversion rate of a targeted message associated with a POI, executed on a server containing a processor and a long-term memory associated with the processor, which stores instructions executed by the processor to execute the method, and including: - determination by the server at the use stage of a plurality of users to whom, within a first predetermined time period, a targeted message associated with a POI was shown, each user of the plurality of users using a wireless device containing a local communication module and having a wireless device identifier associated with local communication module; - receiving at the server at the stage of using a plurality of sets of signs of local communication of local communication modules of wireless devices based on the identifiers of wireless devices of a plurality of users; - input by the server into the MLA algorithm as input of the stage of using a plurality of sets of features of local communication of wireless devices; - receiving by the server as the output of the use phase from the MLA algorithm a set of indications based on a set of local communication feature sets of local communication modules of wireless devices, while an indication from the set of indications is associated with a user from a set of users and indicates whether this user has visited the POI within a second predetermined time period after showing the targeted message to him; - determination by the server at the stage of use, based on a set of indications, of a conversion rate parameter indicating the conversion rate of the targeted message, which is the percentage of the set of users who visited the POI within the second predetermined time period after they were shown the targeted message. 18. A method for determining a site attendance growth metric (PVL) performed on a server containing a processor and a long-term memory associated with the processor, which stores instructions executed by the processor to execute the method, and including: - determining by the server the visitors_site parameter equal to the number of users who (a) were shown a targeted message directing users to the POI, and who (b) visited the POI after showing them the targeted message; - determination by the server of the bypassers_site parameter, equal to the number of users who did not visit the POI after showing them a targeted message; - determination by the server of the visitors_non-site parameter, equal to the number of users who visited the POI without first showing them a targeted message; - determination by the server of the bypassers_non-site parameter equal to the number of users who were not shown the targeted message and who did not visit the POI; And - determination by the server of the PVL metric by the formula

, where at least one of the parameters visitors_site, bypassers_site, visitors_non-site and bypassers_non-site is defined using the method of claim 16. 19. The method of claim 17, wherein the targeted message is a targeted online message. 20. The method of claim 17, wherein the targeted message is a targeted offline message.

Images