RU2805513C1 - Method and server for sending targeted message to user's electronic device - Google Patents

Abstract

FIELD: data transmission.

SUBSTANCE: methods and systems for sending a targeted message to a user's electronic device. The technical result of the proposed technical solution is achieved by the fact that in the claimed solution, in response to a request, the following is carried out: receiving a plurality of targeted electronic messages-candidates; receiving a plurality of templates having a specific layout of one or more targeted electronic messages; selecting by the server, for each template from the plurality of templates, a corresponding set of targeted electronic messages so as to provide an optimized set of parameters for each template from the plurality of templates.

EFFECT: improved relevance of the targeted electronic message to the user.

30 cl, 10 dwg

Description

Field of technology to which the invention relates

[1] This technology relates generally to data communications and, in particular, to methods and systems for sending a targeted message to a user's electronic device.

State of the art

[2] Various global and local networks (Internet, World Wide Web, local networks, etc.) provide the user with access to a huge amount of information. The information includes a variety of context-sensitive topics such as news and current events, maps, business information, financial information and resources, traffic information, gaming and entertainment information, and more. Users use a variety of client devices (desktop, laptop, smartphone, tablet, etc.) to access rich information content (images, audio, video, animation and other multimedia content from such networks).

[3] A user can access information on the Internet in various ways. He can directly access a specific resource by typing the resource's address on the keyboard (usually a Universal Resource Locator) or by following a link available in an email message or other Web resource. This is especially convenient when the user knows the resource he is interested in.

[4] The user can also search for a resource of interest using a search engine. The latter option is especially convenient when the user knows the topic of interest, but does not know the exact address of the resource of interest. There are many search engines available to the user. Some of them are considered universal search engines (such as Yandex™, Google™, Yahoo™, etc.). Others are considered vertical search engines that focus on a specific area, such as Momondo™, a specialized airline ticket search engine. Any search engine is usually capable of accepting a search request from a user, performing a search, and returning the user a ranked search engine results page (SERP, Search Engine Results Page).

[5] When accessing a specific website or conducting a search, the user may be presented with general purpose content or personalized content. General purpose content is content that is shown to any user who has access to a specific resource, without taking into account user parameters such as user interests, user activity history, etc. Personalized content is content that is specifically targeted to a user based on predicted or known parameters of the user's interests. Thus, personalized content can be considered targeted content.

[6] For example, the ranking of search results on any SERP page may be generic (using a general purpose search engine results ranking algorithm) or personalized to the user (i.e., based at least in part on available information about the user, such such as past queries, actions with past search results, etc.). In addition, a web resource can provide both general purpose content and personalized content to the user. Personalized content can be presented in a variety of ways, such as a targeted message embedded in website content that is selected as a relevant message for the user based on available information about the user. The targeted message may be provided by the general purpose content provider of the web property or, alternatively, it may be obtained from a third party source and inserted into the general purpose content otherwise available on the web property.

[7] For example, on a SERP page, some search results may be general purpose results ranked using a general purpose search engine results ranking algorithm, and other results may be targeted emails. These may be presented in the form of electronic banner ads or as search results associated with a “special placement” on the SERP page. This special placement may or may not correspond to the rank of these specially placed search results if their rank is determined using a general purpose search engine results ranking algorithm.

[8] Typically, such special placement is chosen to draw the user's attention to such specially placed targeted electronic message and thus trigger the user to perform a predetermined desired action. The predetermined desired action may be a user action on a targeted email message, a user visiting a resource associated with the source of the targeted email message, a user performing a predetermined action associated with an online resource, or the like.

[09] Thus, special placement of targeted electronic communication generally involves a special arrangement whereby the source of such specially placed targeted electronic message must pay a search engine service provider or electronic advertising platform service provider for such placement.

[10] In FIG. 1 illustrates a known exemplary system (not indicated) for delivering personalized content to a user 104 of an electronic device 102 via a communications network 106 . For this purpose, a content provider server 108 and a personalized content provider server 110 are connected to the communication network 106 . It can be assumed that the user 104 wishes to have access to a web resource 112 hosted on the content provider server 108 . For this purpose, the electronic device 102 may run a browser application, such as a GOOGLE Chrome™ browser, a YANDEX™ browser, a SAFARI™ browser, or the like. It may also be contemplated that user 104 may keyboard-type a URL associated with network resource 112.

[11] In response to an attempt by user 104 to access network resource 112, electronic device 102 generates a resource request 120 in accordance with an appropriate communication protocol and sends it through communications network 106 to content provider server 108. The content provider server 108 receives a resource request 120 and determines that the user 104 is searching for a network resource 112 (eg, based on a URL, using a process known as “address resolving”). The content provider server 108 determines that the network resource 112 contains a piece of general purpose content and a personalized piece of content. The content provider server 108 may also determine that the content for the general portion is stored on the content provider server 108 and the content for the personalized portion is stored on the personalized content provider server 110 . Therefore, the content provider server 108 sends a request 122 for personalized content through the communication network 106 (or through a separate network, as appropriate) to the personalized content provider server 110. The personalized content request 122 includes, among other things, an indication of the user ID and/or an indication of the user's interests associated with the user 104.

[12] Based on the user ID and/or interests of the user, the personalized content provider server 110 selects content for a targeted content chunk and sends the targeted content chunk data 124 via the communications network 106 to the content provider server 108 . The targeted content portion data 124 may, for example, contain a script for embedding into the personalized portion of the network resource 112 that, when executed on the electronic device 102, causes the personalized content to be downloaded from the personalized content provider server 110.

[13] Shown in FIG. 1, the system is capable of placing targeted email messages on user-facing SERP pages, where the content provider server 108 may be implemented as a search engine server. In these embodiments, the personalized content provider server 100 may be implemented as part of a search engine server.

Disclosure of the Invention

[14] The purpose of the present invention is to eliminate at least some of the disadvantages of known solutions.

[15] Non-limiting embodiments of the present technology relate to a method and system for delivering targeted content (ie, a targeted electronic message) to a user's electronic device. In at least some non-limiting embodiments of the present technology, the system is capable of selecting a targeted email message based on an auction model, where sources of targeted email messages bid for inclusion of their targeted email messages for display to a user (as part of a SERP page, as part of network resource, etc.).

[16] In some non-limiting embodiments of the present technology, the method and system described herein is implemented in a search and content advertising system (which are examples of targeted electronic messages). Sources of such targeted emails pay for “clicks” on targeted emails (such as banner ads, etc.), which may be displayed in ad units on the SERP page and/or through advertising networks (for example, through the Yandex Advertising Network or through advertising exchanges).

[17] In at least some non-limiting embodiments of the present technology, the “click cost” (i.e., the amount charged to the source of the targeted email message for the user “clicking” on the targeted email message or for the user to complete another predetermined desired actions in relation to a targeted email) is not fixed. Instead, the cost per click may be determined based on the results of an auction among advertisers (i.e., among multiple sources of targeted email messages that, for example, compete for placement on a SERP page). In those embodiments of the invention where a targeted email message is to be placed on a SERP page, the auction is conducted in real time while the user loads the search results page or page containing ad units.

[18] In general, the purpose of non-limiting embodiments of the present technology is to generate a set of targeted electronic messages (for a specific display to a user) based on an auction. In a sense, some non-restrictive embodiments of the present technology treat the auction as an optimization problem. In general, the goal of an optimization problem is to select a set of targeted emails that maximizes the value of the set of targeted emails. The concept of value in accordance with non-limiting embodiments of the present technology is multifaceted.

[19] Examples of elements that influence value include, but are not limited to, the following.

[20] User-dependent factors - the relevance of the targeted email message to the user.

[21] Source-dependent factors - factors associated with the source of the targeted email: predicted click-through rate (CTR, Click-Through Rate), price offers predetermined by the source, etc.

[22] Combined factors - predicted conversion (i.e., the likelihood that a user who was shown a targeted email message will take the desired action on the targeted email message or the online resource associated with the source of the targeted email message).

[23] In accordance with the first aspect of the present technology, a method of sending a targeted electronic message to a user of an electronic device is implemented. The method is performed by a server connected to an electronic device via a communication network. The method includes: the server receiving a request for a targeted electronic message to be sent to an electronic device; retrieving a plurality of targeted candidate emails from the email database; receiving multiple templates, each with a specific layout for one or more targeted email messages; selection by the server for each template from a plurality of templates of a corresponding set of targeted electronic messages containing one or more targeted electronic messages, wherein the selection is made in such a way that for each template from a plurality of templates the parameter of the total value of the corresponding set of targeted electronic messages is maximized from possible combinations among the set of targeted ones candidate emails; analysis by the server of multiple templates, each of which is assigned a corresponding set of targeted electronic messages, in order to select a template intended to be sent to an electronic device and having the highest value of the total value parameter; sending to the electronic device a template intended to be sent with a corresponding set of targeted email messages to display a set of targeted email messages using a specific layout.

[24] In some embodiments of the method, the server's selection of an appropriate set of targeted electronic messages includes executing a best-match optimization algorithm.

[25] In some embodiments of the method, the server's selection of an appropriate set of targeted emails includes, for each template from a plurality of templates: iteratively placing one of the plurality of candidate targeted emails at each of the placement positions of each template from the plurality of templates, and selecting a combination of targeted emails candidate messages that maximizes the total value parameter and determines the corresponding set of targeted electronic messages.

[26] In some embodiments of the method, the individual value parameter for one of the plurality of targeted email candidate messages placed at a given template placement position of the plurality of templates is determined based on at least one of the following parameters: the price offer associated with one of the targeted candidate emails; predicted “click” probability parameter; predicted relevance parameter; predicted parameter of the probability of a given action; and a predetermined reserved price.

[27] In some embodiments of the method, the aggregate value parameter is based on the aggregation of all individual value targets of a set of targeted email messages assigned to a given template from a plurality of templates.

[28] In some embodiments of the method, the predicted click probability parameter and/or the predicted relevance parameter and/or the predicted probability parameter of a given action is determined by a machine learning algorithm (MLA, Machine Learning Algorithm).

[29] In some embodiments of the method, the predicted click probability parameter, the predicted relevance parameter, and the predicted probability parameter of a given action are determined by the MLA algorithm.

[30] In some embodiments of the method, each of the predicted click probability parameter, the predicted relevance parameter, and the predicted probability parameter of a given action is determined by a corresponding MLA algorithm.

[31] In some embodiments of the method, the predetermined reserved price is preset by the server operator.

[32] In some embodiments of the method, the price quote associated with one of the targeted candidate emails is predetermined by the source of that one of the targeted candidate emails.

[33] In some embodiments of the method, the price quote associated with one of the targeted email candidate messages is predetermined by the source of that one of the targeted email candidate messages, independent of other prices predefined by other sources of other candidates from the plurality of targeted email messages - candidates.

[34] In some embodiments of the method, each template of the plurality of templates is associated with a weight parameter.

[35] In some embodiments of the method, the weight parameter is a replacement rate of at least some positions in a particular layout.

[36] In some embodiments of the method, the weighting parameter also indicates the relative advertising traffic coverage of at least some of the positions in a particular layout compared to at least some other positions in the particular layout.

[37] In some embodiments of the method, the weighting parameter is determined independently of the corresponding price offer associated with one of a plurality of targeted candidate emails.

[38] In some embodiments of the method, the weighting parameter is determined prior to receiving a request for a targeted electronic message to be sent to the electronic device.

[39] In some embodiments of the method, the largest value of the total value parameter is selected based on the corresponding total value parameter of each pattern from the plurality of patterns, adjusted by the weight parameter.

[40] In some embodiments of the method, the corresponding total value parameter is determined by summing the individual values of the targeted electronic messages forming a corresponding set of targeted electronic messages.

[41] In some embodiments of the method, the corresponding total value parameter is further determined by (a) adding the weight parameter to the sum of the individual value parameters of the targeted electronic messages forming the corresponding set of targeted electronic messages, or (b) multiplying this sum by the weight parameter.

[42] In some embodiments of the method, the weight parameter includes a static component and a dynamic component.

[43] In some embodiments of the method, the static component is predetermined for each template of the plurality of templates.

[44] In some embodiments of the method, the static component is predetermined based on experiment or by simulation based on historical data.

[45] In some embodiments of the method, the corresponding static component is predefined for each template of the plurality of templates using a black box optimization algorithm.

[46] In some embodiments of the method, the appropriate static component is selected by a black box optimization algorithm so as to optimize the function with a minimum threshold for the auxiliary function.

[47] In some embodiments of the method, the dynamic component is dynamically calculated for each template of the plurality of templates while the server parses the plurality of templates.

[48] In some embodiments of the method, the dynamic component for one template of the plurality of templates is calculated depending on at least (a) features of the corresponding set of targeted email messages assigned to that template of the plurality of templates, and (b) features of the request for targeted email a message intended to be sent to an electronic device.

[49] In some embodiments of the method, the dynamic component is used to filter out at least some of the templates from the plurality of templates before selecting a template to send.

[50] In some embodiments of the method, selecting a template to send includes selecting from the filtered remaining templates from the plurality of templates based on the static portion and the dynamic portion of the filtering remaining templates from the plurality of templates.

[51] In some embodiments, the method further includes determining an appropriate price to be charged to the source of each targeted email message from a set of targeted email messages forming part of a template of the plurality of templates to be sent.

[52] In some embodiments of the method, the appropriate price is calculated based on (a) the price quote and (b) the placement in a specific layout of each message from a set of targeted email messages forming part of a template of a plurality of templates to be sent.

[53] In some embodiments of the method, the appropriate price for a message from a set of targeted emails is calculated based on the substitution value of that message from the set of targeted emails relative to the remaining messages from the set of targeted emails.

[54] In some embodiments of the method, determining the appropriate price (P) for one of a set of candidates assigned to one of a plurality of templates involves applying the following formula:

[55] P = V ₁ - (TV - V ₁ ),

[56] where

[57] P - corresponding price;

[58] V ₁ is the total value of the set of candidates assigned to a given template from a plurality of templates;

[59] TV - V ₁ is the hypothetical total value of the set of candidates assigned to a given template from a set of templates, in the event of the absence of one candidate from this set of candidates.

[60] In some embodiments of the method, the specific layout includes at least one of the following: the number of targeted electronic messages included; relative placement on the screen of an electronic device of targeted electronic messages included in the corresponding template from a plurality of templates; the relative size of the targeted email messages included in the corresponding template from the plurality of templates; and information to include in each targeted email.

[61] In some embodiments of the method, the relative placement of the targeted email may be one of the following: above an "organic" search result; below the “organic” search result; on the side of the “organic” search result; as part of a suggested search query; and as part of a search result fragment.

[62] In some embodiments of the method, the information mentioned above includes at least one of the following: a text description, an image, a map, a link, and a user interaction element.

[63] In some embodiments of the method, the specific layout comprises an indication of a dynamic layout of at least some targeted electronic messages for inclusion in a corresponding template of the plurality of templates.

[64] In some embodiments of the method, the dynamic placement scheme includes a scheme for fixing the display of the message with the highest rank among the targeted electronic messages and/or a scheme for repeating the display of the message with the highest rank among the targeted electronic messages.

[65] In some embodiments of the method, the plurality of targeted email candidates comprises a first set of targeted email candidates, and retrieving the plurality of targeted email candidates from the email database includes: obtaining from the email database a second plurality of targeted emails candidate messages; and selecting from the second set of targeted email candidates a first set of targeted email candidates that is a subset of the second set of targeted email candidates, wherein the selection is made based at least on a user relevance parameter of each candidate from the second set of targeted emails -candidates.

[66] In some embodiments of the method, receiving a request for a targeted electronic message to be sent to an electronic device is performed while receiving a search query, wherein the relevance parameter is based on the relevance to the sent search query of each candidate from the second set of targeted candidate electronic messages.

[67] In some embodiments of the method, receiving a request for a targeted electronic message to be sent to an electronic device includes receiving a request from the electronic device.

[68] In some embodiments of the method, receiving a request for a targeted electronic message to be sent to an electronic device includes receiving the request from a back-up server.

[69] In some embodiments of the method, receiving a request for a targeted electronic message to be sent to an electronic device includes the server generating the request in response to a user request for access to a network resource hosted on that server.

[70] In some embodiments of the method, the targeted electronic message is an electronic banner advertisement.

[71] In some embodiments of the method, the targeted email is “inorganic” search results intended for inclusion on a SERP page.

[72] In some embodiments of the method, the steps of selecting a corresponding set of targeted emails from possible combinations among a plurality of candidate targeted emails and selecting a template from a plurality of templates to which a corresponding set of targeted emails is assigned are performed based on the principle of truth in electronic auctions.

[73] In accordance with another aspect of the present technology, a server is implemented for sending a targeted electronic message to a user of an electronic device. The server contains: a communication interface for connecting with an electronic device through a communication network, a processor that processes computer commands, the execution of which initiates the server to perform the following actions: receiving a request for a targeted electronic message intended to be sent to the electronic device; retrieving a plurality of targeted candidate emails from the email database; receiving multiple templates, each with a specific layout for one or more targeted email messages; selection for each template from a plurality of templates of a corresponding set of targeted electronic messages containing one or more targeted electronic messages, wherein the selection is made so that for each template from a plurality of templates it is possible to maximize the total value parameter of the corresponding set of targeted electronic messages from possible combinations among the set of targeted ones candidate emails; analysis by the server of a plurality of templates, each of which is assigned a corresponding set of targeted electronic messages, in order to select from a plurality of templates one template intended to be sent to an electronic device and having the highest value of the total value parameter; sending to an electronic device a template intended to be sent with a corresponding set of targeted email messages to display a set of targeted email messages using a specific layout.

[74] As used herein, the term “server” means a computer program executed by associated hardware and capable of receiving requests (eg, from electronic devices) via a network and executing or initiating execution of those requests. The hardware may be one physical computer or one computer system, which is not essential for this technology. In the present context, the expression "server" does not mean that every task (for example, a received command or request) or any specific task is accepted, executed or launched by the same server (i.e. the same software and/or hardware ). This expression means that any number of software or hardware can receive, send, execute, or initiate execution of any task or request, or the results of any tasks or requests. All of these software and hardware may be a single server or multiple servers, both of which are implied by the expression “at least one server.”

[75] As used herein, the term “electronic device” means any computer hardware capable of executing programs suitable for solving a given task. Thus, some (non-limiting) examples of electronic devices include personal computers (desktops, laptops, netbooks, etc.), smartphones and tablets, and network equipment such as routers, switches and gateways. It should be noted that in this context, a device that functions as an electronic device can also function as a server in relation to other electronic devices. The use of the expression “electronic device” does not exclude the use of multiple electronic devices to receive, send, perform, or initiate the execution of any task or request, or the results of any tasks or requests, or steps of any method described herein.

[76] As used herein, the term “database” means any structured collection of data, regardless of its specific structure, database management software, or computer hardware for storing, using, or otherwise making use of that data. The database may reside in the same hardware as the process that stores or uses the information stored in the database, or the database may reside in separate hardware, such as a dedicated server or multiple servers.

[77] As used herein, the expression “information” includes information of any kind or kind capable of being stored in a database. Thus, information includes audiovisual works (images, films, sound recordings, presentations, etc.), data (location data, numerical data, etc.), text (opinions, comments, questions, messages, etc. .etc.), documents, spreadsheets, etc., but is not limited to them.

[78] As used herein, the expression “search query result set relevance score” means the likelihood that a user submitting a search query expected to see data from the search query result set.

[79] As used herein, the expression “component” includes a designation of software (suitable for certain hardware) necessary and sufficient to perform a specific function or functions.

[80] As used herein, the expression “search query result set” refers to a list of results that is returned in response to a search query by a search engine, which may contain one or more standard or customized search modules. The search query result set may comprise a list of results returned by the web search module or one or more vertical search modules, or a combination of results returned by the web search module and one or more vertical search modules. The search query result set may also contain no results.

[81] As used herein, the expression “search engine results page” means a list of results to be displayed to a customer on an electronic device and generated by combining a set of search query results with targeted messages.

[82] As used herein, the expression “computer-readable storage medium” means media of any kind or form, including RAM, ROM, disks (CD-ROM, DVD, floppy disks, hard disks, etc.), USB - drives, solid-state drives, magnetic tape drives, etc.

[83] In the context of the present description, the numerals “first”, “second”, “third”, etc. are used only to indicate differences between the nouns they refer to, but not to describe any specific relationships between those nouns. For example, it should be clear that the use of the terms "first server" and "third server" does not imply any particular order, type, chronology, hierarchy or classification, in this case, of servers, nor that their use (in itself) does not imply the presence of a “second server” in any situation. Moreover, as occurs here in another context, reference to a “first” element and a “second” element does not preclude the possibility that the two elements may in fact be the same element. Thus, for example, in some cases the “first” server and the “second” server may be the same software and/or hardware, and in other cases they may be different software and/or hardware.

[84] Each embodiment of the present technology relates to at least one of the above-mentioned objectives and/or aspects, but not necessarily all of them. It should be understood that some aspects of the present technology associated with attempting to achieve the above purpose may not be consistent with that purpose and/or may be consistent with other purposes not expressly mentioned herein.

[85] Additional and/or alternative features, aspects and advantages of embodiments of the present technology are set forth in the following description, the accompanying drawings, and the claims.

Brief description of drawings

[86] The following description is provided for a better understanding of the present technology, as well as other aspects and their features, and should be used in conjunction with the accompanying drawings.

[87] In FIG. 1 shows a known system for delivering personalized content to an electronic device.

[88] In FIG. 2 is a diagram of a system implemented in accordance with non-limiting embodiments of the present technology.

[89] In FIG. 3 illustrates a network resource containing a piece of general purpose content and a piece of targeted content and available in the system of FIG. 2.

[90] In FIG. 4 is a screen shot of a SERP page containing general search results and targeted emails, wherein the general search results and targeted emails are arranged according to a first example layout according to non-limiting embodiments of the present technology.

[91] In FIG. 5 is a screenshot of another SERP page containing general search results and targeted emails, wherein the general search results and targeted emails are arranged according to a second example layout according to non-limiting embodiments of the present technology.

[92] In FIG. 6 is a screenshot of the one shown in FIG. 5, a SERP page containing general search results and targeted emails, wherein the general search results and targeted emails are arranged in accordance with a third example layout, which is a dynamic layout, according to non-limiting embodiments of the present technology.

[93] In FIG. 7 is a screenshot of the one shown in FIG. 6 a SERP page containing general search results and targeted emails, wherein the general search results and targeted emails are arranged in accordance with a fourth example layout, which is another dynamic layout, according to non-limiting embodiments of the present technology. .

[94] In FIG. 8 is a diagram of the process for selecting a subset of targeted messages for inclusion in a SERP page, implemented in accordance with non-restrictive embodiments of the present technology.

[95] In FIG. 9 is a diagram of a plurality of targeted candidate messages, a first template (having a first placement position and a second placement position) and a second template (having a first placement position and a second placement position), and a step of iteratively placing a plurality of targeted candidate messages at different placement positions. performed as part of the process shown in FIG. 8.

[96] In FIG. 10 is a flowchart of a method for selecting a targeted electronic message to be sent to an electronic device implemented in the embodiment shown in FIG. 2 system according to non-limiting embodiments of the present technology.

Carrying out the invention

[97] In FIG. 2 is a diagram of a system 200. It should be clear that the system 200 is only one possible embodiment of the present technology. Thus, the further 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 this technology. In some cases, useful examples of modifications to the system 200 are described. They provide insight, but also do not define the scope or boundaries of the present technology. These modifications do not constitute an exhaustive list. As will be appreciated by one skilled in the art, other modifications are likely possible. In addition, if in some cases modifications are not described (i.e., there are no examples of modifications), this does not mean that they are impossible and/or that the description contains the only possible implementation of a particular element of this technology. One skilled in the art will appreciate that this is not the case. In addition, it should be understood that the system 200 may, in some cases, provide simplified implementations of the present technology, and that such embodiments are presented to promote a better understanding. Those skilled in the art will appreciate that various embodiments of the present technology can be significantly more complex.

[98] In FIG. 2 illustrates a system 200 in accordance with one non-limiting embodiment of the present technology. System 200 includes an electronic device 202. The electronic device 202 typically communicates with a user 204 and may sometimes be referred to as a client device. It should be noted that the interaction of the electronic device 202 with the user does not need to specify or assume any mode of operation, such as login, registration, or the like. System 200 may include a variety of electronic devices similar to or different from electronic device 202.

[99] There are no particular restrictions placed on the implementation of the electronic device 202. For example, electronic device 202 may be implemented as a personal computer (desktop, laptop, netbook, etc.), wireless electronic device (cell phone, smartphone, tablet, etc.), or network equipment (router, switch, or gateway ).

[100] The general implementation of electronic device 202 is known in the art and therefore is not described in detail here. Suffice it to say that the electronic device 202 includes an input user interface (such as a keyboard, mouse, touchpad, touch screen, etc.) for receiving user input, an output user interface (such as a screen, touch screen, printer, etc.), and an output user interface (such as a screen, touch screen, printer, etc.) etc.) for outputting visual and audio information to the user 204, a network data interface (such as a modem, network card, etc.) for bidirectional data transmission through the communication network 206, and a processor coupled to the user input interface, the user interface output and a network data interface, while the processor is capable of performing various procedures, including those described below. To this end, the processor may store or have access to machine-readable instructions that, when executed, cause the processor to perform various procedures described herein.

[101] The electronic device 202 includes hardware and/or application software and/or firmware (or a combination thereof) for the electronic device 202 to execute a browser application 208. In general, the browser application 208 is designed to provide access to a user 204 over a network 206 communication to one or more network resources, such as network resource 212 hosted on network resource server 210. In some non-limiting embodiment of the present technology, network resource 212 may be a search engine, and network resource server 210 may be a search engine server.

[102] However, it should be noted that non-limiting embodiments of the present technology may be applied to other types of network resource 212 and/or other types of network resource server 210. In general, non-limiting embodiments of the invention can be implemented for any kind of network resource 212 providing targeted electronic messages selected based on auction principles as described above.

[103] System 200 also includes the aforementioned communications network 206 . In some non-limiting embodiments of the present technology, the communications network 206 may use the Internet. In other embodiments of the present technology, the communications network 206 may be implemented differently, for example, as a random wide area communications network, a local communications network, a private communications network, etc.

[104] System 200 also includes the aforementioned network resource server 210 coupled to communications network 206 via a communications link (not separately identified). The network resource server 210 may be implemented as a traditional computer server. In an example implementation of the present technology, the network resource server 210 may be implemented as a Dell™ PowerEdge™ server running the Microsoft™ Windows Server™ operating system. It will be appreciated that the network resource server 210 may be implemented using any other suitable hardware and/or application software and/or firmware, or a combination thereof. In the illustrated non-limiting embodiment of the present technology, the network resource server 210 is a single server. In other non-limiting embodiments of the present technology, the functions of the network resource server 210 may be distributed among multiple servers.

[105] Briefly, the network resource server 210 includes a communication interface (not shown) whose structure and settings allow communication with various elements (such as the electronic device 102 and other devices that may be connected to the communication network 206) through the communication network 206 . The network resource server 210 also includes at least one computer processor (not shown) associated with a communications interface that is structured and configured to perform the various processes described herein.

[106] The network resource server 210 is capable of providing access to a plurality of network resources, such as the network resource 212 shown in FIG. 2. The user 204 may, using the electronic device 202, access the network resource 212 in various ways, for example, by entering a URL associated with the network resource 212 into the browser application 208 executed by the electronic device 202, by following a link (contained in an email message, on another online resource or on a SERP page) linked to online resource 212, or by any other suitable means.

[107] As described above, the online resource 212 may be a search engine or SERP page, and may also be a website, web portal, search results page, or the like.

[108] In FIG. 3 shows the content structure of the network resource 212. The network resource 212 contains a portion 302 of general purpose content and a portion 304 of targeted content. In the embodiments described herein, general purpose content chunk 302 comprises a portion of a SERP page containing results ranked using a search engine server's organic ranking algorithm. These can be documents available on the Internet, or documents found in various verticals (for example, images, maps, films, etc.). Targeted content portion 304 contains content specifically targeted to a user (such as user 204). The targeted content piece 304 may contain search results made more visible through special placement - on electronic advertising banners and the like.

[109] In general, in other embodiments of the present technology, general purpose content chunk 302 contains content that can be displayed to any user accessing the network resource 212 without regard to user parameters such as the user's interests, user activity history, etc. P. Targeted content portion 304 contains content specifically personalized (ie, selected) for a user (such as user 204) based on predicted or known parameters of the user's interests.

[110] For example, for a newspaper portal that is a network resource 212, a general purpose content piece 302 may contain the latest news that can be shown to any user who has access to the newspaper portal, and a targeted content piece 304 may contain news related to a specific topic of interest to user 204 (i.e., targeted content piece 304 may contain a set of news stories specifically selected for a particular user, e.g., user 204). Additionally, in this example of a newspaper portal, the targeted content piece 304 may contain a targeted message, such as an advertisement specifically selected for the user 204.

[111] Although in FIG. 3 schematically shows one instance of a general purpose content chunk 302 and one instance of a targeted content chunk 304, this is not necessary for every embodiment of the present technology. It is contemplated that the network resource 212 may contain one or more instances of a general purpose content chunk 302 and one or more instances of a targeted content chunk 304.

[1112] It should also be noted that there are no special restrictions on the placement of one or more instances of a general purpose content piece 302 and one or more instances of a targeted content piece 304.

[113] In some embodiments, some or all of the targeted content chunks 304 may be visually distinct from one or more instances of the general purpose content chunk 302. In such embodiments, any piece of targeted content 304 may be positioned above, below, to the right, or left of any one or more instances of general content piece 302. In some embodiments, any piece of targeted content 304 may be positioned between any two instances of general content piece 302 (but still be visually different, e.g., in appearance, font, color, border, etc.).

[114] In other embodiments, some or all of the targeted content chunks 304 may be visually indistinguishable from one or more instances of the general purpose content chunk 302. In these embodiments, any piece of targeted content 304 can be sandwiched between any two instances of a piece of general purpose content. In particular, this approach is appropriate in (but not limited to) embodiments of the invention where the network resource 212 represents a SERP page, the general purpose content chunk 302 represents search results ranked according to a general purpose ranking algorithm, and the targeted content chunk 304 represents are search results ranked according to a user-personalized or rank-enhancing ranking algorithm.

[115] According to non-limiting embodiments of the present technology, a plurality of templates 224 stored in a template database 220 are available to the network resource server 210, as shown in FIG. 2. Generally, each template of the plurality of templates 224 is associated with a particular layout of one or more targeted electronic messages intended to be displayed in the browser application 208 as part of the displayed network resource 212 requested by the electronic device 202.

[116] According to non-limiting embodiments of the present technology, the number of patterns in the plurality of patterns 224 can be predetermined. For example, the number of patterns in the set of patterns 224 may be 10, 20, 45, or any other suitable number specified by the operator of the network resource 212.

[117] According to non-limiting embodiments of the present technology, a particular layout includes at least one of the following: a number of targeted electronic messages included; the relative placement on the screen of the electronic device 202 of targeted electronic messages included in a corresponding template from a plurality of templates; the relative size of the targeted email messages included in the corresponding template from the plurality of templates; and the amount of data included in each targeted email message.

[118] Various non-limiting embodiments of the present technology employ one of the following relative placements of the targeted email message: above the "organic" search result; below the “organic” search result; on the side of the “organic” search result; as part of a suggested search query; as part of a search result fragment.

[119] In other non-limiting embodiments of the present technology, the specific layout includes an indication of a dynamic layout of at least some targeted electronic messages to be included in a corresponding template from the plurality of templates 224. The dynamic layout may be a commit pattern displaying the highest ranked message among targeted email messages, a pattern of repeating display of the highest ranking message among targeted email messages, or a combination thereof.

[120] Described below are various non-limiting examples of specific layouts that may be defined by a plurality of templates 224.

[121] In FIG. 4 shows a first example of a screenshot 400. Screenshot 400 shows an online resource 212 as a SERP page 402 displayed on the screen of an electronic device 202. The SERP page 402 contains a variety of search results 404. The network resource 212 generated a search result set 404 in response to a search query received from the electronic device 202 and containing at least one search term indicative of the search intent of the user 204.

[122] The search result set 404 includes a first organic search result 406 and a second organic search result 408, which are shown in FIG. 4 shaded figures. The search results set 404 also contains a set of targeted messages, namely, a first targeted email 410, a second targeted email 412, and a third targeted email 414. The first targeted email 410, second targeted email 412, and third targeted email 414 are located in according to the layout specified by the corresponding template from the plurality of templates 224.

[123] In this example, the layout specifies the number of targeted emails to be included, namely three targeted emails (first targeted email 410, second targeted email 412, and third targeted email 414).

[124] The layout also specifies the relative position of the targeted electronic messages included (first targeted electronic message 410, second targeted electronic message 412, and third targeted electronic message 414).

[125] In this example, the highest ranking targeted email (first targeted email 410) is placed at the very top of the 402 SERP above all “organic” search results. Two other targeted emails (second targeted email 412 and third targeted email 414) are placed adjacent to each other between the organic search results (namely, located between the first organic search result 406 and the second organic search result 408 ).

[126] The layout also specifies the relative size of the targeted electronic messages included (first targeted electronic message 410, second targeted electronic message 412, and third targeted electronic message 414). In this example, the highest ranked targeted email (first targeted email 410) is relatively larger in size compared to the other two targeted emails (second targeted email 412 and third targeted email 414).

[127] In this example, the second targeted email 412 and the third targeted email 414 are of a similar size, smaller than the size of the first targeted email 410. It should also be noted that the size of the first targeted email 410 is generally similar to the size of the "organic" search results ( the first organic search result 406 and the second organic search result 408).

[128] The above description is for illustrative purposes only. The number of general purpose search results, the number of targeted emails, and the placement and size of general search results and targeted emails may vary.

[129] In FIG. 5 shows a second example screenshot 500. Screenshot 500 shows an online resource 212 as a SERP page 502 displayed on the screen of an electronic device 202. The SERP page 502 contains a variety of search results 504. The network resource 212 generated a search result set 504 in response to a search query received from the electronic device 202 and containing at least one search term indicative of the search intent of the user 204.

[130] The search results set 504 includes a first organic search result 506 and a second organic search result 508, shown in FIG. 5 shaded figures. The search results set 504 also contains a set of targeted messages, namely, a first targeted email 510 and a second targeted email 512. The first targeted email 510 and the second targeted email 512 are arranged in accordance with a layout defined by another corresponding template from the plurality of templates 224.

[131] In this example, the layout specifies the number of targeted emails included, namely two targeted emails (first targeted email 510 and second targeted email 512).

[132] The layout also specifies the relative position of the targeted electronic messages included (first targeted electronic message 510 and second targeted electronic message 512). In this example, the highest ranking targeted email (first targeted email 510) is placed at the very top of the 502 SERP page above all “organic” search results. Other targeted emails (second targeted email 512) are placed between the organic search results (namely, between the first organic search result 506 and the second organic search result 508).

[133] The layout also specifies the relative size of the targeted electronic messages included (first targeted electronic message 510 and second targeted electronic message 512). In this example, the highest ranked targeted email (first targeted email 510) is the same size as another targeted email (second targeted email 512).

[134] It should also be noted that the size of the first targeted email 510 and the second targeted email 512 is generally similar to the sizes of the organic search results (the first organic search result 506 and the second organic search result 508).

[135] In this alternative example, the layout also specifies the amount of data included in the targeted electronic messages (first targeted electronic message 510 and second targeted electronic message 512). In the illustrated example, the first targeted email 510 contains a data set 520, and the second targeted email 512 contains a data set 522. Obviously, data set 520 is larger than data set 522. In the illustrated example, the first targeted email 510 additionally contains an image 524 that is not present in the second targeted email 512.

[136] Examples of various elements that may form the first data set 520 (or the second data set 522) include a URL, text fragment, representative links, quick links, address, other contact information, etc. Image 524 may be a representative image from a corresponding resource, a map, or the like.

[137] Dynamic layouts are described below. Clearly, SERP page 502 contains many more search results than the four results shown in FIG. 5. The user 104 can scroll the screen image in direction 580, thereby displaying additional search results not shown in screenshot 500. Traditionally, when the user 104 scrolls, the search results shown in FIG. 5 will be sequentially hidden in the direction opposite to that of 580, and new results will sequentially appear.

[138] According to non-limiting embodiments of the invention, a template of the plurality of templates 224 may define a dynamic placement pattern as a fixation pattern. In this example, the highest ranking result (the first targeted email 510 in this case) will be captured at the top of the SERP page 502 even if the user 104 scrolls in the direction 580. This is shown in FIG. 6, which shows a screenshot 500 of a SERP page 602 containing an updated search result set 604 and representing the version of the SERP page 502 after the user scrolled 104.

[139] As shown in FIG. 6, SERP page 602 displays the first targeted email 510 in the top position and a set of 606 new “organic” search results. It should be noted that as the user 104 scrolls, the first targeted email may remain at the top of the SERP page 602 while the user is presented with additional search results.

[140] It should also be noted that although FIG. 6 shows a SERP page 602 containing a set 606 of new organic search results, the SERP page 602 may also contain additional targeted email messages placed between the results from the set 606 of new organic search results.

[141] Another dynamic placement pattern may specify a pattern to repeatedly display the highest ranked targeted email (the first targeted email 510 in this case).

[142] This is shown in FIG. 7, which shows a screenshot 500 of a SERP page 702 containing an updated search results set 704 and representing the version of the SERP page 502 after the user has scrolled 104 compared to the SERP page 602.

[143] As shown in FIG. 7, the SERP page 702 displays a first targeted email 510 located between the first set of organic search results 710 and the second set 712 of organic search results. It should be noted that as the user 104 scrolls, the first targeted email 510 is hidden while additional sets of “organic” search results are displayed. In other words, in these embodiments, the highest ranking targeted email may be repeatedly shown (one or more times) to the user 104 when the user 104 scrolls through the search results.

[144] It should be noted that various example layouts defined by the plurality of templates 224 are provided for illustration purposes only. Other types of placement schemes are possible. Various elements of layouts can also be mixed and matched.

[145] Given the various examples presented above, it is clear that different placement patterns provide different advertising reach for the different targeted email messages included in the email mix. In the case of the example shown in FIG. 4, the first targeted email 410 has greater advertising reach compared to the second targeted email 412 and the third targeted email 414. This greater advertising reach is achieved at least through higher (i.e., more prominent) placement and larger size. compared to other targeted emails.

[146] In the case of the example shown in FIG. 5, the first targeted email 510 has greater advertising reach compared to the second targeted email 512. This greater advertising reach is achieved at least due to higher (i.e., more prominent) placement and more data included in the first targeted email 510.

[147] At least some non-limiting embodiments of the present technology are based on the developers' understanding that greater advertising reach can result in more traffic associated with a targeted email message. In other words, the greater the advertising reach of a targeted email message, the greater the likelihood that the user 104 will perform the desired action in relation to the targeted email message. The desired action may be a “click” on a targeted email, an action performed on a resource to be presented after the user 104 “clicks” on a targeted email, etc.

[148] Whereas at least some non-limiting embodiments of the invention are implemented in an environment where targeted electronic messages are selected based on an auction principle, where the selection of targeted electronic messages depends at least on the price bid associated with the targeted electronic message message, it should be noted that the price quote for a targeted email is provided by the source of the targeted email and typically represents the price that source is willing to pay for the targeted email's advertising reach event. An advertising outreach event may correspond to the display of a targeted email message to a user (such as user 104) or the user 104 performing a “click” (or other action) on the targeted email message.

[149] At least some non-limiting embodiments of the present technology are based on the assumption that the source of a targeted electronic message generates a price quote independent of other sources of other targeted electronic messages. In other words, at least some non-limiting embodiments of the present technology are based on the assumption that the source of the targeted electronic message forms a price proposal based on various factors specific to that source (such as financial factors, the competitive environment in which the source operates, other commercial considerations, etc.).

[150] It should also be noted that in at least some non-limiting embodiments of the present technology, targeted electronic messages are selected based on the principle of truth in electronic auctions. In other words, targeted emails are selected to provide fairness to the source of the targeted email, the user 104, and the auction platform (such as online resource 212). At least some non-limiting embodiments of the present technology are based on the assumption that if targeted electronic messages are selected based on the truthfulness principle of electronic auctions, then the sources of targeted electronic messages will be “truthful” in selecting their respective price offer.

[151] According to non-limiting embodiments of the present technology, each pattern of the plurality of patterns 224 may be associated with a weight parameter. In some non-limiting embodiments of the present technology, the predetermined weighting parameter represents the replacement rate of at least some positions in a particular layout.

[152] In the case of the example shown in FIG. 4, the first targeted email 410 has greater advertising reach compared to the second targeted email 412 and the third targeted email 414. Thus, the first targeted email 410 will displace relatively more traffic. In other words, the first targeted email 410 will likely receive more traffic compared to the second targeted email 412 and the third targeted email 414. It can be assumed that the 402 SERP page does not contain the first targeted email 410, but only the second targeted email email 412 and third targeted email 414. In this hypothetical situation, the replacement value would be less because the second targeted email 412 and third targeted email 414 likely have less advertising reach than the first targeted email 410.

[153] Thus, the predetermined weighting parameter can also be said to indicate the relative advertising reach of traffic for at least some of the positions in a particular placement compared to at least some other positions in the particular placement.

[154] According to non-limiting embodiments of the present technology, a predetermined weight parameter is determined before performing the methods described below. It should be noted that the predetermined weighting parameter is determined independently of the price offer associated with the plurality of targeted candidate emails.

[155] As described in more detail below, the weight parameter is used to determine the largest total value parameter value for each template from the plurality of templates 224, which in turn is used to select a particular template from the plurality of templates 224 for use.

[156] As shown in FIG. 2, system 200 also contains a database 230 of targeted electronic messages. The targeted email database 230 stores a plurality of targeted emails 238 . According to non-limiting embodiments of the present technology, the plurality of targeted electronic messages 238 may be considered as a plurality of candidate targeted electronic messages to be sent to the electronic device 202.

[157] A plurality of 238 targeted electronic messages may be provided by a plurality of sources (not shown) of a plurality of 238 targeted electronic messages. To this end, each source interested in having their targeted email displayed by the network resource 212 may provide a targeted email body 234 and a price quote 236. The targeted email body 234 contains all of the information that could potentially be presented to the user 204. depending on the specific template used from the plurality of templates 224. The body 234 of the targeted email may also contain a targeting template, i.e. an indication of the properties or characteristics of the user 204 to whom the corresponding targeted email message is to be presented.

[158] Corresponding to non-limiting embodiments of the present technology, the process of selecting a subset of targeted messages for inclusion in the advertising set generated by the network resource 212 is not described here. The example below is based on a web resource 212, which is a search engine capable of generating SERP pages in response to user-submitted search queries.

[159] In FIG. 8 illustrates a process 800 for selecting a subset of targeted messages for inclusion on a SERP page.

[160] Initiating the process of selecting one or more targeted emails

[161] Process 800 begins with subprocess 802, where network resource 212 receives a request for a targeted email message to be sent to electronic device 202. In this example, the request arrives as part of a search query submitted by user 204 containing one or more search terms indicating the target search by user 204. The search purpose can be anything and can be represented as a free-form search query entered from the keyboard by user 204, such as “Cheap flights from Paris to Nice” or “Latest news about the fire at Notre Dame Cathedral.”

[162] However, in alternative embodiments of the present technology, subprocess 802 may be implemented differently.

[163] For example, receiving a request for a targeted electronic message to be sent to electronic device 202 may be implemented by receiving a request from a back-up server (such as a back-up advertising platform or the like).

[164] In other embodiments of the invention, receiving a request for a targeted electronic message to be sent to electronic device 202 can be implemented by having the network resource 212 itself generate a request simply in response to a user 204 request to access the network resource 212 (for example, by contacting by URL).

[165] Selecting Targeted Email Candidates

[166] Next, during execution of subprocess 804, the network resource 212 receives a plurality of target email candidates from the target email plurality 238.

[167] In some non-limiting embodiments of the present technology, network resource 212 receives and further processes (as described below) the entire set of targeted email messages 238 stored in the targeted email database 230 .

[168] In other non-limiting embodiments of the present technology, network resource 212 receives a plurality of targeted email messages 238 from targeted email database 230 and filters them for further processing.

[169] In some non-limiting embodiments of the present technology, filtering is performed to select a subset of a plurality of targeted candidate emails. In some embodiments, such selection is made based at least on the user relevance parameter of each candidate from the second set of targeted candidate emails. This selection may be made based on the user profile associated with each user and indicating the interests and socioeconomic or demographic characteristics of the user.

[170] In other embodiments of the present technology, most but not exclusively applicable, receiving a request for a targeted electronic message to be sent to an electronic device is performed at the time the search query is received, and the relevance parameter is based on the relevance of each subset of the set to the sent search query targeted candidate emails.

[171] Selecting a Subset of Targeted Emails for Each Template from a Set of 224 Templates

[172] Next, during execution of subprocess 806, the network resource 212 receives from the template database 220 a plurality of templates 224, each of which has a corresponding specific layout of one or more targeted e-mail messages (examples described above with reference to FIGS. 4-7) .

[173] When executing subprocess 806, the network resource 212, for each template from the plurality of templates 224, selects a corresponding set of targeted emails containing one or more targeted emails. In general, one or more targeted emails for each template from the plurality of templates 224 are selected such that, for each template from the plurality of templates, the total value parameter of the corresponding set of targeted emails is maximized from the possible combinations among the plurality of candidate targeted emails.

[174] In some embodiments of the present technology, selecting an appropriate set of targeted electronic messages includes running a best-fit optimization algorithm. This algorithm is also known as the assignment problem or the generalized assignment problem. This optimization problem is similar to the task of assigning workers to various jobs. There are various ways to solve such optimization problems, for example, the Hungarian algorithm, also known as the Kuhn-Munkres algorithm or the Munkres assignment algorithm.

[175] Specifically, selecting an appropriate set of targeted email messages includes, for each template from a plurality of templates 224: iteratively placing one of a plurality of candidate email targeted messages at each of the placement positions of each template from a plurality of templates 224, and selecting a combination of targeted emails -candidates that maximize the total value parameter and determine the appropriate set of targeted electronic messages.

[176] This process is shown schematically in FIG. 9, which includes a plurality of targeted email candidates 902, a first template 904 (comprising a first placement item 908 and a second placement item 910) and a second template 906 (containing a first placement item 912 and a second placement item 914). The iterative placement process is represented by 920 lines (some are solid and some are dashed for better understanding).

[177] In some non-limiting embodiments of the present technology, the individual value parameter of a targeted electronic message placed at the placement position of a particular template from the plurality of templates 224 is calculated based on at least some of the following values:

- a price offer associated with one of the targeted candidate email messages;

- predicted “click” probability parameter;

- predicted relevance parameter;

- predicted parameter of the probability of a given action;

- a predetermined reserved price.

[178] The predicted Click-Through Rate (CTR) may be the original value of the predicted rate or may be calculated as a function of the predicted rate (e.g., the predicted rate adjusted using a weighting parameter, additive parameter, or other functions).

[179] In some non-limiting embodiments of the present technology, the predicted click probability parameter and/or the predicted relevance parameter and/or the predicted probability parameter of a given action is determined by the MLA algorithm. In some non-limiting embodiments of the present technology, the predicted click probability parameter, the predicted relevance parameter, and the predicted action probability parameter are determined by the MLA algorithm.

[180] In some non-limiting embodiments of the present technology, each of the predicted click probability parameter, the predicted relevance parameter and the predicted probability parameter of a given action is determined by an appropriate MLA algorithm.

[181] In some non-limiting embodiments of the present technology, a predetermined reserved price is preset by the server operator.

[182] In some non-limiting embodiments of the present technology, the price quote associated with one of the targeted email candidates is predetermined by the source of that one of the targeted email candidates. In some non-limiting embodiments of the present technology, a price quote associated with one of the target email candidate candidates is predetermined by the source of that one targeted email candidate, independent of other prices predefined by other sources of other candidates from a plurality of targeted email candidates. candidate messages.

[183] In some non-limiting embodiments of the present technology, a predicted click probability parameter is predicted by a trained MLA algorithm based on a variety of factors, which may include historical data from past auctions, attributes of a targeted candidate email, and the like.

[184] In some non-limiting embodiments of the present technology, a predicted relevance parameter is predicted by a trained MLA algorithm based on a variety of factors associated with the targeted candidate email and/or user 204. These features may include a profile of interests of the user 204, socio-demographic parameters of the user 204, browsing history of the user 204, search history of the user 204, etc.

[185] In some non-limiting embodiments of the present technology, the predicted probability parameter of a given action is predicted by a trained MLA algorithm. The predicted probability parameter of a given action may be a conversion parameter. The conversion parameter indicates the likelihood of a user 204 performing a desired specified action after the user 204 is shown a targeted candidate email.

[186] In some non-limiting embodiments of the present technology, a Total Value Parameter (TVP) is defined as a function or by combining all targeted emails from a set of targeted emails assigned to a template.

[187] In the specific example shown in FIG. 9, as a result of execution of subprocess 806, the following sets of targeted electronic messages were determined.

[188] For the first pattern 904:

- the first placement item 908 is assigned a targeted email message C;

- Targeted email message A is assigned to the second placement item 910.

[189] For the second template 906:

- the first placement item 912 is assigned a targeted email message E;

- the second placement item 914 is assigned a targeted email message B.

[190] It should be noted that in non-limiting embodiments of the invention, the calculation of individual value parameters may be based on a linear function based on the price offer. It should be noted that in at least some embodiments of the present technology, the linear coefficient is independent of the price offer.

[191] Thus, in some non-limiting embodiments of the present technology, the calculation of individual value parameters may be based on the following linear function:

VP = A⋅t + B, (1)

where VP is the parameter of individual value; A - predicted CTR value or, in general, can be calculated depending on the CTR value; t - price offer; B is a linear coefficient that does not depend on the price offer.

[192] The linear coefficient may be based on one or more of the following parameters:

- predicted “click” probability parameter;

- predicted relevance parameter;

- predicted parameter of the probability of a given action;

- a predetermined reserved price.

[193] In some non-limiting embodiments of the present technology, the total value parameter of a corresponding set of targeted emails from possible combinations among a plurality of candidate targeted emails is calculated as the sum of the value parameters of each individual targeted email placed in the placement position of one of the plurality of templates . In other words, the targeted email message can be said to be selected for placement position of one of the plurality of templates 224 based on the “local” maximization of the value parameter.

[194] Selecting a winning template from a plurality of 224 templates

[195] Next, during subprocess 808, the network resource 212 analyzes the plurality of templates 224, each of which has been assigned a corresponding set of targeted emails, to select the winning template from the plurality of templates that has the highest overall value parameter value. According to non-limiting embodiments of the present technology, a winning template from a plurality of templates may also be considered a “template to send to electronic device 202” selected by using the processes and methods described herein.

[196] It should be noted again that each pattern of the plurality of patterns 224 is associated with a weight parameter. Thus, in some non-limiting embodiments of the present technology, the highest value of the total value parameter is selected based on the corresponding total value parameter of each template from the plurality of templates, adjusted by the weight parameter.

[197] In particular, in some non-limiting embodiments of the method, the corresponding overall value parameter is determined by summing the individual value parameters of the targeted electronic messages forming part of the corresponding set of targeted electronic messages. In some non-limiting embodiments of the present technology, the corresponding total value parameter is further determined by (a) adding a weight parameter to the sum of the individual values of targeted electronic messages forming part of the corresponding set of targeted electronic messages, or (b) by multiplying this sum by the weight parameter .

[198] In alternative non-limiting embodiments of the present technology, the weight parameter includes a static component (C) and a dynamic component (q).

[199] According to non-limiting embodiments of the present technology, a static component (C) may be predefined for each pattern of a plurality of patterns 224. For example, a corresponding static component (C) may be predefined for each of a plurality of patterns 224 using optimization algorithm of the “black box” type. In some non-limiting embodiments of the present technology, the static component is predetermined based on experiment or by simulation based on historical data.

[200] In these non-limiting embodiments of the present technology, the appropriate static component (C) may be selected by a black box optimization algorithm so as to optimize the function with a minimum threshold for the auxiliary function. For example, the function may represent the revenue to be generated by the operator of the network resource 212, and the supporting function may represent the satisfaction of the user 204 and/or the minimum advertising reach for the source of the targeted email. Of course, the function and the auxiliary function can be swapped.

[201] The dynamic component may be dynamically calculated for each template of the plurality of templates 224 during analysis of the plurality of templates 224 that are assigned a set of targeted emails to select a winning template.

[202] The dynamic component for a certain template from the plurality of templates 224 is calculated depending on at least (a) features of the corresponding set of targeted email messages assigned to that template from the plurality of templates 224, and (b) features of the request for the targeted email intended to send to an electronic device.

[203] In the case of the example shown in FIG. 6, it can be assumed that the image 524 is a map. In this example, the feature being analyzed could indicate whether the targeted message request is geographically related. In addition, features of the electronic device 202 may be taken into account, such as current location, normal location, and the like. In addition, attributes of the source of the electronic targeted message may be taken into account, such as the proximity of the source location to the user's current or usual location. For sources with multiple locations (such as a coffee shop chain), the map of one of those locations closest to the current location or to the user 204's normal location may be selected.

[204] In those examples where the image 524 is an image from a network resource associated with the source of the targeted email, the features considered may include the relevance of the image content to the search query sent, the quality of the image (determined using pattern recognition or other electronic analysis tools images) etc.

[205] It should be noted that the dynamic component (q) is calculated for the entire set of targeted emails assigned to a certain template from the set of templates 224. In other words, the dynamic component (q) is a parameter of the global value of the set of targeted emails assigned to that template from a variety of 224 templates.

[206] In some non-limiting embodiments of the present technology, the dynamic term (q) is calculated by an MLA algorithm trained on tagged sets of SERP pages. SERP pages can be tagged by experts based on the question: “Does it make sense to include a map as part of the SERP page for this search query?”

[207] In some non-limiting embodiments of the present technology, the dynamic component is used to filter out at least some of the templates from the plurality of templates 224 before selecting a winning template from the plurality of templates 224. In some non-limiting embodiments of the present technology, the dynamic component may compared to a predetermined threshold, and patterns from the plurality of patterns 224 for which the dynamic component is less than the predetermined threshold may be filtered out and excluded from further processing. In some embodiments of the present technology, a predetermined threshold is selected to filter out patterns from the plurality of templates 224 with an unsatisfactory global value parameter from the entire set of targeted emails assigned to one of the plurality of templates 224.

[208] In these embodiments of the present technology, selecting a winning template from the plurality of templates 224 includes selecting from the remaining templates after filtering from the plurality of templates 224, which is made based on the static component (C) and the dynamic component (q) remaining after filtering templates from a variety of 224 templates.

[209] Sending the winning template from the set of templates 224 and the corresponding subset of targeted email messages to the electronic device 202

[210] Next, during execution of subprocess 810, the network resource 212 sends to the electronic device 202 the winning template from the plurality of templates having the highest total value, with a corresponding set of targeted emails, to display the set of targeted emails using a specific layout. Various possible non-limiting examples of such display are described above using the screenshots shown in FIGS. 4-7.

[211] Determining the appropriate price to be charged to the source of one or more electronic messages included in the winning template from a plurality of templates 224

[212] Next, during execution of subprocess 812, network resource 212 determines the appropriate price to be charged to the source of each targeted email message from the set of targeted email messages forming part of the winning template of the plurality of templates.

[213] In some non-limiting embodiments of the present technology, the associated price is calculated based on (a) the price bid and (b) the placement in a specific layout of each message from a set of targeted emails forming part of a winning template from a plurality of templates. In some non-limiting embodiments of the present technology, the reserved price set for the auction is additionally taken into account when calculating the appropriate price.

[214] In some non-limiting embodiments of the present technology, the appropriate price for a message in a set of targeted emails is calculated based on the substitution value for that message in the set of targeted emails relative to the rest of the messages in the set of targeted emails.

[215] In some non-limiting embodiments of the present technology, determining the appropriate price (P) for a candidate from a set of candidates assigned to a particular template from a plurality of templates involves applying the following formula:

P = V ₁ - (TV - V ₁ ), (2)

where P is the corresponding price; V ₁ is the total value of the set of candidates assigned to this template from the set of templates; TV - V ₁ is the hypothetical total value of the set of candidates assigned to this template from a set of templates, in the event of the absence of one candidate from this set of candidates.

[216] In some embodiments of the present technology, the process may be repeated (starting with subprocess 802) for another auction instance to provide another targeted email message to another user.

[217] Trusted Auctions

[218] As briefly described above, at least some embodiments of the present technology are based on the principle of truth in electronic auctions. According to various non-limiting embodiments of the present technology, performing subprocess 806 (selecting an appropriate set of targeted emails from possible combinations among a plurality of candidate targeted emails) and subprocess 808 (selecting a template from a plurality of templates to which a plurality of targeted emails are assigned) candidates) contributes, at least in part, to compliance with the principle of fairness in electronic auctions.

[219] In at least some other non-limiting embodiments of the present technology, the selection of weighting parameters for each of the plurality of templates 224, independent of the price bid, also at least partially contributes to the integrity of electronic auctions.

[220] Given the architecture described above, it is possible to implement a method of sending a targeted electronic message to a user 204 of an electronic device 202. The method may be performed by a network resource server 210.

[221] In FIG. 10 is a flow diagram of a method 1000 implemented in accordance with non-limiting embodiments of the present technology.

[222] Step 1002: The server receives a request for a targeted electronic message to be sent to the electronic device.

[223] Method 1000 begins at step 1002, in which network resource 212 receives a request for a targeted electronic message to be sent to electronic device 202.

[224] Step 1004: Retrieving a plurality of targeted candidate emails from the email database.

[225] At step 1004, network resource 212 obtains a plurality of targeted candidate emails from the email database.

[226] Step 1006: Receive a plurality of templates, each of which has a specific layout for one or more targeted email messages.

[227] At step 1006, the network resource 212 receives a plurality of templates 224, each of which has a particular layout of one or more targeted electronic messages.

[228] Step 1008: the server selects, for each of the plurality of templates, a corresponding set of targeted emails containing one or more targeted emails, wherein the selection is made so as to, for each of the plurality of templates, maximize the total value parameter of the corresponding set of targeted emails from the possible combinations among a variety of targeted candidate emails.

[229] At step 1008, the network resource 212 selects, for each template from a plurality of templates, a corresponding set of targeted emails containing one or more targeted emails, the selection being made such that for each template from a plurality of templates 224, the total value parameter of the corresponding a set of targeted email messages from possible combinations among a plurality of candidate targeted email messages.

[230] Step 1010: The server analyzes the plurality of templates, each of which is assigned a corresponding set of targeted email messages, to select the winning template from the plurality of templates that has the highest total value parameter value.

[231] At step 1010, the network resource 212 analyzes the plurality of templates 224, each of which has been assigned a corresponding set of targeted emails, to select the winning template from the plurality of templates 224 that has the highest total value parameter value.

[232] Step 1012: Sending the winning template from the plurality of templates having the highest total value with a corresponding set of targeted emails to the electronic device to display the set of targeted emails using a specific layout.

[233] At step 1012, the network resource 212 sends to the electronic device 202 the winning template from the plurality of templates 224 having the highest overall value with a corresponding set of targeted emails to display the set of targeted emails using a specific layout.

[234] Possible changes and improvements to the above-described embodiments of the present technology may be apparent to one skilled in the art. The foregoing description is provided for illustrative purposes only and is not intended to limit the scope of the invention. The scope of protection of this technology is determined solely by the scope of the attached claims.

Claims (43)

1. A method of sending targeted electronic messages to a user of an electronic device, performed by a server connected to the electronic device through a communications network, and including: - receiving a request for a targeted electronic message intended to be sent to an electronic device, performed while receiving a search request from the user’s electronic device through a communication network; - selecting, from a plurality of targeted email messages contained in the database, a population of targeted email candidates, the selection being based on a user profile associated with said user and/or the relevance of the population of targeted email candidates to said search query; - obtaining a plurality of templates, each of which has an individual layout of one or more targeted electronic messages, wherein the individual layout includes at least one of the following: the number of targeted electronic messages included; relative placement on the screen of an electronic device of targeted electronic messages included in the corresponding template from a plurality of templates; the relative size of the targeted email messages included in the corresponding template from the plurality of templates; and the amount of data included in each of the targeted electronic messages; - selection, for each of a plurality of templates, of a corresponding set of targeted electronic messages from a set of targeted candidate electronic messages containing one or more targeted electronic messages, while for each targeted electronic message from a set of targeted candidate electronic messages, individual parameters associated with placement positions of a given targeted email in said template, wherein the individual parameter is determined according to at least a relevance parameter calculated based on at least the user profile and/or the user's browsing/search history, wherein as a corresponding set of targeted emails messages for said template is determined, through optimization, a combination of one or more targeted electronic messages, respectively placed in placement positions, for which the total parameter is maximum, and the total parameter is obtained by combining the individual parameters of targeted electronic messages in said combination, taking into account the current location electronic device; - selecting a template from a plurality of templates that has the highest value of the total parameter to be sent to the electronic device; - sending to the electronic device a selected template with a corresponding set of targeted electronic messages to display a set of targeted electronic messages using a custom layout. 2. The method according to claim 1, wherein, upon said determination, a maximum matching optimization algorithm is performed through optimization. 3. The method according to claim 1, in which, with the mentioned determination of individual parameters for each targeted electronic message from the set of targeted candidate electronic messages, iterative placement of the targeted electronic message is carried out in each of the placement positions of the mentioned template. 4. The method according to claim 1, wherein said determination of individual parameters is further based on at least one of the following: - predicted “click” probability parameter; - predicted parameter of the probability of a given action. 5. The method according to claim 4, in which the predicted click probability parameter and/or the relevance parameter and/or the predicted probability parameter of a given action is determined by a machine learning algorithm (MLA). 6. The method according to claim 4, in which the predicted “click” probability parameter, the relevance parameter and the predicted probability parameter of a given action are determined by the MLA algorithm. 7. The method of claim 4, wherein each of the predicted click probability parameter, the relevance parameter, and the predicted probability parameter of a given action is determined by a corresponding MLA algorithm. 8. The method according to claim 1, wherein each template of the plurality of templates is associated with a weight parameter. 9. The method according to claim 8, wherein the weight parameter is a replacement rate of at least some positions in the particular layout. 10. The method according to claim 8, in which the weight parameter is determined before said receipt of a request for a targeted electronic message. 11. The method of claim 8, wherein the largest value of the total parameter is selected based on the corresponding total parameter of each of the plurality of templates, adjusted for the weight parameter. 12. The method according to claim 11, in which the total parameter is obtained specifically by summing the individual parameters of the targeted electronic messages in said combination. 13. The method of claim 12, wherein the total parameter is additionally obtained by (a) adding the weight parameter to said sum of the individual parameters or (b) multiplying this sum by the weight parameter. 14. The method according to claim 8, in which the weight parameter includes a static component and a dynamic component. 15. The method of claim 14, wherein the static component is predetermined for each of the plurality of templates. 16. The method of claim 15, wherein the static component is predetermined based on experiment or by modeling based on historical data. 17. The method according to claim 15, in which the corresponding static component is preset for each of the plurality of templates using a black box optimization algorithm. 18. The method according to claim 17, in which the corresponding static component is selected by a black box optimization algorithm in such a way as to optimize the function with a minimum threshold for the auxiliary function. 19. The method according to claim 14, further comprising, before said template selection, analysis by the server of a plurality of templates, wherein this analysis comprises the step of calculating, for each of the plurality of templates, a dynamic component of the template weight depending on at least the relevance of the targeted email request to the content of the set of targeted emails selected for the template and/or the current location of the electronic device. 20. The method according to claim 19, wherein said analysis further comprises the step of obtaining a reduced set of templates by filtering from the plurality of templates those templates whose dynamic component of the weight parameter is less than a predetermined threshold, wherein said template selection is carried out from the reduced set of templates . 21. The method of claim 1, wherein the customized layout further includes information to be included in each of the targeted email messages. 22. The method of claim 21, wherein the relative placement of the targeted electronic message is one of the following: - higher than the “organic” search result; - below the “organic” search result; - on the side of the “organic” search result; - as part of a proposed search query; - as part of a search result fragment. 23. The method of claim 21, wherein said information includes at least one of a text description, an image, a map, a link, and a user interaction element. 24. The method of claim 1, wherein the custom layout comprises an indication of a dynamic layout of at least some targeted electronic messages for inclusion in a corresponding template of the plurality of templates. 25. The method of claim 24, wherein the dynamic placement scheme comprises a scheme for fixing the display of the message with the highest rank among the targeted electronic messages and/or a scheme for repeating the display of the message with the highest rank among the targeted electronic messages. 26. The method according to claim 1, wherein upon said receipt of a request for a targeted electronic message, the request is received from an electronic device. 27. The method of claim 1, wherein upon said receipt of a request for a targeted electronic message, the request is received from a secondary server. 28. The method according to claim 1, in which, upon receiving a request for a targeted electronic message, this request is generated by the server in response to a user request for access to a network resource hosted on the server. 29. The method of claim 1, wherein the targeted email is “inorganic” search results intended for inclusion in a search engine results page. 30. A server for sending a targeted electronic message to a user of an electronic device, comprising a communication interface for connecting to the electronic device via a communications network and a processor that processes computer commands that, when executed, instruct the server to carry out the method of any one of the preceding claims.

Images