RU2775561C1 - Method and system for identification of electronic devices of real clients of organizations - Google Patents

Abstract

FIELD: anti-fraud systems.

SUBSTANCE: invention relates to anti-fraud methods and systems. The method comprises: receiving from a first user electronic device a first Uniform Resource Locator (URL) from a first user browsing history; generating a first vector representation of the URL of the transaction confirmation pages corresponding to the first URL; comparing the first vector representation of the URL of the transaction confirmation pages with the template vector representation of the URL of the transaction confirmation pages from the plurality of template vector representations of the URL of the transaction confirmation pages; and in response to the fact that the difference between the first vector representation of the transaction confirmation pages URL and the template vector representation of the transaction confirmation pages URLs is below the threshold value of the difference of the vector representations of the transaction confirmation pages URLs, adding a first vector representation of the transaction confirmation page URLs to the identity database, and identifying the first user electronic device as the first client electronic device associated with the first real client of the first organization.

EFFECT: improving the accuracy of identifying the authenticity of client electronic devices.

35 cl, 10 dwg

Description

FIELD OF TECHNOLOGY

[1] The present technology relates to anti-fraud methods and systems, in particular to a method and system for identifying electronic devices of real customers of organizations.

BACKGROUND OF THE INVENTION

[2] User-generated recommendation platforms provide user-generated ratings and reviews on various organizations, their products and services. Recommendation platforms are often consulted by customers before making a purchase decision. However, some users are paid to comment and intentionally upvote or downvote organizations, even if they have not purchased anything from those organizations or otherwise interacted with the organization's services. This behavior significantly skews ratings and misleads future customers who are looking for genuine reviews and ratings on recommendation platforms.

[3] To prevent fraud and ensure the validity of the ratings, the user who provided the recommendation for the organization must be a real client of the organization. Therefore, the identity of the clients must be confirmed. The currently known technology verifies the user's credit card information and geographic location. The data provided by the user is compared with the data available in credit card services. However, this technology requires credit card processing companies to request information or otherwise access the financial data of electronic device users, which in turn may raise privacy or other issues.

SUMMARY OF THE INVENTION

[4] The purpose of the present technology is to eliminate at least some of the disadvantages of the prior art. Embodiments of the present technology may provide and/or expand the scope of approaches and/or methods for achieving the goals and objectives of the present technology.

[5] Embodiments of the present technology were developed based on the understanding by developers that the transaction confirmation page of different organizations may have similar and recognizable layouts, since many websites are created using the same template from a content management system (CMS). To identify whether a user has visited a transaction confirmation page, base repeat patterns and/or uniform resource locator (URL) masks can be identified in the browsing history of the user's electronic device.

[6] In accordance with a first broad aspect of the present technology, there is provided a computer-implemented method for identifying, among user electronic devices, client electronic devices associated with an organization's current clients, the method being executed by a server. The method comprises: receiving by the server from the first user electronic device a first Uniform Resource Locator (URL) from the first user browsing history; generating a first vector representation corresponding to the first URL; comparing the first vector representation with a template vector representation of the plurality of template vector representations, the plurality of template vector representations having been retrieved from the identity database; and in response to the fact that the difference between the first vector representation and the template vector representation is below a vector representation difference threshold, adding the first vector representation to the authenticity database.

[7] In some non-limiting embodiments of the method, the method further comprises, in response to the fact that the difference between the first vector representation and the template vector representation is below a vector representation difference threshold: designating the first user electronic device as the first client electronic device associated with the first real client of the first organization; and causing the first feedback received from the first user electronic device to be displayed on future customer electronic device displays.

[8] In some non-limiting embodiments of this method, the method further comprises generating an authentication database by: receiving, by the server, a plurality of URL template transaction confirmation pages of the actual authentication pages of the corresponding template organizations; and for each transaction confirmation template page URL of the transaction confirmation template page URLs, generating a new template vector view and storing the new template vector view in the authentication database, the new template vector view being one of the plurality of template vector views.

[9] In some non-limiting embodiments of the method, at least one of the authentication confirmation pages is an online purchase confirmation page indicating the completion of a purchase by a real customer.

[10] In some non-limiting embodiments of this method, the method further comprises, in response to the fact that the difference between the first vector representation and the first template vector representation is below a vector representation difference threshold: associating the first vector representation with the URL template scheme.

[11] In some non-limiting embodiments of this method, the method further comprises, in response to the difference between the first vector representation and the first template vector representation being below a vector representation difference threshold: generating a first URL scheme for the first URL; and adding the first URL scheme to the authentication database.

[12] In some non-limiting embodiments of the method, generating the first URL scheme comprises a first dynamic part of the first URL and a first static part of the first URL. Generating the first URL scheme may comprise identifying at least one mask of a first dynamic scheme part of the first URL scheme itself and identifying a first static scheme part of the first URL scheme itself.

[13] In some non-limiting embodiments of this method, the method may further comprise: receiving by the server from the second user electronic device a second URL; comparing the second URL with the first URL scheme; and in response to matching the second URL with the first URL scheme: designating the second user electronic device as the second client electronic device associated with the second real client of the second organization, and causing the display of the second feedback about the second organization received from the second user electronic device on the displays of electronic devices of future customers.

[14] In some non-limiting embodiments of this method, the method may further comprise: in response to the second URL being different from the first URL scheme: accessing a recommendation database; and performing one of: (i) designating the second recall associated with the second entity and received from the second user electronic device as fraudulent; and (ii) deleting the second response associated with the second entity and received from the second user electronic device from the recommendations database; and in response to the fact that the second static schema part is different from the first static schema part: generating a second vector representation corresponding to the second URL; comparing the second vector representation with a third vector representation, the third vector representation being one of a plurality of template vector representations; and in response to the fact that the difference between the second vector representation and the third vector representation is below the vector representation difference threshold: adding the second vector representation to the authenticity database.

[15] In some non-limiting embodiments of this method, the method may further comprise: receiving a second URL from the second user electronic device by the server; generating a second vector representation corresponding to the second URL; comparing the second vector representation with the first vector representation; and in response to the fact that the difference between the second vector representation and the first vector representation is below the vector representation difference threshold: adding the second vector representation to the authenticity database.

[16] In at least one embodiment, that the difference between the first vector representation and the template vector representation is below the vector representation difference threshold indicates that the first URL and the template URL share at least partially the same URL scheme. In at least one embodiment, generating the first vector representation comprises applying the Word2Vector algorithm to the first URL.

[17] The method may further comprise: in response to the fact that the difference between the first vector representation and the template vector representation is below a vector representation difference threshold: accessing a database of recommendations and indicating feedback associated with the first organization and received from the first electronic device like genuine comments. The method may further comprise: in response to the fact that the difference between the first vector representation and the template vector representation is below a vector representation difference threshold, excluding from the recommendations database a subset of reviews not associated with a plurality of real customer electronic devices.

[18] The method may further comprise: in response to the difference between the first vector representation and the template vector representation being greater than or equal to a vector representation difference threshold: accessing a recommendation database; and performing one of: (i) designating feedback associated with the first entity and received from the first user electronic device as fraudulent; and (ii) deleting the feedback associated with the first organization and received from the first user electronic device from the recommendations database.

[19] In some non-limiting embodiments of the method, the method may further comprise: prior to receiving the first URL, requesting the first user electronic device to provide the first URL from the first user browsing history. The method may further comprise: in response to the fact that the difference between the first vector representation and the template vector representation is below the vector representation difference threshold, and prior to calling the display of the first feedback: sending a feedback request to the first electronic device of the real client requesting comments about the first organization, wherein the recall request is configured to be displayed on the display of the present customer's first electronic device. The first organization may be at least one of an online store, a reservation service, and a service provider.

[20] In accordance with another broad aspect of the present technology, a computer-implemented system is provided to identify, among user electronic devices, client electronic devices associated with organizations' actual clients. The system contains: database of authenticity; and a server configured to: receive by the server from the first user electronic device a first uniform resource locator (URL) from the first user browsing history; generating a first vector representation corresponding to the first URL; comparing the first vector representation with a template vector representation of the plurality of template vector representations, the plurality of template vector representations having been retrieved from the identity database; and in response to the fact that the difference between the first vector representation and the template vector representation is below a vector representation difference threshold, adding the first vector representation to the authenticity database.

[21] In some non-limiting embodiments of this system, the server is further configured to: in response to the fact that the difference between the first vector representation and the template vector representation is below a vector representation difference threshold: designating the first user electronic device as the first client electronic device, associated with the first real client of the first organization; and causing the first feedback received from the first user electronic device to be displayed on future customer electronic device displays.

[22] In some non-limiting embodiments of the present system, the authentication database is generated by: receiving by the server a plurality of URL template transaction confirmation pages of the actual authentication pages of the corresponding template organizations; and for each transaction confirmation template page URL of the transaction confirmation template page URLs, generating a template vector view and storing the template vector view in the authentication database, the template vector view being one of the plurality of template vector representations.

[23] In some non-limiting embodiments of this system, the server is further configured, in response to the fact that the difference between the first vector representation and the first templated vector representation is below a vector representation difference threshold: to associate the first vector representation with a URL templated scheme, wherein a URL templating scheme is associated with a templating vector representation.

[24] In some non-limiting embodiments of this system, the server is further configured to, in response to the difference between the first vector representation and the first template vector representation being below a vector representation difference threshold: generating a first URL scheme for the first URL; and comparing the first URL scheme with the template URL scheme stored in the authentication database.

[25] In some non-limiting embodiments of this system, the server is further configured to identify the first static part of the first URL, the first static part of the scheme corresponding to said first static part, and the first dynamic part of the first URL.

[26] In some non-limiting embodiments of this system, the server is further configured to: identify the first mask of the first URL scheme to generate the first URL scheme; and comparing the first mask with the URL template wildcard to compare the first URL scheme with the URL wildcard.

[27] In some non-limiting embodiments of this system, the server is further configured to: in response to the fact that the first static scheme part is different from the URL template static scheme part: adding the first URL scheme to the authentication database, designating the first user electronic devices as the first client electronic device associated with the first present client of the first organization, and causing the first feedback received from the first user electronic device to be displayed on the displays of electronic devices of future clients; in response to the first static schema part matching the template static schema part and in response to the first URL schema matching the URL template: designating the first user electronic device as the first client electronic device associated with the first real client of the first organization , and causing the first feedback received from the first user electronic device to be displayed on the displays of electronic devices of future customers.

[28] In some non-limiting embodiments of the system, the server is further configured to: in response to the first static schema part matching the template static schema part and in response to the first URL scheme being different from the URL template: accessing recommendations database; and performing one of: (i) designating the first revocation received from the first user electronic device and associated with the first entity associated with the first URL as fraudulent; and (ii) deleting the first response associated with the first organization and received from the first user electronic device from the recommendation database.

[29] In some non-limiting embodiments of this system, the server is further configured to: receive a second URL from the second user electronic device; generating a second URL scheme corresponding to the second URL by identifying a second static scheme part and a second dynamic scheme part of the second URL; in response to the second static schema part matching the first static schema part and in response to the second URL schema matching the first URL schema: designating the second user electronic device as a second client electronic device associated with the second real client of the first organization , and causing the display of the second feedback about the first organization, received from the second user electronic device, on the displays of electronic devices of future customers.

[30] In some non-limiting embodiments of the system, the server is further configured to: in response to the second static schema portion being the same as the first static schema portion and the second URL schema being different from the first URL schema: accessing a recommendation database; and performing one of: (i) designating the second recall associated with the first entity and received from the second user electronic device as fraudulent; and (ii) deleting the second response associated with the first organization and received from the second user electronic device from the recommendations database; and in response to the fact that the second static schema part is different from the first static schema part: generating a second vector representation corresponding to the second URL; comparing the second vector representation with the third vector representation, the third vector representation being one of a plurality of template vector representations; and in response to the fact that the difference between the second vector representation and the third vector representation is below the vector representation difference threshold: adding the second vector representation to the authenticity database.

[31] In some non-limiting embodiments of this system, the server is further configured to: receive a second URL from the second user electronic device; generating a second vector representation corresponding to the second URL; comparing the second vector representation with the first vector representation; and in response to the fact that the difference between the second vector representation and the first vector representation is below the vector representation difference threshold: adding the second vector representation to the authenticity database.

[32] In some non-limiting embodiments of the system, that the difference between the first vector representation and the template vector representation is below the vector representation difference threshold indicates that the first URL and the template URL share at least partially the same URL scheme.

[33] In some non-limiting embodiments of this system, the server is further configured to generate a first vector representation by applying the Word2Vector algorithm to the first URL.

[34] In some non-limiting embodiments of the present system, in response to the fact that the difference between the first vector representation and the template vector representation is below a vector representation difference threshold, the server is further configured to: access a recommendation database and designate feedback associated with by the first organization and accepted from the first electronic device as genuine comments.

[35] In some non-limiting embodiments of this system, the server is further configured to, in response to the fact that the difference between the first vector representation and the template vector representation is below a vector representation difference threshold, excluding from the recommendation database a subset of comments not associated with many electronic devices of real customers.

[36] In some non-limiting embodiments of the present system, in response to the fact that the difference between the first vector representation and the template vector representation is greater than or equal to a vector representation difference threshold, the server is further configured to: access a recommendation database; and performing one of: (i) designating reviews associated with the first entity and received from the first user electronic device as fraudulent; and (ii) deleting the comments associated with the first organization and received from the first user electronic device from the recommendations database.

[37] In some non-limiting embodiments of this system, in response to the fact that the difference between the first vector representation and the template vector representation is below the vector representation difference threshold, and prior to calling the display of the first feedback, the server is further configured to: send a feedback request to the first to the present customer's electronic device requesting comments about the first organization, wherein the feedback request is configured to be displayed on the display of the present customer's first electronic device.

[38] As referred to herein, reviews (also referred to herein as “comments”) requested and provided by a user on the Recommendation Platform include, without limitation, feedback/comments (including neutral, positive, and/or negative), ratings and/or recommendations. As an example, the rating may include a number of points and/or stars (eg, 4 out of 5 or 8 out of 10). In some embodiments, the feedback may include text and/or images. Testimonials can be provided for the most relevant organization, product, service, etc.

BRIEF DESCRIPTION OF THE DRAWINGS

[39] For a better understanding of the present technology, as well as other aspects and additional features thereof, reference is made to the following description, which is to be used in conjunction with the accompanying drawings, in which:

[40] FIG. 1 illustrates a schematic representation of a system implemented in accordance with non-limiting embodiments of the present technology;

[41] FIG. 2 illustrates a non-limiting example screen shot of an e-commerce page of an e-commerce platform in accordance with non-limiting embodiments of the present technology;

[42] FIG. 3 illustrates a non-limiting example screenshot of a feedback page in accordance with non-limiting embodiments of the present technology;

[43] FIG. 4 schematically illustrates a non-limiting example of a transaction confirmation page in accordance with at least one embodiment of the present technology;

[44] FIG. 5 illustrates a portion of a user's browsing history on a user's electronic device in accordance with non-limiting embodiments of the present technology;

[45] FIG. 6 illustrates a schematic representation of a non-limiting example of a portion of an authenticity database in accordance with non-limiting embodiments of the present technology;

[46] FIG. 7 illustrates a schematic representation of an authentication application and its interaction with system databases in accordance with non-limiting embodiments of the present technology;

[47] FIG. 8 illustrates a flow diagram of a computer-implemented method for identifying, among user electronic devices, client electronic devices associated with real customer organizations in accordance with non-limiting embodiments of the present technology; and

[48] FIG. 9 illustrates a flow diagram of a computer-implemented method for identifying, among user electronic devices, client electronic devices associated with real customer organizations in accordance with other non-limiting embodiments of the present technology.

[49] FIG. 10 illustrates a flow diagram of a computer-implemented method for identifying, among user electronic devices, client electronic devices associated with real customer organizations in accordance with other non-limiting embodiments of the present technology.

[50] It is to be understood that throughout the accompanying drawings and related descriptions, the same features are identified by the same reference numerals. In addition, it should also be understood that the drawings and the following descriptions are for illustrative purposes only and that such disclosures do not provide any limitation on the scope of the claims.

DETAILED DESCRIPTION OF NON-LIMITING EMBODIMENTS

[51] With reference to FIG. 1 illustrates a schematic representation of a system 100, with system 100 being suitable for implementing non-limiting embodiments of the present technology. It should be clearly understood that the system 100 is illustrated only as an exemplary implementation of the present technology. Thus, the following description is only intended to be used as a description of illustrative examples of the present technology.

[52] The examples and conventions used in this document are intended primarily to help the reader understand the principles of this technology, and not to limit its scope to such specific examples and terms. It should be understood that those skilled in the art will be able to develop various mechanisms that, although not explicitly described in this document, nevertheless embody the principles of the present technology and are included in its essence and scope. In addition, the following description may describe implementations of the present technology in a relatively simplified manner for purposes of ease of understanding. Those skilled in the art will appreciate that various implementations of the present technology may be more complex.

[53] In addition, all statements contained herein that identify principles, aspects, and implementations of the present technology, as well as specific examples thereof, are intended to cover both structural and functional equivalents, whether or not they are currently known. time or will be developed in the future. Thus, for example, those skilled in the art will appreciate that any block diagrams herein represent conceptual views of an illustrative circuit embodying the principles of the present technology. Likewise, it will be understood that any flowcharts, flow charts, state transition diagrams, and the like represent various processes that may be as such represented on computer-readable media and executed by a computer or processor, whether or not such a computer or processor is depicted. explicitly or not.

[54] The functions of the various elements shown in the figures, including any functional unit labeled "processor", can be provided by specialized hardware, as well as hardware capable of executing software and associated with proper software. When provided by a processor, the functions may be provided by one dedicated processor, one shared processor, or multiple individual processors, some of which may be shared. In some embodiments of the present technology, the processor may be a general purpose processor such as a central processing unit (CPU) or a processor dedicated to a specific purpose such as a graphics processing unit (GPU). In addition, explicit use of the term "processor" should not be construed as referring solely to hardware capable of executing software, and may implicitly include, without limitation, digital signal processor (DSP) hardware, network processor, integrated circuit special purpose (ASIC), field programmable gate array (FPGA), read-only memory (ROM) for software storage, random access memory (RAM), and non-volatile storage. Other hardware, traditional and/or specialized, may also be included.

[55] Given these fundamentals, consider some non-limiting examples to illustrate various implementations of aspects of the present technology.

[56] System 100 includes an electronic device 102. Electronic device 102 is commonly associated with user 101 and, as such, may sometimes be referred to as a "user electronic device". It should be noted that the fact that the electronic device 102 is associated with the user 101 does not imply or imply any mode of operation such as needing to log in, needing to register, or the like.

[57] As used herein, unless otherwise indicated, "electronic device" is any computer hardware that is capable of executing software suitable for the respective task in question. Thus, some (non-limiting) examples of electronic devices include personal computers (desktops, laptops, netbooks, etc.), smartphones, and tablets.

[58] Referring again to FIG. 1, the electronic device 102 includes a persistent storage 104 of the user device. User device persistent storage 104 may span one or more storage media and typically provides storage space for computer-executable instructions executed by user processor 106. As an example, user device persistent storage 104 may be implemented as a computer-readable storage media including read-only memory (ROM), hard disk drives (HDD), solid state drives (SSD) and flash memory cards.

[59] The electronic device 102 also contains hardware and/or software and/or firmware (or a combination thereof) for executing one or more applications 108 for viewing. Generally speaking, the browsing application 108 corresponds to electronic applications available to the electronic device 102 and configured to provide access to various web pages. For example, one or more browsing applications 108 may include a Yandex.Browser™ web browsing application, an Internet Explorer™ browser, a Firefox™ browser, a Yandex.Market™ marketplace application, a Firefox™ browser mobile operating system application. Android™ etc.

[60] The Uniform Resource Locators (URLs) of the websites visited by the application 108 to be viewed on the user's electronic device are stored in the user's browsing history 110 in the permanent storage 104 of the user device. User browsing history 110 may comprise the URLs of websites visited by one or more browsing applications on one or more user electronic devices 102. It should be noted that user browsing history 110 may be physically located on electronic device 102 or elsewhere. For example, user history 110 views may be at least partially located in an online storage such as Yandex.Cloud™ storage and the like. In some non-limiting embodiments, user browsing history 110 may comprise URLs of websites visited by user 101 while using browsing applications 108 on one or more electronic devices 102.

[61] The user electronic device 102 also includes a user electronic device display 105 configured to display various information to the user 101. It should be understood that when the user 101 and the electronic device 102 “visit” a website, the website information is displayed on the display. 105 of the user electronic device of the user electronic device 102 itself.

[62] Using the browsing application 108, the user 101 can visit the websites of the e-commerce platforms. As an example, the e-commerce platform 128 may be the Yandex.Market™ e-commerce platform, the Amazon™ e-commerce platform, and the like. The e-commerce platform 128 may sell products and/or may also direct the user 101 to websites of various organizations.

[63] Entity 120 may be an online store, reservation service, online shopping platform, and/or service provider that has websites for entities that sell their products or services. An organization server 122 running an organization website is schematically illustrated in FIG. one.

[64] As an example, the organization's website may be the Beru.ru online store, the Samsung™ online store, and the like. It should be understood that the e-commerce platform may also have the same name as entity 120 and/or may be the same entity as entity 120. For example, e-commerce platforms such as the Amazon™ e-commerce platform or e-commerce www.bestbuy.com, sell their products directly (from their own online store) and provide links to other sales organizations (such as www.batteries.com or www.batteriesexpert.com). The organization referred to in this document may also operate an online store that is native to the e-commerce platform (such as, for example, www.bestbuy.com or www.amazon.ca).

[65] The user electronic device 102 has an electronic device identifier (ID), which may also be referred to as a user device ID. When visiting websites, the electronic device 102 provides a user device ID to websites that may be registered with hosting servers, such as those of organizations 122 .

[66] FIG. 2 illustrates a non-limiting example screenshot of an e-commerce page 200 in an e-commerce platform 128 that may be displayed on a display 105 of a user electronic device when a user 101 accesses the e-commerce platform 128 from a user electronic device 102. It should be noted that the implementation of page 200 e-commerce is unlimited. By interacting with the e-commerce page 200, the user 101 can purchase an item (such as the bicycle 210 illustrated in FIG. 2) online.

[67] In some embodiments, goods (eg, bicycle 210) may be purchased directly from e-commerce platform 128. In other embodiments, the products may be purchased from one or more entities 120. For example, an e-commerce platform website may include a description of the item, a list of entities that sell the item, and links to those entities. In response to activation of a link (e.g., via a click) to entity 120 on e-commerce page 200, e-commerce page 200 may redirect application 108 for viewing to entity server 122 of corresponding entity 120.

[68] It should be noted that after the user 101 has accessed the e-commerce page 200 displaying the product 210 and before accepting the purchase transaction confirmation, the display 105 of the user electronic device may display other web page(s) at any time. (s) from an e-commerce website or other website(s).

[69] Referring again to FIG. 1, the electronic device 102 includes a communications interface (not shown) for providing two-way communication with the communications network 114 via the communications link 116. In some non-limiting embodiments of the present technology, communication network 114 may be implemented as the Internet. In other embodiments of the present technology, communication network 114 may be implemented differently, such as any wide area network, local area network, private network, and the like.

[70] How link 116 is implemented is not particularly limited and depends on how electronic device 102 is implemented. By way of example only and not limitation, in those embodiments of the present technology where electronic device 102 is implemented as a wireless communication link (such as a smartphone), link 116 may be implemented as a wireless link (such as, but not limited to, a 3G network link, a 4G network link, a 5G network link, Wireless Fidelity, or Wi-Fi® for short, Bluetooth® or similar) or wired (eg Ethernet connection).

[71] It should be clearly understood that implementations for electronic device 102, link 116, and network 114 are provided for illustrative purposes only. Thus, those skilled in the art will readily understand other specific implementation details for electronic device 102, link 116, and network 114. As such, the examples provided herein above are in no way intended to limit the scope of the present technology.

[72] The platform 130 user-generated recommendations has a server 132 recommendations. Recommendation server 132 may span one or more storage media and typically provides storage space for computer-executable instructions executed by recommendation processor 134 . The recommendation server 132 has access to the recommendation database 136 . As an example, recommendation database 136 may be implemented as a computer-readable storage medium including read only memory (ROM), hard disk drives (HDDs), solid state drives (SSDs), and flash memory cards.

[73] The recommendation server 132, the organization server 122, and the databases described herein are connected to the communication network 114 via communication lines 116 .

[74] The recommendation server 132 provides a recommendation website that is configured to provide an interface for users to provide feedback on various organizations 120.

[75] FIG. 3 illustrates a non-limiting example screenshot of a feedback page 300 in accordance with at least one non-limiting embodiment of the present disclosure. The feedback page 300 may be displayed on the display 105 of the user electronic device when the user 101 accesses the recommendation server 132 from the user electronic device 102.

[76] It should be noted that the implementations of the e-commerce page 200 and the testimonials page 300 are not limited.

[77] In at least one embodiment, the e-commerce page 200 and the testimonials page 300 are associated with each other. The testimonials page 300 can be accessed from the e-commerce page 200, for example, by clicking on a button or the like. For example, the screenshot 200 shown in FIG. 2 may be the landing page of an e-commerce website for product 210. The e-commerce page 200 may have a link 215 that allows the corresponding review page 300 to be displayed on the display 105 of the electronic device 102. It should be noted that the testimonials page 300 may be part of the e-commerce website landing page for product 210, and clicking on link 215 may simply cause the landing page to scroll down to display product reviews 210. In some embodiments, part of the e-commerce page 200 commerce and 300 testimonials pages can be displayed on the same page at the same time.

[78] In some embodiments, the recommendations website may be the same as an e-commerce website. In some embodiments, users provide reviews of services and/or products directly on the e-commerce website. In some embodiments, users may provide feedback on services and/or products purchased directly from the respective e-commerce platform. In some other embodiments, users may provide feedback on services and/or products purchased from organizations 120, such as to which an e-commerce website has redirected application 108 to view electronic device 102.

[79] The recommendation server 132 is configured to generate and periodically update a recommendation database 136 based on the reviews and ratings provided by the user 101 (and other users potentially using the system 100) via the reviews page 300 interface. Referring to FIG. 1, 3, for each review, the recommendation database 136 may store the organization ID 360 of the organization 120 itself, as well as the text of the review 362, including the rating provided by the user 101. In addition, the recommendation database 136 may contain information such as ID 364 corresponding user and/or ID 366 of the electronic device of the actual electronic device 102 from which said recall 362 was received. It should be noted that review 362 could be an organization review and/or a product review 210.

[80] The recommendation database 136 may also contain an authenticity label 368 that may indicate whether the user who provided the review is a real customer or a fraudulent user.

[81] A legitimate customer or, in other words, a “real customer” of organization 120 provides feedback based on his or her actual experience with the relevant organization 120, because the real customer purchased a product or service through the e-commerce platform 128 or organization 120, or interacted with organization 120 otherwise. For example, organization 120 may offer merchandise or downloads (such as, for example, music downloads) for free or as part of another online experience.

[82] However, some users provide feedback despite the fact that they are not real customers of the e-commerce platform 128 or the corresponding organization 120. Feedback from such users may be paid by the organization 120 itself to improve the ranking of that organization 120 in the reference data base 136 . Alternatively, an organization's competitor may pay some users to provide negative feedback. Such false reviews distort the information on the 130 recommendations platform and should be identified and eliminated in a timely manner.

[83] The system and method described herein is configured to identify whether the user 101 who provided the feedback is indeed a true customer, that is, to identify whether he/she purchased goods and/or services on the e-commerce platform 128 or from relevant organization 120.

[84] Referring again to FIG. 1, system 100 includes an authentication server 140 connected to a communication network 114. The authentication server 140 may be implemented as a computer server. In an exemplary embodiment of the present technology, the authentication server 140 may be implemented as a Dell™ PowerEdge™ server running a Microsoft™ Windows Server™ operating system. Needless to say, the authentication server 140 may be implemented in any other suitable hardware and/or software and/or firmware, or combinations thereof. In the illustrated non-limiting embodiment of the present technology, the authentication server 140 is a one-of-a-kind server. In alternative non-limiting embodiments of the present technology, the functionality of the authentication server 140 may be distributed and may be implemented across multiple servers.

[85] The authentication server 140 includes a communication interface (not shown) structured and configured to communicate with various entities (such as electronic device 102 and other devices potentially associated with communication network 114) via communication network 114.

[86] The authentication server 140 also includes an authentication server memory 142 that contains one or more storage media and typically provides storage space for computer-executable program instructions executed by the authentication server processor 144. For example, the authentication server memory 142 may be implemented as a tangible computer-readable storage medium including Read Only Memory (ROM) and/or Random Access Memory (RAM). The authentication server memory 142 may also include one or more latched storage devices in the form of, for example, hard disk drives (HDDs), solid state drives (SSDs), and flash memory cards.

[87] In accordance with non-limiting embodiments of the present technology, the authentication server 140 is configured to execute the revocation authentication application 146 .

[88] The authentication server 140 can be connected to the recommendation database 136 directly. In alternative embodiments, the recommendation database 136 may be communicatively connected to the authentication server 140 via the communications network 114 . While the recommendation database 136 is schematically illustrated herein as a single entity, it is contemplated that the recommendation database 136 can be executed in a distributed fashion.

[89] When a real customer visits an organization's website and makes a purchase, the user's browsing application 108 is typically redirected to a transaction confirmation page (also referred to herein as an "authentication page").

[90] FIG. 4 illustrates a non-limiting example of a transaction confirmation page 400 in accordance with at least one embodiment of the present technology.

[91] In some web design patterns, the transaction confirmation page 400 may also be referred to as a "thank you page". In at least one embodiment, the transaction confirmation page 400 is an online purchase confirmation page indicating completion of a purchase by a real customer. The fact that the user 101 has made a purchase and is therefore a genuine customer can be confirmed if the URL of such a transaction confirmation page 400, ie. The transaction confirmation page URL 411 can be found in the user history 110 views.

[92] In order to determine whether a particular user is a fraudulent user or a real client, the authentication server 140 is configured to check whether the user history 110 has a transaction confirmation page URL 411 . In order to recognize the transaction confirmation page URL 411 in the user's browsing history 110, the authentication server 140 needs to parse the URLs stored in the user's browsing history 110.

[93] It should be noted that the purchase transaction may be processed by the e-commerce platform 128 and/or the organization's server 122. For example, the user viewer application 108 may display an e-commerce page 200 with information about the bicycle 210. As described above, the e-commerce page 200 may have links to various organizations that may sell the bicycle 210. In some embodiments, the e-commerce page 200 may provide comparison of prices and/or reviews of relevant organizations 120, for example, page 300 reviews. The user may follow one or more of the links provided on the e-commerce web page 200 to access the web pages of one or more organizations 120.

[94] If a user purchases a bicycle 210 from an organization 120 (eg, biking.ru online store or europebike.com online store), the viewer application 108 displays a transaction confirmation page 400 . In this case, the browsing history 110 of the user electronic device 102 stores the URL 411 of the transaction confirmation page. It should be noted that the transaction confirmation page URL 411 may have the domain name of the organization 120. In some embodiments, the transaction confirmation page URL 411 has a different domain name than the domain name of the corresponding organization 120 from which the user 101 purchased the item.

[95] Alternatively, the user 101 may purchase the bike 210 directly from the e-commerce website, and the transaction confirmation page URL 411 may have the domain name of the e-commerce website (eg, the domain name "amazon.ca").

[96] FIG. 5 illustrates a portion 500 of a user browsing history 110 on a user electronic device 102 in accordance with non-limiting embodiments of the present technology.

[97] As illustrated in FIG. 5, the user story 110 views may have, among other URLs, the following URLs: URL2 512 (“https://www.amazon.ca/”) and URL3 513 (“https://www.amazon.ca/EUROBIKE-mountain -folding-suspension-bicycle/dp/B0757J74FR/ref=sr_1_3_sspa?keywords=bike&qid=1562271743&s=gateway&sr=8-3-spons&psc=1” or just “https://www.amazon.ca/eurobike-mountain-folding- suspension-bicycle”).

[98] If a transaction (for example, buying a bike 210 from the europebike.com online store) was confirmed and the transaction confirmation page 400 was displayed, the user history 110 views also contains the URL of the transaction confirmation page 400: URL1 411 (“https:// www.europebike.com/cart/show-uid=123456789/9a87v654321asdf”).

[99] The transaction confirmation page URL 411 represents the URL of the transaction confirmation page 400 to which the application 108 for viewing on the electronic device 102 was redirected by the organization's server 122 to confirm the purchase transaction. If the transaction confirmation page URL 411 is present in the user's browsing history 110, the browsing application 108 displays the transaction confirmation page 400 and this confirms that the user 101 purchased the bike 210 from the respective organization 120 (such as the europebike.com online store in this non-limiting example). ).

[100] Still referring to FIG. 5, in at least one embodiment, the URLs 411, 512, 513 in the user's browsing history 500, and in particular the transaction confirmation page URL1 411, each have a static part 531 and may have a dynamic part 532 of the URL.

[101] Static part 531 URL contains the web address of the organization, such as, for example, <domain name> (for example, "europebike.com" or "biking.com"). In some embodiments, the implementation of the static part 531 may also include another term that is static. For example, the static part 531 may include terms such as "market", "cart", "store", and/or "transaction", etc., for example.

[102] Dynamic parts 532 of URLs 411, 512, 513 may contain a user ID 534 (eg, "show-uid=123456789" in FIG. 5) and/or a transaction ID 536 (eg, "9a87v654321asdf" in FIG. 5). In some embodiments, the user ID 534 may be included in the transaction ID 536. In some embodiments, the implementation of the dynamic part 532 contains the ID of the purchased product, so that the dynamic part 532 varies depending on the product purchased, in contrast to the domain name.

[103] In some embodiments, the dynamic part 532 of the URL may contain a combination of a user ID and/or transaction ID.

[104] The mask referred to in this document is a sequence of characters that represents part of the URL. Wildcards can be used to recognize similar URL schemes. In some embodiments, the authentication server 140 may associate a user ID with a user ID mask. The authentication server 140 may also associate a transaction ID with a mask of the transaction ID. The authentication server 140 can identify masks, such as a user ID mask and/or a transaction ID mask, in the dynamic part 532 of the URL 411.

[105] The URL scheme referred to herein is a sequence of strings that represent a URL. The URL template may contain a combination of the domain name and/or product ID of the static URL part 531, a user ID mask, and a transaction ID mask. The part of the URL scheme that corresponds to the static part of the URL may be referred to as the static part of the scheme. Another part of the URL scheme that corresponds to the dynamic part of the URL and contains one or more masks, such as a user ID mask and a transaction ID mask, may be referred to as the dynamic part of the scheme.

[106] The URL schemes of the URL 411 of the transaction confirmation pages of different organizations may be similar. Referring to FIG. 5, two transaction confirmation page URLs 411 and 540 of different organizations may have similar and recognizable URL schemes since these websites are created using the same template from content management systems (CMS). Transaction confirmation page URL schemes may contain similar user ID masks and/or transaction ID masks.

[107] In some embodiments, masks (such as, for example, user ID masks and/or transaction ID masks) may contain characters, such as, for example, special characters (such as "?", #, etc.), which can represent the corresponding parts of the URL with a mask code. For example, in such a mask code, the question mark symbol "?" can represent a single character. The asterisk character "*" can be a wildcard character (that is, one or more characters). The pound sign # can represent a single digit in the corresponding part of the URL. It should be noted that other codes and encodings may be used to represent the character portions of the URL.

[108] In a non-limiting example, URL1 411 in FIG. 5 is "https://www.europebike.com/cart/show-uid=123456789/9a87v654321asdf". The user ID is 123456789. Using the above mask code, the user ID mask may be "####*" for example. The transaction ID is "9a87v654321asdf". Using the above mask code, the Transaction ID mask could be, for example, "#?##?#####*". The URL scheme of URL1 411 in FIG. 5 is: <domain name>/cart/show-uid=<user ID mask>/<transaction ID mask>. Using wildcards, the URL scheme of URL1 411 might look like this: "www.europebike.com/cart/show-uid=####/#?##?#####*". The static part of the URL scheme of URL1 is <domain name>/cart. The dynamic part of the URL scheme of URL1 is "show-uid=####/#?##?#####*".

[109] In at least one embodiment, the authentication server 140 is configured to generate embeddings corresponding to the URL. To generate a vector representation, the authentication server 140 may apply the Word2Vector algorithm to the URL. For example, such a URL may be URL1 411. In some embodiments, the authentication server 140 may apply the Word2Vector algorithm to URL masks to generate vector representations based on the masks. The authentication server 140 may apply the Word2Vector algorithm to the URL scheme to generate vector representations based on the URL schemes. The authentication server 140 may train the neural network based on the Word2Vec algorithm such that the neural network generates vectors that are closest in multidimensional space when the associated URLs are similar (i.e., have similar URL schemes, for example) and vectors that are not. closest when the associated URLs are not similar.

[110] To determine whether the user 101 has made a purchase and therefore whether the transaction confirmation page 400 has been displayed to the user on the electronic device 102, a vector representation of the new (unconfirmed) URL may be determined and then compared to vector representations known to be they correspond to the transaction confirmation pages of the template organizations. As discussed above, the URLs of transaction confirmation pages for two different organizations with websites created using a template from a CMS can be similar. This can be determined by comparing vector representations of these transaction confirmation page URLs.

[111] Two vector representations can be considered similar when the difference between the two vector representations is less than a vector representation difference threshold. Similarly, one URL can be considered similar to the scheme of another URL (in other words, the same as the scheme of another URL) when the difference between the second URL and the scheme of the first URL is less than a vector representation difference threshold. A part of the URL can also be compared against a mask, so that they are considered similar to each other (in other words, the same as each other) when the difference between them is less than the vector representation difference threshold.

[112] In some embodiments, the difference threshold (including the vector representation difference threshold, the schema difference threshold, and the mask difference threshold) may be, for example, a difference of ten percent. For example, the schema difference threshold may correspond to a match of 9 symbols out of 10. The exact value of the difference threshold may be selected based on the desired level of accuracy of the system 100.

[1113] Returning to FIG. 1, the authentication server 140 is connected to the authentication database 150 . Authentication database 150 contains a list of template vector representations of template URLs for transaction confirmation pages of template organizations. Template organizations are organizations with well-known transaction confirmation page URLs. Transaction confirmation pages of template organizations are generated using CMS templates.

[114] In some embodiments, to create the authentication database 150, the authentication server 140 may receive a plurality of template transaction confirmation URLs, such as, for example, URL1 411, of the transaction confirmation pages of the respective template organizations. For example, the authentication server 140 may accept template URLs for marketplace confirmation pages. Amazon™, BestBuy.com™ online store, etc.

[115] In at least one embodiment, for each template URL of the plurality of template URLs of template organization transaction confirmation pages, the authentication server 140 is configured to generate a corresponding template vector representation. The authentication server 140 then stores the generated template vector representation of the template organization(s) in the authentication database 150 .

[116] FIG. 6 illustrates a non-limiting example of a portion 600 of an identity database 150 in accordance with non-limiting embodiments of the present technology. The authentication database 150 may contain template URLs 610 of transaction confirmation pages and template vector views 620 that correspond to template URLs 610 of transaction confirmation pages.

[117] Vector representations 620 may be URL-based vector representations 622, schema-based vector representations 624, and/or mask-based vector representations 626, which may be determined based on a URL, URL scheme, and/or masks, respectively.

[118] As described above, the URL scheme 641 contains a user ID mask 634 and a transaction ID mask 636 of the dynamic part of the URL. While the IDs themselves may be different for the two URLs, the URL schemes 630 and/or dynamic URL part masks of the transaction confirmation page URLs 610 may be similar.

[119] Referring again to FIG. 1, system 100 may also include a real customer database 170 that stores organization IDs and user IDs and/or electronic device IDs of real customers. After determining that the user 101 is a real customer of a particular organization, the system 100 may store the data in the real customer database 170. With such a database 170 of real customers, the recommendation server 132 may be consulted from time to time. It is also contemplated that the recommendation server 132 can perform such validation on the fly without storing anything or consulting the real customer database 170 .

[120] FIG. 7 illustrates a schematic representation of a revocation authentication application 146 and its interaction with system databases 100 and user electronic device 102 implemented in accordance with non-limiting embodiments of the present technology.

[121] The revocation validation application 146 includes an authenticity database subroutine 710, a browsing history analysis subroutine 712, and a revocation administration subroutine 714. It is contemplated that other routines may be present in the revocation validation application 146 .

[122] Authentication database routine 710 is configured to generate authentication database 150. In at least one embodiment, the authentication database 710 receives a plurality of template transaction confirmation page URLs 610 of the actual authentication pages of the respective template organizations. Then, for each transaction confirmation page template URL (eg, transaction confirmation page URL 411), the authentication database routine 710 generates a URL template vector representation 621 and stores it along with other URL template vector representations 620 in the authentication database 150 .

[123] Authentication database routine 710 may generate URL templates (eg, URL template 631) for received URL templates 610 and store them in authentication database 150 with user ID mask 634 and transaction ID mask 636 .

[124] Browsing history analysis routine 712 accesses recommendation database 136 and finds unapproved review 362 for one organization 120 received from electronic device 102 with device ID 366. Such unconfirmed feedback has not yet been verified, and it is not known whether the user 101 who provided such unconfirmed feedback 362 is a true customer of organization 120, which has organization ID 360. In other words, it is not known if the user 101 has purchased something from the organization 120 for which he provided feedback 362. For example, the recommendation base 136 may store an authenticity label 368. Authenticity label 368 may identify the corresponding response 362 as "unverified", for example. For example, other testimonials may have a label of authenticity such as "Verified Genuine" or "Verified Fraud".

[125] In some embodiments, for each user electronic device 102 and/or for each user 101 that provided a review 362 on a recommendation website and that is stored in the recommendation database 136, the browsing history analysis routine 712 analyzes the user's browsing history 110 on an electronic device corresponding to device ID 366. It should be noted that a single user 101 with a user ID stored in the recommendation database 136 may have browsing histories 110 on various user electronic devices or on the network, as discussed above, and the browsing history parsing routine 712 may be configured to parse such electronic devices in series or parallel.

[126] In some embodiments, the potential customer database 707 may store the device ID 705 of the electronic devices themselves, the authenticity of which is to be verified. Browsing history analysis subroutine 712 can access the lead database 707 and check electronic devices that match the device ID 705 . In some embodiments, the lead database 707 may store the device IDs of users who have already submitted reviews. The recommendation server 132 may collect the device IDs of users who have followed the link from the e-commerce platform to the respective organization's website.

[127] For example, for an organization that is an online store "www.lemon.com", the organization ID 360 could be "lemon.com" or "lemon". As described above, the recommendations database 136 may store an organization ID 360 (eg, "lemon.com"), a user ID 364 of the user 101 who provided the feedback, and/or an ID 366 of the electronic device from which the recommendation database 136 received the feedback.

[128] In some embodiments, the browsing history analysis routine 712 may request the user electronic device 102 corresponding to the electronic device ID 366 and/or user ID 364 to provide, at least in part, the browsing history 110. In some embodiments, the browsing history parsing routine 712 may request a URL from the user's browsing history 110 from various electronic devices 102 that match the user ID 364. The browsing history analysis routine 712 may determine if the user's browsing history 110 of the user's electronic device 102 has URLs that are similar to the template URLs 610.

[129] The browsing history analysis routine 712 is configured to receive from the user electronic device 102 URLs stored in the user's browsing history 110. For example, a new URL 540 (also referred to herein as "first URL 540") may be received from the user's electronic device 102. A non-limiting example of a new URL 540 is illustrated in FIG. 5: "https://www.lemon.com/mart/3485/333420456789". The browsing history parsing routine 712 then parses the new URL 540.

[130] In at least one embodiment, to determine if the newly received new URL 540 is a transaction confirmation page locator (URL), the browsing history parsing routine 712 is configured to generate a new vector representation 722 of that new URL 540. This new vector the representation 722 is then compared to template vector representations 620 from the authenticity database 150 .

[131] At least a portion 724 of the template vector representations 620 can be retrieved from the identity database 150 for comparison with the new vector representation 722.

[132] To compare the new vector representation 722 with the template vector representation 621, the browsing history analysis routine 712 determines the difference between the new vector representation 722 and the corresponding (i.e., URL-based) template vector representations 621 received from the authentication database 150.

[133] In some embodiments, if the browsing history analysis routine 712 determines that the difference between one of the template URL-based vector views 724 and the new URL-based vector view 722 (also referred to herein as the "URL-based vector view difference ”) is less than the vector representation difference threshold, then the electronic device 102 can be considered as a real client device.

[134] If the URL-based vector representation difference between each of the template URL-based vector representations 724 and the new URL-based vector representation 722 is equal to or greater than the vector representation difference threshold, then the electronic device 102 from which the new URL 540 was received, can be considered as an electronic device of a fraudulent user.

[135] In some embodiments, the browsing history analysis routine 712 may further parse the new URL 540 to determine if the new URL contains a static part that corresponds to the organization 120 that can be viewed in the recommendation database 136.

[136] In some embodiments, the browsing history analysis routine 712 may generate a new scheme 742 from the new URL 540. The new scheme 742 may be generated in the same form as template schemes 630. For example, if the new URL is "https:// www.lemon.com/mart/3485/333420456789", new scheme 742 could be: <static part of new URL>/<user ID mask>/<transaction ID mask>, which could be = "www.lemon.com/mart /####/############". In this new schema 742, the static part is "www.lemon.com/mart". The dynamic schema portion of the new schema 742 is "####/############".

[137] In some embodiments, to generate a new schema 742, the browsing history parsing routine 712 may identify one or more masks of the new schema 742, such as a user ID mask and/or a transaction ID mask. The user ID mask is "####" and the transaction ID mask is "############".

[138] After generating the new schema 742, the browsing history analysis routine 712 can compare the static schema part of the new URL schema 742 with the templated static schema parts of the templated schemas 630. As discussed above, the templated schemas 630 are located in the authenticity database 150.

[139] In some embodiments, the static parts of the new URL 540 can be compared directly with the templated static schema parts of the templated schemas 630. The URL, which also matches the domain name of that URL, are linked to each other. The static part of the new URL 540 may be determined at the time the new scheme 742 is generated.

[140] Thus, the browsing history analysis routine 712 can determine whether template schemas 630 have static parts similar to the new domain (eg, "lemon.com") of the new URL 540. In some embodiments, the new schema 742 can be compared to those template schemas that correspond to template vector representations found to be similar to the new vector representation 722.

[141] In some embodiments, if no template static schema portions of template schemas 730 are similar to the static schema portion of the new schema 742, then the new URL schema 742 may be added to the authentication database 150 along with the corresponding new vector representation 722 and other data. available for the new URL 540.

[142] Based on the knowledge that the new vector representation 722 is similar to at least one of the template vector representations, the electronic device 102 from which the new URL 540 was received can be considered to be the real client's electronic device. The first user electronic device may be labeled as a client electronic device associated with the current client of the respective organization (eg, "lemon.com"). The name of the respective organization may correspond to the domain name of the new URL 540. The feedback received from the client electronic device about the respective organization may be displayed on the displays of future customers' electronic devices.

[143] FIG. 8 illustrates flowchart embodiments of a computer-implemented method for identifying, among user electronic devices, client electronic devices associated with real customer organizations in accordance with non-limiting embodiments of the present technology.

[144] At step 810, a new URL 540 is received. At step 815, a new vector representation 722 of the new URL 540 is determined. The system then checks whether the new URL 540 belongs to the electronic device of a real customer or a fraudulent user that can potentially provide feedback or has already provided feedback. about the first organization.

[145] At block 820, this new vector representation 722 is compared with template vector representations 630 stored in the authenticity database 150 . If there is no template vector representation in the authenticity database 150 such that the difference between the new vector representation 722 and said template vector representation is less than the vector representation difference threshold, then the electronic device from which the new URL was received can be considered as belonging to the user - cheater (step 850).

[146] If the difference between the new vector representation and the at least one template vector representation is less than the vector representation difference threshold, the new vector representation 722 is added to the authentication database 150 in step 825. The electronic device 102 from which the new URL 540 was received, may be considered at step 860 as belonging to the real client.

[147] A new schema 742 corresponding to the new URL 540 may be generated at step 830. At step 840, the new schema 742 may be added to the authentication database 150 .

[148] If, for one matching template vector representation, the vector representation difference is less than a vector representation difference threshold, the browsing history analysis routine 712 may associate the new vector representation 722 with the URL scheme associated with the corresponding (matched) template vector representation.

[149] In some embodiments, if the difference between the new URL-based vector representation and the templated URL-based vector representation is less than the vector representation difference threshold, this may indicate that the new URL and the templated URL share at least partially the same URL scheme.

[150] In some embodiments, the implementation of the new vector representation 722, once it is confirmed that it corresponds to a genuine transaction confirmation page, can be added to the authenticity database 150. In some embodiments, when parsing browsing history 110, browsing history analysis subroutine 712 may parse only URLs that have been used over a certain period of time. For example, if the user 101 provided a review 362 for a particular date and time, the browsing history analysis routine 712 may be configured to parse only those URLs in the browsing history 110 that were saved or changed prior to the date and time of the review.

[151] If the differences between the URL vector representations from the browsing history 110 are greater than the vector representation difference threshold, then the user 101 who provided feedback 362 may be considered a fraudulent user.

[152] After the browsing history analysis routine 712 has determined whether the electronic device 102 matches a real client or a fraudulent user, the authentication server 140 revocation administration routine 714 is configured to access and modify the recommendations database 136, the real database 170. clients and/or database 180 of fraudulent users.

[153] In at least one embodiment, the feedback administration routine 714 may flag reviews from real customers as genuine. For example, a text label or image may be added to the recommendation database 136 to indicate to readers of the recommendation website that the review was provided by a real client. In at least one embodiment, in response to the fact that the difference between the unconfirmed vector representation and the template vector representation is below a vector representation difference threshold, the feedback administration routine 714 may access the recommendations database 136 and designate the comments associated with the corresponding organization and received from the first electronic device 102 of a real customer as genuine comments. This may include changing the authenticity label 369 in the recommendation database 136 .

[154] In some embodiments, the feedback administration routine 714 may add the user ID of the verified real customer of the respective organization to the real customer database 170 .

[155] In at least one embodiment, feedback administration routine 714 can request feedback by sending a comment request to the present customer's electronic device. A comment request, which asks the user to provide comments, may be displayed on the display 105 of the user electronic device 102 of the present client.

[156] In some embodiments, feedback administration routine 714 may exclude a subset of feedback not associated with real customer electronic devices that are referenced or listed in real customer database 170.

[157] If the browsing history analysis routine 712 has determined that the review 362 was added by a user who is not a genuine customer, the review administration routine 714 can access the recommendation database 136 and flag the reviews that were accepted from that user as fraudulent. The feedback administration subroutine 714 may also remove the corresponding feedback 362 from the recommendation database 136 . For example, the authenticity label 368 corresponding to the recall 362 in the recommendation database 136 can be changed.

[158] In some embodiments, reviews provided by a real customer's non-electronic device may also be used with a weight that reduces their impact, or with a label that indicates that these reviews are not verified or may be associated with fake (fraudulent) users.

[159] In some embodiments, the feedback administration routine 714 may access and exclude reviews in the recommendation database 136 that were provided by the fraudulent user.

[160] In some embodiments, if the browsing history analysis routine 712 has determined that the review 362 was added by a user who is not the actual customer, the reviews administration routine 714 may log information about that fraudulent user in the fraudulent user database 180. The data that may be stored in the fraudulent user database 180 may include, for example, a user ID and/or one or more IDs of one or more electronic devices 102 that correspond to such a fraudulent user.

[161] FIG. 9 illustrates a computer-implemented method for identifying, among user electronic devices, client electronic devices associated with real customer organizations in accordance with non-limiting embodiments of the present technology. Method 900 is executed in the system described above by an authentication server 140 that communicates with the electronic devices 102 and the servers and databases described herein.

[162] In step 910, a new Uniform Resource Locator (URL) 540 of their user browsing history 110 is received from the electronic device 102 of the questionable user by the authentication server 140 .

[163] At block 912, a new vector representation 722 is generated that corresponds to the new URL 540. In some embodiments, a new URL scheme may also be generated. To generate a new URL scheme, the validator server 140 may identify the static part 531 and the dynamic part 532 of the new URL 540. The validator server 140 may also identify the new mask 550 for the dynamic part 532 of the new URL 540.

[164] At step 914, the new vector representation 722 is compared with the template vector representation 724 from the set of template vector representations 620. In some embodiments, the implementation of the set of template vector representations 620 can be retrieved from the authenticity database 150.

[165] In step 916, the difference between the new vector representation 722 and the template vector representation 724 is determined and compared to a vector representation difference threshold.

[166] If the determined difference between the new vector representation 722 and the template vector representation 724 is below a vector representation difference threshold, the new vector representation may be associated with the URL template associated with the template vector representation 724. At block 918, the new vector representation 722 may be added to the database 150 authenticity data. At step 920, the electronic device 102 of the questionable user is marked as a new client electronic device associated with the new real client of the corresponding first organization.

[167] At step 921, the authentication server 140 may cause the response received from the first user electronic device and corresponding to the present client to be displayed on the displays of future client electronic devices.

[168] If the difference between the new vector representation 722 and the template vector representation 724 is equal to or greater than the vector representation difference threshold, the system checks at step 930 whether all template vector representations have been compared to the new vector representation. If all available template vector representations have been compared to the new vector representations, and none of the template vector representations has a difference with the new vector representation that is less than a vector representation difference threshold, recommendation database 136 is accessed at step 922. Revocation , associated with the corresponding first organization and received from the electronic device 102 of the dubious user, can be marked at step 924 as fraudulent. The response associated with the respective first organization and received from the questionable user's electronic device 102 may be deleted at step 926 from the recommendation database 136 .

[169] FIG. 10 illustrates another embodiment of a flowchart of a computer-implemented method for identifying, among user electronic devices, client electronic devices associated with real customer organizations in accordance with non-limiting embodiments of the present technology.

[170] At 1010, another URL (also referred to herein as a second URL) may be received by the server. For example, the second URL may be received from a different user electronic device (not shown) or the same user electronic device.

[171] At 1015, a second vector representation of the second URL may be generated. At block 1020, the second vector representation is compared with the new vector representation discussed above and other template vector representations of the identity database. If the difference between the second vector representation and any of the template vector representations, including the new vector representation 722 described above, is less than the vector representation difference threshold, then the electronic device from which the second URL was received is considered at block 1030 by the present client device.

[172] If the difference between the second vector representation and all template vector representations, including the new vector representation 722 described above, is greater than or equal to the vector representation difference threshold, then the electronic device from which the second URL was received is considered at step 1040 rogue user's device.

[173] As an alternative to generating and parsing the second vector representation in steps 1015-1020, system 100 may compare, in step 1035, the second URL directly with the new URL scheme generated earlier and other template URL schemes from the identity database 150.

[174] If the second URL is similar to the new URL scheme or any other URL scheme from the identity database 150, then the electronic device from which the second URL was received is considered to be the real client device. If the second URL does not match any of the wildcard schemes, including the new URL scheme, then the electronic device from which the second URL was received is considered to be the device of the rogue user.

[175] In some embodiments, a second static part of the second URL and a second dynamic part of the second URL can be identified. The second URL can be compared with the new URL scheme that was previously stored in the authenticity database 150 .

If the second URL matches the new scheme 742, the second electronic device can be considered to belong to the real customer of the second organization. If the second URL does not match the first scheme, the second electronic device may be considered to belong to the fraudulent user. It should be noted that in some embodiments, the implementation of the second organization may be similar to the first organization.

[176] It should be understood that in at least one embodiment, the second URL matches the first scheme when the second dynamic part of the second URL matches the first dynamic part of the scheme. In at least one embodiment, the second dynamic part of the second URL is compared with the first dynamic part of the scheme and with one or more masks of the first dynamic part of the scheme. The second URL may be considered to match the first scheme when their difference is less than a scheme difference threshold, which may be, for example, ten percent deviation or the like.

[178] If it has been determined that the second electronic device belongs to a fraudulent user, the recommendation database may be accessed to designate the recall associated with the first entity and received from the second user electronic device as fraudulent. The second feedback associated with the first organization and received from the second user electronic device may be removed from the recommendation database.

[179] It should be clearly understood that not all technical effects mentioned in this document must be provided in every embodiment of the present technology. For example, embodiments of the present technology may be implemented without providing the user with some of the technical effects mentioned, while other embodiments may be implemented with providing the user with other technical effects or without providing any technical effects.

[180] Modifications and improvements to the above-described implementations of the present technology may become apparent to those skilled in the art. The preceding description is intended to be exemplary and not limiting. Therefore, the scope of the present technology is intended to be limited only by the scope of the appended claims.

[181] Although the above implementations have been described and shown with reference to specific steps performed in a specific order, it should be understood that these steps can be combined, subdivided, or reordered without deviating from the teachings of the present technology. Accordingly, the order and grouping of steps is not a limitation of the present technology.

Claims (109)

1. A computer-implemented method for identifying client electronic devices from among user electronic devices as client electronic devices associated with real clients of organizations, the method being executed by a server and comprising: receiving by the server from the first user electronic device a first uniform resource locator (URL) from the first user browsing history; generating a first vector representation of a transaction confirmation page URL corresponding to the first URL; comparing the first transaction confirmation page URL vector representation with the transaction confirmation page URL template vector representation of the plurality of transaction confirmation page URL template vector representations, wherein the plurality of transaction confirmation page URL template vector representations have been retrieved from the authentication database; and in response to the difference between the first vector representation of the transaction confirmation pages URL and the templated vector representation of the transaction confirmation pages URLs being below the threshold value of the difference of the vector representations of the transaction confirmation pages URLs, adding the first vector representation of the transaction confirmation pages URLs to the authentication database, and identifying the first user electronic device as the first client electronic device associated with the first real client of the first organization. 2. The method of claim 1, wherein the method further comprises: in response to the fact that the difference between the first vector representation of the transaction confirmation pages URL and the templated vector representation of the transaction confirmation pages URLs is below the threshold of the difference in vector representations of the transaction confirmation pages URLs: designating the first user electronic device as the first client electronic device associated with the first real client of the first organization; and displaying the first feedback received from the first user electronic device on the displays of electronic devices of future customers. 3. The method of claim 1, which further comprises generating an identity database by: receiving, by the server, a plurality of URL template transaction confirmation pages of the actual authentication pages of the respective template organizations; and for each transaction confirmation template page URL from the transaction confirmation page URL templates, generating a new transaction confirmation page URL template vector representation, and storing the new transaction confirmation page URL template vector representation in the authentication database, the new transaction confirmation page URL template vector being one of sets of template vector representations of URLs of transaction confirmation pages. 4. The method of claim 3, wherein at least one of the authentication pages is an online purchase confirmation page indicating that the real customer has completed the purchase. 5. The method of claim. 1, further comprising, in response to the fact that the difference between the first vector representation of the URL of the transaction confirmation pages and the first template vector representation of the URL of the transaction confirmation pages is below the threshold value of the difference of vector representations of the URL of the transaction confirmation pages: associating the first vector representation of the URL of the transaction confirmation pages with a URL template. 6. The method of claim. 1, further comprising, in response to the fact that the difference between the first vector representation of the URL of the transaction confirmation pages and the first template vector representation of the URL of the transaction confirmation pages is below the threshold value of the difference of vector representations of the URL of the transaction confirmation pages: generating a first URL scheme for the first URL; and adding the first URL scheme to the authentication database. 7. The method according to claim 6, in which generating the first URL scheme comprises identifying the first dynamic part of the first URL and the first static part of the first URL. 8. The method according to claim 7, in which: generating the first URL scheme comprises identifying at least one mask of the first dynamic scheme part of the first URL scheme itself and identifying the first static scheme part of the first URL scheme itself. 9. The method of claim 8, further comprising: receiving by the server from the second user electronic device a second URL; comparing the second URL with the first URL scheme; and in response to the second URL matching the first URL scheme: marking the second user electronic device as the second client electronic device associated with the second real client of the second organization, and causing the display of the second feedback about the second organization, received from the second user electronic device, on the displays of electronic devices of future customers. 10. The method according to claim 9, in which: in response to the second URL being different from the first URL scheme: access the database of recommendations; and execution of one of: (i) designating the second recall associated with the second entity and received from the second user electronic device as fraudulent; and (ii) deleting the second response associated with the second entity and received from the second user electronic device from the recommendation database. 11. The method according to p. 1, while the method further comprises: receiving by the server from the second user electronic device the second URL; generating a second vector representation of the URL of the transaction confirmation pages corresponding to the second URL; comparing the second vector representation of the URL of the transaction confirmation pages with the first vector representation of the URL of the transaction confirmation pages; and in response to the fact that the difference between the second vector representation of the transaction confirmation page URLs and the first vector representation of the transaction confirmation pages URLs is below the threshold value of the vector representation difference of the transaction confirmation pages URLs: adding a second vector representation of the transaction confirmation page URLs to the authentication database. 12. The method of claim 1, wherein the difference between the first transaction confirmation page URL vector representation and the transaction confirmation page URL template vector representation is below a transaction confirmation page URL vector representation difference threshold, indicating that the first URL and the template URL have at least partially a common URL scheme. 13. The method of claim 1, wherein generating the first vector representation of the transaction confirmation page URLs comprises applying the Word2Vector algorithm to the first URL. 14. The method of claim 1, further comprising, in response to the fact that the difference between the first transaction confirmation page URL vector representation and the transaction confirmation page URL template vector representation is below a transaction confirmation page URL vector representation difference threshold: accessing a recommendation database, and designating the reviews associated with the first organization and received from the first electronic device as genuine reviews. 15. The method of claim 14, further comprising, in response to the fact that the difference between the first vector representation of the transaction confirmation pages URL and the template vector representation of the transaction confirmation pages URLs is below a threshold value of the difference in vector representations of the transaction confirmation pages URLs, excluding from the recommendations database a subset of testimonials that are not associated with a variety of electronic devices of real customers. 16. The method of claim. 1, further comprising, in response to the fact that the difference between the first vector representation of the URL of the transaction confirmation pages and the template vector representation of the URL of the transaction confirmation pages is higher than or equal to the threshold value of the difference of vector representations of the URL of the transaction confirmation pages: access to a database of recommendations; and execution of one of: (i) designating reviews associated with the first entity and received from the first user electronic device as fraudulent; and (ii) deleting the feedback associated with the first organization and received from the first user electronic device from the recommendations database. 17. The method according to p. 1, further comprising: before receiving the first URL, requesting the first user electronic device to provide the first URL from the first user browsing history. 18. The method of claim 2, further comprising responsive to the fact that the difference between the first transaction confirmation page URL vector representation and the transaction confirmation page URL template vector representation is below a transaction confirmation page URL vector representation difference threshold, and prior to calling the first revocation display : sending a feedback request to the first electronic device of the present client requesting feedback on the first organization, wherein the feedback request is configured to be displayed on the display of the first electronic device of the present client. 19. The method of claim 2, wherein the first organization is at least one of an online store, a reservation service, and a service provider. 20. A computer-implemented system for identifying client electronic devices from among user electronic devices, as client electronic devices associated with real clients of organizations, and the system contains: authenticity database; and a server configured to: receiving by the server from the first user electronic device a first uniform resource locator (URL) from the first user browsing history; generating a first vector representation of a transaction confirmation page URL corresponding to the first URL; comparing the first transaction confirmation page URL vector representation with the transaction confirmation page URL template vector representation of the plurality of transaction confirmation page URL template vector representations, wherein the plurality of transaction confirmation page URL template vector representations have been retrieved from the authentication database; and in response to the difference between the first vector representation of the transaction confirmation pages URL and the templated vector representation of the transaction confirmation pages URL is below the threshold of the difference of the vector representations of the transaction confirmation pages URLs, adding the first vector representation of the transaction confirmation pages URLs to the authentication database, and identifying the first user electronic device as the first client electronic device associated with the first real client of the first organization. 21. The system of claim 20, wherein the server is further configured to: in response to the fact that the difference between the first vector representation of the transaction confirmation pages URL and the templated vector representation of the transaction confirmation pages URLs is below the threshold for the difference in vector representations of the transaction confirmation pages URLs: designating the first user electronic device as the first client electronic device associated with the first real client of the first organization; and causing the display of the first feedback received from the first user electronic device on the displays of electronic devices of future customers. 22. The system of claim 20, wherein the identity database is generated by: receiving, by the server, a plurality of URL template transaction confirmation pages of actual authentication pages of the respective template organizations; and for each transaction confirmation template page URL from the transaction confirmation page URL templates, generating a new transaction confirmation page URL template vector representation, and storing the new transaction confirmation page URL template vector representation in the authentication database, the new transaction confirmation page URL template vector being one of sets of template vector representations of URLs of transaction confirmation pages. 23. The system of claim 22, wherein the server is further configured, in response to the fact that the difference between the first transaction confirmation page URL vector representation and the first transaction confirmation page URL template vector representation is below a transaction confirmation page URL vector representation difference threshold : associating the first vector representation of the URL of the transaction confirmation pages with a URL template. 24. The system of claim 22, wherein the server is further configured, in response to the fact that the difference between the first transaction confirmation page URL vector representation and the first transaction confirmation page URL template vector representation is below a transaction confirmation page URL vector representation difference threshold : generating a first URL scheme for the first URL; and adding the first URL scheme to the authentication database. 25. The system of claim 24, wherein the server is further configured to identify the first dynamic part of the first URL and the first static part of the first URL. 26. The system of claim 25, wherein the server is further configured to: identifying at least one mask of a first dynamic schema portion of the first URL schema itself; and identifying a first static schema portion of the first URL schema itself. 27. The system of claim 26, wherein the server is further configured to: receiving by the server from the second user electronic device a second URL; comparing the second URL with the first URL scheme; and in response to the second URL matching the first URL scheme: designating the second user electronic device as the second client electronic device associated with the second real client of the second organization, and causing the display of the second feedback about the second organization, received from the second user electronic device, on the displays of electronic devices of future customers. 28. The system of claim 27, wherein the server is further configured to: in response to the second URL being different from the first URL scheme: accessing the database of recommendations; and execution of one of: (i) designating the second recall associated with the second entity and received from the second user electronic device as fraudulent; and (ii) deleting the second response associated with the second entity and received from the second user electronic device from the recommendation database. 29. The system of claim 22, wherein the server is further configured to: receiving by the server from the second user electronic device the second URL; generating a second vector representation of the URL of the transaction confirmation pages corresponding to the second URL; comparing the second vector representation of the URL of the transaction confirmation pages with the first vector representation of the URL of the transaction confirmation pages; and in response to the fact that the difference between the second vector representation of the transaction confirmation page URLs and the first vector representation of the transaction confirmation pages URLs is below the threshold value of the vector representation difference of the transaction confirmation pages URLs: adding a second vector representation of the transaction confirmation page URLs to the authentication database. 30. The system of claim 20, wherein the difference between the first transaction confirmation page URL vector representation and the transaction confirmation page URL template vector representation is below a transaction confirmation page URL vector representation difference threshold, indicating that the first URL and the template URL have at least partially a common URL scheme. 31. The system of claim 20, wherein the server is further configured to generate a first vector representation of the transaction confirmation page URLs by applying the Word2Vector algorithm to the first URL. 32. The system of claim 20, wherein, in response to the fact that the difference between the first vector representation of the transaction confirmation pages URL and the template vector representation of the transaction confirmation pages URLs is below a threshold value of the difference of the vector representations of the transaction confirmation pages URLs, the server is further configured to : accessing a database of recommendations and designating reviews associated with the first organization and received from the first electronic device as genuine reviews. 33. The system of claim 32, wherein the server is further configured to, in response to the difference between the first transaction confirmation page URL vector representation and the transaction confirmation page URL vector representation being below a transaction confirmation page URL vector representation difference threshold, exclusion from the recommendations database of a subset of reviews that are not associated with a variety of electronic devices of real customers. 34. The system of claim 20, wherein, in response to the fact that the difference between the first vector representation of the transaction confirmation pages URL and the template vector representation of the transaction confirmation pages URLs is higher than or equal to a threshold value of the difference of vector representations of the transaction confirmation pages URLs, the server is further executed with an opportunity: accessing the database of recommendations; and execution of one of: (i) designating reviews associated with the first entity and received from the first user electronic device as fraudulent; and (ii) deleting the feedback associated with the first organization and received from the first user electronic device from the recommendations database. 35. The system of claim 21, wherein, in response to the fact that the difference between the first vector representation of the transaction confirmation pages URL and the template vector representation of the transaction confirmation pages URLs is below a threshold value of the difference of the vector representations of the transaction confirmation pages URLs, and before calling the display of the first revocation, the server is additionally configured to: sending a feedback request to the first electronic device of the present client requesting feedback on the first organization, the feedback request being configured to be displayed on the display of the first electronic device of the present client.

Images