System and method for generating individual content for users of a service

Abstract

The present invention relates to systems and methods for generating individual content for users of a service. In one aspect, an exemplary method includes: collecting data related to the behavior of a user of a computing device; training a user behavior model based on the collected data, wherein the trained data identifies the user with a predetermined reliability; and providing individual content to the user of the service based on a predetermined service environment according to the trained model received from a model transmitter.

Description

Technical Field

[0001] The present invention relates to the field of data analysis and machine learning, and more particularly, to a system and method for generating individual content for users of a service. Background Art

[0002] The rapid development of computer technology in the past decade and the widespread use of various computing devices (personal computers, laptops, tablet computers, smartphones, etc.) have become a strong stimulus for the use of these devices in various activities and a huge number of tasks. Some examples of tasks include Internet surfing, banking activities (e.g., remittances), transferring electronic files, etc. However, along with the rapid growth in the number of computing devices and the amount of software running on these devices, malicious activities are also growing rapidly. For example, the number of malicious programs, the number of methods for unauthorized access to data being processed by the devices, and the number of fraudulent methods for using the data continue to grow. Therefore, the issue of information security has become particularly important.

[0003] Widespread attention has begun to be paid not only to ensuring the information security of users' personal data, but also to protecting users' personal data. The protection of personal data can be directed against hypothetical (sometimes even unlikely) threats, such as successful targeted attacks on servers storing users' personal data, etc. In the European Union, solutions based on the concept of voluntary consent have been introduced for this purpose (e.g., the General Data Protection Regulation (GDPR)). In addition, solutions have also been introduced to ensure the maximum storage of users' personal data on their computing devices and the minimum possible analysis and use of such data at external services.

[0004] At present, the use of trained models (usually, and machine learning) has been widely used in various fields of human activities. Efforts are also made to use trained models to save users' personal data and use users' personal data for various tasks. For example, models trained based on users' personal data can allow: predicting users' preferences, identifying users, and so on in certain fields of business or other information technology. One of the main advantages of trained models is that personal data used to train the model cannot be extracted from the trained model. Therefore, the trained model can use personal user data without disclosing personal user data (or transferring data from one computing device to another computing device, which will have a negative impact on the information security of user data). The above method uses technology to handle the following tasks: identifying users through their behavior (actions). However, the user action itself will never be hidden; the data (personal user data) is sent to other computing devices (including remote computing devices) for analysis (identification). Therefore, user actions may be vulnerable to unauthorized access.

[0005] Therefore, there is a need for an optimal method to provide services to users based on personal user data without sending the data over the network. Summary of the invention

[0006] Various aspects of the present invention are directed to generating individual content for users of a service while providing information security by transmitting to the service a user behavior model that has been trained using personal user data without transmitting the actual personal user data. Another aspect of the present invention is directed to providing security for personal user data by identifying users based on a user behavior model that has been trained using personal user data rather than identifying users based on the actual personal user data.

[0007] In an exemplary aspect, a method for generating separate content for users of a service is provided, the method comprising: collecting data related to behavior of users of a computing device; training a user behavior model based on the collected data, wherein the trained data identifies the user with a predetermined reliability; and generating separate content for users of the service based on a predetermined service environment according to the trained behavior model received from a model transmitter.

[0008] In one aspect, the data related to the user's behavior describes at least one of the following: one or more applications that the user is using, the time the user uses each of the one or more applications, the actions the user performs in each of the one or more applications, the search queries issued by the user, one or more websites visited by the user, the actions performed by the user on corresponding websites of the one or more websites visited, and the forms filled out by the user.

[0009] In one aspect, the user behavior model comprises a model based on a neural network with weights, wherein the model is used to predict the user's next behavior.

[0010] In one aspect, the method further comprises sending the trained user behavior model to the content generator, wherein the sending is performed at least at a predetermined frequency.

[0011] In one aspect, the method further includes sending the trained user behavior model to the content generator, wherein the sending is performed based at least on a request of the content generator.

[0012] In one aspect, the behavior model database contains user behavior models collected during user registration for the service.

[0013] In one aspect, the method further includes, before generating individual content to the user of the service, authorizing the user to use the service based on a comparison result of the user behavior model provided by the model transmitter and one or more models obtained from the behavior model database.

[0014] In one aspect, the one or more behavioral models are compared to each other using at least one of: a scatter plot matrix, individual scatter plots, and a statistical significance test.

[0015] In one aspect, the predetermined service environment includes an environment for accessing at least one of: a list of goods and services to which the user is to be authorized to access, and data to which the user is authorized to access based on analysis of the behavior model received from the model transmitter.

[0016] According to one aspect of the present invention, a system for generating separate content for users of a service is provided, the system comprising a hardware processor configured to: collect data related to the behavior of a user of a computing device; train a user behavior model based on the collected data, wherein the trained data identifies the user with a predetermined reliability; and generate separate content for users of the service based on a predetermined service environment according to the trained behavior model received from a model transmitter.

[0017] In one exemplary aspect, a non-transitory computer-readable medium is provided, on which a set of instructions for generating separate content for users of a service is stored, wherein the set of instructions includes instructions for: collecting data related to behavior of users of a computing device; training a user behavior model based on the collected data, wherein the trained data identifies the user with a predetermined reliability; and generating separate content for users of the service based on a predetermined service environment according to the trained model received from a model transmitter.

[0018] The method and system of the present invention are designed to generate individual content to users of the service while providing security of information. Therefore, a user behavior model is transmitted instead of the user's personal data. In addition, the user is identified based on the user behavior model that has been trained using personal user data instead of the actual personal user data. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate one or more exemplary aspects of the present invention and, together with the detailed description, serve to explain the principles and implementations of these exemplary aspects.

[0020] Figure 1 A system for generating individual content for users of a service according to aspects of the present invention is shown.

[0021] Figure 2 An exemplary method for generating individual content for users of a service according to aspects of the invention is shown.

[0022] Figure 3 An example of a general purpose computer system is shown upon which aspects of the invention may be implemented. DETAILED DESCRIPTION

[0023] Exemplary aspects are described herein in the context of systems, methods, and computer programs for generating separate content for users of services. It will be appreciated by those of ordinary skill in the art that the following description is merely illustrative and is not intended to be limiting in any way. Other aspects will readily suggest themselves to those skilled in the art who understand the advantages of the present invention. Reference will now be made in detail to the implementation of the exemplary aspects as shown in the accompanying drawings. Throughout the accompanying drawings and the following description, the same reference numerals will be used as much as possible to refer to the same or similar items.

[0024] In one aspect, the present invention describes a system for generating separate content for users of a service, the system being implemented on a computing system that includes real-world devices, systems, components, and groups of components that are implemented using hardware such as integrated microcircuits (Application Specific Integrated Circuits, ASICs) or field programmable gate arrays (FPGAs), or in the form of a combination of software and hardware (e.g., a microprocessor system and a program instruction set), and also on a neurosynaptic chip. The functions of such devices of the system may be implemented solely by hardware, and may also be implemented in a combined form, with some functions of the system devices being implemented by software and some functions being implemented by hardware. In some aspects, some or all of the components, systems, etc. may be implemented on a general-purpose computer such as a computer system. Figure 3In addition, the system components may be implemented in a single computing device or may be distributed among multiple interconnected computing devices.

[0025] Figure 1 A system 100 for generating individual content for users of a service according to aspects of the present invention is shown. In one aspect, the system 100 includes a data collector 110, a model trainer 120, a user behavior model 121, a model transmitter 130, a behavior model database 131, a content generator 140, service content 141, individual content 142, and an analyzer 150.

[0026] In one aspect, the main purpose of the system 100 is to generate individual user content from predetermined service content of a remote service.The individual user content is generated based on an analysis of the user's behavior when using his / her computing device.

[0027] In one aspect, the remote service includes at least one of the following:

[0028] Online store;

[0029] Advertising platforms (including, for example, informational websites) that present advertisements to users; and

[0030] · Resources that provide information services.

[0031] In one aspect, when the remote service includes an online store, the service content includes at least a collection of commodities available in the online store. In one aspect, the service content also includes information about the commodities in the collection.

[0032] In one aspect, when the remote service includes an advertising platform, the service content includes an advertising block including media data for advertising goods and services, such as a banner advertisement.

[0033] In one aspect, when the remote service includes resources for providing information services, the service content includes: information, news, and any number of other data blocks sent to users. The information, news, etc. sent to users can be similar to the content sent through the RSS channel.

[0034] In one aspect, the data collector 110 collects data related to the user's behavior (ie, log data) during use of the user's computing device and sends the collected data to the model trainer 120 .

[0035] In one aspect, the user's computing device includes at least:

[0036] Personal computer;

[0037] Laptop computer;

[0038] Tablet computers; and

[0039] Mobile or cell phone.

[0040] In one aspect, the computing device includes a collection of computing devices on which software is run, wherein the computing devices of the collection are interconnected through a single user account. For example, the computing device may be a user's personal computer, the user's iPhone and iPad tablet computer, and the software running on the user's personal computer, the user's iPhone and iPad tablet computer connects to a remote service or connects to a single cloud service (e.g., a cloud service based on Handoff technology) through a single account (e.g., YouTube under a single account), that is, the computing device is a computing device on which a single sign-on (SSO) technology is used.

[0041] In one aspect, the collected data related to the user's behavior includes at least one of the following:

[0042] The names of the applications that the user is using, the time that the user uses each application, and the actions that the user performs in each application;

[0043] Search queries issued by users;

[0044] The addresses of the (online) websites that the user visits, the actions that the user performs on each website (e.g. which links the user uses to jump from the website's pages, extract and copy text, etc.);

[0045] Data from forms filled out by users;

[0046] Information related to cursor movements of mice, touch pads, and other data input devices;

[0047] Jump from the login page to other pages;

[0048] The identifier of the data entry tool or device (hardware ID); and

[0049] An identifier for the client application.

[0050] For example, the system may collect requests made to a voice assistant (e.g., Alisa, Alexa, Siri, etc.) as data related to the user's behavior.

[0051] In another example, as data related to the user's behavior, the system collects Google search queries and addresses of websites visited by the user after each search query.

[0052] In one aspect, the data collected by the data collector 110 is analyzed such that at least one of the following is performed:

[0053] When the content generator 140 generates individual content for users of the service, data with a low priority is not recorded;

[0054] Do not record duplicate data; and

[0055] Correct incorrect (erroneous) data.

[0056] In one aspect, the model trainer 120 trains the user behavior model 121 based on the collected data and sends the trained user behavior model 121 to the model transmitter 130 .

[0057] In one aspect, the user behavior model 121 comprises a model based on a neural network with weights, which is used to predict the next behavior of the user.

[0058] For example, a trained user behavior model 121 may be generated based on data related to a user's search queries, so that when the model 121 is used, it is possible to determine which known search queries (including modified known search queries, i.e., queries composed based on a combination of multiple known queries) will be executed by the user in the future, and select the known search query with the greatest probability.

[0059] In one aspect, the user behavior model 121 is trained so that the user behavior model 121 identifies the user with a given degree of reliability.

[0060] In another aspect, training the user behavior model 121 includes:

[0061] Parameterize the collected data about the user's behavior in vector form (e.g., by lexical analysis, converting speech to text, converting text to text vectors);

[0062] Obtaining a cleaned vector by removing low priority information, eliminating noise, etc. from the obtained vector; and

[0063] Send the cleaned vector to the input of the training algorithm.

[0064] As an illustrative example, for two users using the same computing device but performing different actions (using different applications, making different search queries, filling out different forms, etc.), different trained user behavior models 121 will be generated, and these two trained models will undoubtedly be different from each other, and the degree of their difference will characterize the reliability level of user identification of a given trained user behavior model 121. Different metrics (see below) can be used to compare these trained user behavior models 121. In addition, for the same user using a computing device at different times, similar trained user behavior models 121 are generated so that only one user will be identified with a given reliability.

[0065] In one aspect, the user behavior model 121 is first trained to characterize the behavior of a large group of users. For different user groups, multiple different user behavior models 121 can be trained to characterize different user groups.

[0066] Based on the data collected by the data collector 110, one of these user behavior models 121 is sent to the model trainer 120. This approach makes it possible to reduce the demand on the resources of the user's computing device, since no full training of the user behavior model is performed, but only further training is performed.

[0067] In one aspect, a pre-trained user behavior model 121 is generated on a remote service, wherein individual content 142 is generated for each specific user based on the trained user behavior model 121 of each specific user.

[0068] As an illustrative example, assume that the service is used to provide advertising material in the form of banner ads on news websites and online magazines. The service generates different trained user behavior models 121 that characterize different user groups (sports enthusiasts, housewives, students, etc.). Depending on which partner websites of the service a particular user has visited, the system for generating separate content to the user of the service will send the corresponding pre-trained user behavior model. For example, for users visiting an online store for sporting goods, a trained user behavior model 121 that has previously been trained on the behavior of sports enthusiasts will be sent. This pre-trained user behavior model 121 is adjusted (trained) to take into account the search queries made by users on websites with sports themes and the use of sports applications (trackers, heart rate monitors, etc.), and ignores other data collected by the data collector 110 (i.e., other data contributes little to the working results of the trained user behavior model).

[0069] In one aspect, the model transmitter 130 transmits the trained model 121 to the content generator 140 .

[0070] In one aspect, the model transmitter 130 determines whether a condition for transmitting the trained model 121 is met before transmitting the trained model 121 to the content generator 140. For example, user authorization or the like may be required.

[0071] In one aspect, the behavior model database 131 contains user behavior models collected during user registration for the service.

[0072] As an illustrative example, an online book selling service may contain a database of registered users, where the registration information sent is not the user's login name and password, but rather his or her trained behavioral model 121, where the model 121 itself is constructed using software provided by the service, which software includes at least a data collector 110, a model trainer 120, and a model transmitter 130. Thus, after a user first visits the service and after a trained model 121 has been generated and sent to the service, the trained model resides in the behavioral model database 131. During repeated visits of the user to the service, further trained user behavioral models 121 will be sent (specifically, by the model transmitter 130), which may differ from previously sent user behavioral models 121, but not beyond the extent that allows the user to be unambiguously identified on the service.

[0073] In one aspect, the model transmitter 130 includes at least:

[0074] A separate application that intercepts requests to a service, wherein the interception is performed to provide the trained user behavior model 121 as a user identifier to the service to which the request is directed;

[0075] A plug-in installed in the browser of the user's computing device that automatically provides the trained user behavior model 121 to the service when the user visits the website of the corresponding service; and

[0076] · A script (eg, in the form of JS code) that contains elements of the trained user behavior model 121 and is embedded in the service website.

[0077] As an illustrative example, the model transmitter 130 can be a plug-in of the Google Chrome browser, which is installed on the user's computing system by the request of the online book sales store when the website is visited for the first time. Each time the website is visited, this plug-in will send the generated user behavior model 121 as a user identifier.

[0078] In one aspect, the model transmitter 130 transmits the trained user behavior model 121 to the content generator 140 at least as follows:

[0079] at a predetermined frequency; and

[0080] According to the requirements of the content generator 140.

[0081] For example, if the user is using some software provided by the service, the software (as described above) may send an updated (i.e., further trained) user behavior model 121 to the service once a day. As a result, when the user accesses the service, at least:

[0082] be authorized to use the Service at all times; and / or

[0083] · Receive relevant and new information, ie individual content 142 of the service.

[0084] In one aspect, the model transmitter 130 authorizes the user to use the service based on the comparison of the user behavior model 121 sent to the content generator 140 with the model obtained from the behavior model database 131. For example, if two users use the same computing device, the trained user behavior model of user No. 1 should be different from the trained user behavior model of user No. 2 so that the two users can be distinguished with a given degree of reliability. If the same user uses different computing devices at different times, the trained user behavior model generated based on data from computing device No. 1 should be similar to the trained user behavior model generated based on data from computing device No. 2.

[0085] In one aspect, the comparison of two behavioral models is performed by using at least:

[0086] Scatter plot matrix;

[0087] Individual scatter plots; and

[0088] Statistical significance testing.

[0089] As an illustrative example, the parameters of a neural network are used as a scatter plot.

[0090] In one aspect, the two trained behavioral models may be compared to one another based on an analysis of the results when the two trained behavioral models are used on a previously prepared test sample of data.

[0091] In one aspect, test samples can be prepared by calculating what actions a user can (or should) perform on a given service. For example, test samples can include search queries for items in an online store that the user has visited, or search queries for items and services through advertisements displayed to the user, etc.

[0092] For example, assume that: for user No. 1, the behavior model is trained using search queries of user No. 1 and a certain group of other users, while for user No. 2, the behavior model is trained using search queries of user No. 2 and a certain group of other users (not necessarily the same as the training of the behavior model of user No. 1). In order to verify how similar the trained behavior models of user No. 1 and user No. 2 are to each other, a test sample of search queries prepared in advance is put into the model. At the output of the model, the probability of a specific query related to the corresponding user (i.e., user No. 1 for the trained behavior model of user No. 1 and user No. 2 for the trained behavior model of user No. 2) is calculated. The comparison of the results obtained is used to determine whether these user behavior models are similar or different.

[0093] Considering the behavior model 121 provided by the model transmitter 130 , the content generator 140 generates individual content 142 to users of the service based on a predetermined service environment 141 .

[0094] In one aspect, the designated service environment 141 includes at least:

[0095] a list of goods and services to which the user may be authorized to access, where access to goods includes providing the user with information about the goods or services and the ability to order the goods or services, including purchasing (paying for) the ordered goods or services; and

[0096] Data that the user is authorized to access based on analysis of the user behavior model 121 provided by the model transmitter 130 .

[0097] As an illustrative example, assume that a user accesses an online appliance sales service, after which the model transmitter 130 sends the user behavior model 121 that has been trained using the user's search query to the service. The trained user behavior model 121 provided to the service will be used to identify the user and authorize him / her to use a given service. Based on the provided trained user behavior model 121 and based on previous search queries for goods provided by the online merchant service, it is determined which goods (lists) have a higher priority for the user (mainly goods for which the search query was made, related goods, etc.). Then, a catalog and price list are generated for the user based on the determined list of goods.

[0098] Therefore, the user is presented with certain information previously prepared based on his / her personal data (e.g., search queries), and the search queries themselves are not presented to the service, but only the user behavior model 121, which was trained by those search queries, thereby improving the information security of the user's personal data.

[0099] In yet another example, assume that the service is an advertising agency for providing contextual advertising to users through various platforms (various subject websites: news websites, forums, online stores, etc.). In order for the service to operate more accurately, that is, in order to select more relevant advertisements for each specific user, user data needs to be analyzed, wherein the analysis includes data about the user's behavior on the computing device and in the network, which application the user is using, which search queries he or she is conducting, which websites are being visited, and so on. The service needs to collect these types of data. On the one hand, it is illegal to collect a large amount of data without the user's consent; on the other hand, even if the data is collected with the user's consent, it may lead to unauthorized access to the data collected in this way, thereby bringing reputational risks and financial risks to the user. Therefore, the method of the present invention provides benefits to the user. The method of the present invention uses a user behavior model 121 that has been trained using the user's computing device. However, the model does not contain user data (user data includes personal data). In addition, the model can be used to identify the user, predict his or her behavior, interests, and so on.

[0100] In some scenarios, the use of the trained user behavior model 121 may result in high demands on the computing resources of the user's computing device. Even if the training of the user behavior model on the user's computing device is performed in real-time mode, this may affect the performance of the computing device itself and the operation of the application that the user is using in a way that is obvious to the user. For example, when the user behavior model 121 is trained on the user's computing device, the user's computing device may work more slowly, RAM consumption may be higher, and the autonomous operation time of the mobile device may be shorter. In order to prevent excessive demands on the computing resources of the user's computing device, in one aspect, the method of the present invention uses a previously trained user behavior model 121 and updates the model for use by further training the previously trained model.

[0101] The above method can solve the above problems and provide the following significant advantages:

[0102] Providing the user with a user behavior model 121, which is designed to generate individual content 142 for users of a specific service, which greatly improves the accuracy of generating the individual content 142;

[0103] The training (further training) of the user behavior model 121 does not result in the user data being sent outside the computing device (e.g., to a dedicated service in a computer network), which improves the information security of the user data; and

[0104] • Further training of the user behavior model 121, rather than full training of the user behavior model 121, which reduces the demand on computing resources of the user's computing device.

[0105] As an illustrative example, for the above-mentioned advertising agency, different pre-trained user behavior models 121 are used for different platforms (websites). For example, for a news website, the model is pre-trained to take into account the user's search queries for news that they are interested in, determine which topics are of greatest interest to the user, and so on; on a sports website, the model is pre-trained to take into account the sports applications that are being used on the user's computing device, and so on. This type of user behavior model 121 is issued to the user as an identifier for accessing the corresponding website (e.g., in order to receive various news, access forums, personal accounts, etc.), and then provided from the user to the corresponding website (when going to the website, for example, using a special plug-in installed in the browser), and from these websites to the advertising agency (service). Based on the user behavior model 121 that has been fully trained, the advertising agency determines which specific advertisement it needs to display to a specific user.

[0106] Additionally, different advertising agencies (different services) may exchange trained and pre-trained user behavior models 121 with each other to improve the accuracy of generating individual content to users of a particular website.

[0107] In one aspect, the analyzer 150 collects data of the user's actions on the service, analyzes the statistics of the user's actions on the service based on the collected data, and makes a decision on whether the user behavior model 121 needs to be retrained. In one aspect, the results analyzed by the analyzer 150 are used to determine which specific actions the user has performed (e.g., which products he or she is interested in, which products he or she has purchased, etc.) in order to decide whether the user behavior model 121 needs to be retrained or further trained, so as to generate more relevant content 142 of the service for the user when the user visits the service in the future. In one aspect, the time spent on the service, the number of purchases made or rejected, the number of messages sent to customer support, etc. can be used to evaluate the relevance of the analysis results.

[0108] In one aspect, the user behavior model 121 may not be fully generated (trained), but a partially trained ready-made behavior model of a "generic" user may be further trained. For example, when the trained user behavior model 121 is a neural network, further training involves correcting the weights of each layer of the neural network based on data obtained from the content generator 140.

[0109] For example, when the service is an online store, during the user's first visit to the website of the online store, the user is provided with a pre-trained user behavior model 121 as an identifier based on the actions performed by the user in the online store to access his or her personal account (e.g., which catalogs he or she viewed, which goods he or she is looking for, etc.). The model takes into account certain search queries that the user has already made and is adjusted to more accurately predict the content that a given user needs in the online store. The preliminary training of the user behavior model 121 takes into account the behavior of a large group of people with similar queries, so that on the one hand, the user behavior model 121 is applicable to a large number of users, but on the other hand, when generating individual content for users of the online store, using the user behavior model 121 may lead to less correct results. Therefore, it is necessary to further train the provided user behavior model 121 on the user's computing device. In part, this reduces the demand on the computing resources of the user's computing device, and speeds up the training of the user behavior model because it is not necessary to perform an entire training cycle with a large training sample, etc. Furthermore, if the same initially trained user behavior model 121 is provided to two different users, the two models will be different after further training on their computing devices based on the users' actions.

[0110] Figure 2 An exemplary method 200 for generating individual content for users of a service in accordance with aspects of the present invention is shown.

[0111] A structural diagram of a method for generating separate content for users of a service includes: step 210, in which data related to the user's behavior is collected; step 220, in which a user behavior model is trained; step 230, in which data related to the user's actions is collected; step 240, in which the user is authorized to use the service; step 250, in which separate content for the user is generated; and step 260, in which statistical data related to the user's actions on the service are analyzed.

[0112] In step 210 , method 200 collects data related to the behavior of a user of a computing device via data collector 110 .

[0113] In step 220 , the method 200 trains the user behavior model 121 based on the collected data by the model trainer 120 , so that the trained model identifies the user with a predetermined reliability.

[0114] In step 230 , method 200 collects data, via analyzer 150 , regarding the user's actions when accessing the service.

[0115] In step 240, the method 200 authorizes the user to use the service based on the result of comparing the trained user behavior model 121 with any number of models obtained from the behavior model database 131 by the model transmitter 130. The trained user behavior model 121 is received from the model trainer 120. The model transmitter 130 provides the model to the content generator 140. Therefore, in one aspect, the method further includes: before generating separate content for the user of the service, authorizing the user to use the service based on the result of comparing the user behavior model provided by the model transmitter with one or more models obtained from the behavior model database.

[0116] In step 250 , the method 200 generates individual content 142 for users of the service through the content generator 140 based on the predetermined service environment 141 according to the behavior model 121 received from the model transmitter 130 .

[0117] In step 260 , the method 200 analyzes, by the analyzer 150 , statistical data related to the user's actions on the service based on the collected data related to the user's actions.

[0118] In step 265, the method 200 determines, through the analyzer 150, whether it is necessary to retrain the user behavior model 121. If it is necessary to retrain the model, the method proceeds to step 220. Otherwise, the method proceeds to step 210.

[0119] In one aspect, the data related to the user's behavior describes at least one of the following: one or more applications that the user is using, the time the user uses each of the one or more applications, the actions the user performs in each of the one or more applications, the search queries issued by the user, one or more websites visited by the user, the actions performed by the user on each of the one or more websites visited, and the forms filled out by the user.

[0120] In one aspect, the user behavior model comprises a model based on a neural network with weights, wherein the model is used to predict the user's next behavior.

[0121] In one aspect, the method further comprises sending the trained user behavior model to the content generator, wherein the sending is performed at least at a predetermined frequency.

[0122] In one aspect, the method further includes sending the trained user behavior model to the content generator, wherein the sending is performed based at least on a request of the content generator.

[0123] In one aspect, the behavior model database contains user behavior models collected during user registration for the service.

[0124] In one aspect, the one or more behavioral models are compared to each other using at least one of: a scatter plot matrix, individual scatter plots, and a statistical significance test.

[0125] In one aspect, the predetermined service environment includes an environment for accessing at least one of: a list of goods and services to which the user is to be authorized to access, and data to which the user is authorized to access based on analysis of the behavior model received from the model transmitter.

[0126] Figure 3 is a block diagram illustrating a computer system 20 according to an exemplary aspect on which aspects of the systems and methods for generating individual content for users of a service may be implemented. The computer system 20 may be in the form of multiple computing devices or in the form of a single computing device, such as a desktop computer, a laptop computer, a handheld computer, a mobile computing device, a smartphone, a tablet computer, a server, a mainframe, an embedded device, and other forms of computing devices.

[0127] As shown, computer system 20 includes a central processing unit (CPU) 21, a system memory 22, and a system bus 23 connecting various system components, including a memory associated with the central processing unit 21. The system bus 23 may include a bus memory or a bus memory controller, a peripheral bus, and a local bus capable of interacting with any other bus architecture. Examples of buses may include PCI, ISA, PCI-Express, HyperTransport, and the like. TM (HyperTransport TM ), unlimited bandwidth TM (InfiniBand TM ), Serial ATA, I 2C, and other suitable interconnections. The central processing unit 21 (also referred to as a processor) may include a single group or multiple groups of processors having a single core or multiple cores. The processor 21 may execute one or more computer executable codes that implement the technology of the present invention. The system memory 22 may be any memory for storing data used herein and / or computer programs executable by the processor 21. The system memory 22 may include volatile memory (such as random access memory (Random Access Memory, RAM) 25) and non-volatile memory (such as read-only memory (Read-Only Memory, ROM) 24, flash memory, etc.) or any combination thereof. The basic input / output system (Basic Input / Output System, BIOS) 26 may store basic programs for transmitting information between elements of the computer system 20, such as those basic programs when the operating system is loaded using ROM 24.

[0128] The computer system 20 may include one or more storage devices, such as one or more removable storage devices 27, one or more non-removable storage devices 28, or a combination thereof. The one or more removable storage devices 27 and the one or more non-removable storage devices 28 are connected to the system bus 23 via a memory interface 32. In one aspect, the storage devices and corresponding computer-readable storage media are power-independent modules for storing computer instructions, data structures, program modules, and other data for the computer system 20. The system memory 22, the removable storage devices 27, and the non-removable storage devices 28 can use a variety of computer-readable storage media. Examples of computer-readable storage media include: machine memory, such as cache, SRAM, DRAM, zero-capacitor RAM, two-transistor RAM, eDRAM, EDO RAM, DDR RAM, EEPROM, NRAM, RRAM, SONOS, PRAM; flash memory or other storage technology, such as in a Solid State Drive (SSD) or a flash drive; magnetic cassettes, tapes, and disk storage, such as in a hard drive or floppy disk; optical storage, such as in a compact disc (CD-ROM) or a digital versatile disc (DVD); and any other medium that can be used to store the desired data and can be accessed by the computer system 20.

[0129] The system memory 22, removable storage device 27 and non-removable storage device 28 of computer system 20 can be used to store operating system 35, additional application programs 37, other program modules 38 and program data 39. Computer system 20 may include a peripheral interface 46 for transmitting data from input device 40, such as a keyboard, mouse, stylus, game controller, voice input device, contact input device, or other peripheral devices, such as a printer or scanner via one or more I / O ports, such as a serial port, parallel port, Universal Serial Bus (USB), or other peripheral interfaces. Display device 47 (such as one or more monitors, projectors or integrated displays) can also be connected to system bus 23 via output interface 48 (such as a video adapter). In addition to display device 47, computer system 20 may also be equipped with other peripheral output devices (not shown), such as speakers and other audio-visual devices.

[0130] The computer system 20 can work in a network environment using a network connection with one or more remote computers 49. The one or more remote computers 49 can be local computer workstations or servers, which include most or all of the elements described above when describing the properties of the computer system 20. Other devices may also be present in the computer network, such as but not limited to routers, websites, peer devices or other network nodes. The computer system 20 may include one or more network interfaces 51 or network adapters for communicating with the remote computers 49 via one or more networks, such as a local area computer network (LAN) 50, a wide area computer network (WAN), an intranet, and the Internet. Examples of the network interface 51 may include an Ethernet interface, a frame relay interface, a SONET interface, and a wireless interface.

[0131] Various aspects of the present invention may be systems, methods and / or computer program products. The computer program products may include one or more computer-readable storage media having computer-readable program instructions for causing a processor to perform various aspects of the present invention.

[0132] Computer readable storage medium can be a tangible device, and this tangible device can keep and store the program code of the form of instruction or data structure, and this program code can be accessed by the processor of computing device (such as computing system 20).Computer readable storage medium can be electronic storage device, magnetic storage device, optical storage device, electromagnetic storage device, semiconductor storage device or its any suitable combination.As an example, this type of computer readable storage medium can include random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), portable optical disc read-only memory (CD-ROM), digital versatile disc (DVD), flash memory, hard disk, portable computer disk, memory stick, floppy disk or even mechanical coding device, such as punch card or convex structure in the groove of recording instruction thereon.As used in this article, computer readable storage medium should not be regarded as temporary signal itself, temporary signal such as radio wave or other free propagation electromagnetic wave, electromagnetic wave propagated by waveguide or transmission medium or electric signal transmitted by wire.

[0133] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a corresponding computing device, or downloaded to an external computer or external storage device via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network). The network can include copper transmission cables, optical transmission fibers, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network interface in each computing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the corresponding computing device.

[0134] The computer-readable program instructions for performing the operation of the present invention can be assembly instructions, instruction set architecture (Instruction-Set-Architecture, ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data or source code or object code written in any combination of one or more programming languages ​​(including object-oriented programming languages ​​and traditional programmed programming languages).Computer-readable program instructions (as independent software packages) can be performed completely on the user's computer, partly on the user's computer, partly on the user's computer and partly on a remote computer or completely on a remote computer or server.In the latter scenario, the remote computer can be connected to the user's computer through any type of network (including LAN or WAN), or can be connected to an external computer (for example, through the Internet).In some aspects, electronic circuits (including for example programmable logic circuits, field programmable gate arrays (FPGAs) or programmable logic arrays (Programmable Logic Arrays, PLAs)) can be performed by utilizing the state information of computer-readable program instructions to personalize the electronic circuit, thereby performing aspects of the present invention.

[0135] In various aspects, the systems and methods described in the present invention may be processed in terms of modules. The term "module" as used herein refers to, for example, a real-world device, component, or arrangement of components implemented using hardware (e.g., via an Application Specific Integrated Circuit (ASIC) or FPGA), or to a combination of hardware and software, such as a microprocessor system and an instruction set that implements the module's functionality (which, when executed, converts the microprocessor system into a dedicated device). A module may also be implemented as a combination of two modules, where certain functions are facilitated by hardware alone and other functions are facilitated by a combination of hardware and software. In some implementations, at least a portion of a module (and in some cases, all of a module) may be implemented in a computer system (e.g., as described above in Figure 3 Therefore, each module can be implemented in various suitable configurations and should not be limited to any specific implementation illustrated in this document.

[0136] For the sake of clarity, not all routine features of various aspects are disclosed herein. It should be appreciated that in the development of any practical implementation of the present invention, many specific implementation decisions must be made in order to achieve the developer's specific goals, and these specific goals will vary for different implementations and different developers. It should be understood that such development efforts can be complex and time-consuming, but are still a routine task of engineering for those of ordinary skill in the art who understand the advantages of the present invention.

[0137] In addition, it should be understood that the words or terms used herein are for the purpose of description rather than limitation, so the terms or terms of this specification should be interpreted by those skilled in the art according to the teachings and guidance provided herein in combination with the knowledge of (one or more) relevant technical personnel. In addition, it is not intended to attribute any term in this specification or claims to an uncommon or special meaning unless explicitly stated as such.

[0138] The various aspects disclosed herein include the present and future known equivalents of the known modules cited herein in an illustrative manner. In addition, although various aspects and applications have been shown and described, it will be apparent to those skilled in the art who understand the advantages of the present invention that more modifications than those mentioned above are feasible without departing from the inventive concept disclosed herein.

Claims

1. A method for generating individual content for a user of a service, the method comprising: collect data about the behavior of users of computing devices; training a user behavior model based on the collected data, wherein the trained data identifies the user with a predetermined reliability based on the collected data, wherein the data related to the user's behavior includes at least one of the following: one or more applications being used by the user, the time the user uses each of the one or more applications, and an action performed by the user in each of the one or more applications; Authorizing the user to use the service based on a comparison result of the user behavior model provided by the model transmitter and one or more models obtained from a behavior model database; and According to the trained model received from the model transmitter, separate content is generated for the user of the service based on a predetermined service environment, wherein the predetermined service environment includes an environment for accessing at least one of: a list of goods and services to which the user is authorized to access, and data that the user is authorized to access.

2. The method according to claim 1, wherein: The data related to the user's behavior also describes at least one of the following: search queries issued by the user, one or more websites visited by the user, actions performed by the user on each of the one or more websites visited, and forms filled out by the user.

3. The method according to claim 1, wherein: The user behavior model includes a model based on a neural network with weights, wherein the user behavior model is used to predict the next behavior of the user.

4. The method according to claim 1, further comprising: The trained user behavior model is sent to a content generator, wherein the sending is performed at least at a predetermined frequency.

5. The method according to claim 1, further comprising: The trained user behavior model is sent to a content generator, wherein the sending is performed based at least on a requirement of the content generator.

6. The method according to claim 1, wherein: The behavior model database contains user behavior models collected during the user's registration with the service.

7. The method according to claim 1, wherein: One or more behavioral models are compared to each other using at least one of: a scatter plot matrix, individual scatter plots, and a statistical significance test.

8. The method according to claim 1, wherein: The predetermined service environment is determined based on an analysis of a behavior model received from the model transmitter.

9. A system for generating individual content for users of a service, comprising: At least one processor of a computing device, the at least one processor configured to: collect data about the behavior of users of computing devices; training a user behavior model based on the collected data, wherein the trained data identifies the user with a predetermined reliability based on the collected data, wherein the data related to the user's behavior includes at least one of the following: one or more applications being used by the user, the time the user uses each of the one or more applications, and an action performed by the user in each of the one or more applications; Authorizing the user to use the service based on a comparison result of the user behavior model provided by the model transmitter and one or more models obtained from a behavior model database; and According to the trained model received from the model transmitter, separate content is generated for the user of the service based on a predetermined service environment, wherein the predetermined service environment includes an environment for accessing at least one of: a list of goods and services to which the user is authorized to access, and data that the user is authorized to access.

10. The system according to claim 9, wherein: The data related to the user's behavior also describes at least one of the following: search queries issued by the user, one or more websites visited by the user, actions performed by the user on each of the one or more websites visited, and forms filled out by the user.

11. The system according to claim 9, wherein: The user behavior model includes a model based on a neural network with weights, wherein the user behavior model is used to predict the next behavior of the user.

12. The system according to claim 9, wherein: The processor is further configured to: send the trained user behavior model to a content generator, wherein the sending is performed at least at a predetermined frequency.

13. The system according to claim 9, wherein: The processor is further configured to send the trained user behavior model to a content generator, wherein the sending is performed based at least on a requirement of the content generator.

14. The system according to claim 9, wherein: The behavior model database contains user behavior models collected during the user's registration with the service.

15. The system according to claim 9, wherein: One or more behavioral models are compared to each other using at least one of: a scatter plot matrix, individual scatter plots, and a statistical significance test.

16. The system of claim 9, wherein: The predetermined service environment is determined based on an analysis of a behavior model received from the model transmitter.

17. A non-transitory computer-readable medium having stored thereon computer-executable instructions for generating separate content for users of a service, the computer-executable instructions comprising instructions for: collect data about the behavior of users of computing devices; Based on the collected data, a user behavior model is trained, where: Data trained to identify the user with a predetermined reliability based on the collected data, wherein the data related to the user's behavior includes at least one of the following: one or more applications being used by the user, the time the user uses each of the one or more applications, and an action performed by the user in each of the one or more applications; Authorizing the user to use the service based on a comparison result of the user behavior model provided by the model transmitter and one or more models obtained from a behavior model database; and According to the trained model received from the model transmitter, separate content is generated for the user of the service based on a predetermined service environment, wherein the predetermined service environment includes an environment for accessing at least one of: a list of goods and services to which the user is authorized to access, and data that the user is authorized to access.

18. The non-transitory computer readable medium of claim 17, wherein: The data related to the user's behavior also describes at least one of the following: search queries issued by the user, one or more websites visited by the user, actions performed by the user on each of the one or more websites visited, and forms filled out by the user.

Images