EP3494494A1

EP3494494A1 - Means for broadcasting a personalized content in a communication network

Info

Publication number: EP3494494A1
Application number: EP17757809.3A
Authority: EP
Inventors: Eric Mathieu; Cyril MARCH; Bruno PRELY
Original assignee: Qwant SAS
Current assignee: Qwant SAS
Priority date: 2016-08-05
Filing date: 2017-08-04
Publication date: 2019-06-12
Also published as: FR3054903B1; US20210049647A1; KR20190049722A; FR3054903A1; JP2019527906A; CN110199275A; WO2018024999A1; EA201990461A1

Abstract

The present invention is adapted to be implemented for a search engine allowing iterations during the search for content. The invention determines and broadcasts a personalized content, by: • collecting information relating to at least one interaction of a user • determining at least one centre of interest of the user, • determining at least one measurement of the commitment of the user, • determining a context characterization related to the current session and a characterization of a degree of interest of the user • broadcasting, to the navigation module, a content personalized as a function of the characterization of the centre of interest and of the degree of commitment determined.

Description

Means for broadcasting personalized content in a communication network

The present invention relates to the field of broadcasting, in an information system, of personalized information in an automated manner. The invention relates more particularly to means for implementing, in a communication network, spaces for distributing personalized information to a user, according to the interests of said user. The information dissemination spaces can be in particular advertising spaces, or spaces reserved for recommended content, intended to be displayed, or in a page of a website accessed by the user, during a session open by means of a web browser software, either within an application embedded on a mobile device (mobile phone, also called Smartphone according to the Anglo-Saxon term, or tablet) during its consultation. Many websites, and applications on mobile equipment, offer to customize a part of the content - in particular, but not exclusively, advertising messages - distributed to a user, depending on the activities of the latter on the multiple websites previously consulted. For example, online advertisers are particularly interested in the possibility of identifying, upstream, the users interested in a presented content, and in particular the users being likely or intending to buy a product.

A first approach, adapted to websites with a search engine, may consist of displaying personalized content, based for example on keywords entered by the user and used to explain a search query. If this first approach can be effective, the content customization is insufficient for the precise and real-time understanding of the interest, the commitment of the user (that is to say, his receptivity vis-à-vis a content), and its own intentions. Cookies, more commonly referred to by the English term "cookies", managed by the web browser, or by connecting to an application on a mobile device with the user's name and password ( for example when downloading), are typically used for this purpose. In the case of the dissemination of advertising, the cookies thus make it possible to uniquely identify the user as a consumer and to offer him an advertising message on the current site. or the app viewed, in connection with his previously recorded interests and past searches. This method of personalization of content, based on these cookies, is usually referred to by the terms "Advertising Retargeting". Advertising targeting is then based on the individual recognition of the user associated with its history of previous consultations, and allows to target the user based on past interests, hence the term "retargeting" The user is already known to the advertiser.

Thus, most of the customization methods used by existing profiling systems require uniquely identifying the user, and the customization of the content relies on the recording of a data history specific to that user. The recorded data is collected from the same site - for example an e-commerce site - or from several sites, grouped together in the same network of purchase of advertising space, or recommendations from content. In this latter configuration, the same identification and profiling service of the user is typically used to store the profile of the user, including for example information relating to his interests (previous sites and pages consulted, previous searches made with keywords) and its behavior during previous visits to one or more websites.

However, uniquely identifying a user poses many problems. Indeed, although it is not always necessary to know the identity of the user or his marketing profile (for example, his name, if he is a man or a woman, the socio-demographic segment ... ), this practice may, in some cases, be in conflict with the rules on privacy, prescribed by various laws depending on the jurisdiction (s) concerned. In addition, navigation software and additional technical means can limit or even prohibit the use of cookies, which may render this method inoperative, in the absence of a relevant targeting method.

Therefore, there is still a need for means for broadcasting personalized content in a communication network that can guarantee the anonymity of the user, and that are able to operate without the use of cookies. The aim of the invention is to provide means for implementing, in a communication network, spaces for the distribution of personalized information to a user, according to the current interest of the user, and therefore specific to its current consultation, and not to its past navigation, or even to other sites. The information dissemination spaces may be in particular advertising spaces, intended to be displayed on a page of a website accessed by the user, during an open session using a web browser, or within an application on mobile equipment.

The invention also aims to propose technical means adapted to the implementation of an advertising space selling infrastructure, based on the iterative refinement of the user's interest center from data extracted from the content of the website accessed during the current session of the user, and regardless of the user's past browsing history on that same website or other websites. The invention also aims to take into account the commitment of the user with the content presented during a session, taking into account the user's interactions, and the way the content is consumed.

The invention also aims to allow the characterization of the behavior of a user, anonymously, vis-à-vis a subject in terms of the degree of commitment that is to say the receptivity of the subject. The invention also aims to allow the characterization of the behavior of a user vis-à-vis a subject, and anonymously, in terms of the intention of the user, so as to allow the determination of a likelihood that the user will interact with an advertisement, and, if necessary, redirect the user to another site from which a purchase could be made.

The invention also aims at proposing technical means for an advertising space selling infrastructure based on iterative refining, these means being able to enable characterization and improvement in precision of the interest center by iterative refining, in a function of the entire path in the consulted content, namely browsing the pages, the actions on the content, the various searches by menus and filters or by keywords, the content selections in positive and / or in negative , and more generally all the successive actions and requests made for current research, including their concatenations and their variations within the iterative cycle of research. The invention is therefore not limited to the expression of the last action or query of the search in the content. One or more of these objects are fulfilled by the broadcast method, the computer program, the medium, the module, and the system objects of the independent claims. The dependent claims further provide solutions to these objects and / or other advantages.

In the present description, by "center of interest", one refers to what characterizes and synthesizes the "context", namely the content itself of the pages viewed, the metadata of the pages, which may not be displayed, but also the progress of the session, such as the sequence of interactions, the navigation path, any selection of content, and more generally the content itself as well as all the actions to obtain the content displayed in the course of the session. The focus includes, among other things, content selection actions such as:

• navigation actions in the content and themes of the pages viewed;

• the actions available by menu, by filtering;

• the user's comments on content (for example, adding a "like" statement, etc.);

• Content search actions that can use keywords, or any other actions on content: I like / I do not like, ... and can also contribute to the search for content by a textual search engine and multimedia.

In the present description, the term "user commitment" refers to the user's receptivity to the content, which can be measured by the way the content is viewed and the user's interactivity with the user. content. The commitment is essentially of a behavioral nature, of which here are some examples for illustration:

• reading an article until the end, which can be identified by scrolling with the help of the elevator of the corresponding page until the end of the article;

· Allowing a video to be played without closing or terminating the video in advance; • Interacting with a website or an application that allows you to express a preference by "Like" / "I do not like", denotes a greater or lesser interest in the content presented, but also a commitment strong through interactivity;

· Etc.

In the present description, by "engagement variable" in a number of aspects, reference is made to user engagement information, such as, for example, a percentage of the displayed portion of an article. , a display time of the article, a video playback percentage, a number of clicks during the session, etc., as well as any combinations of variables that can be calculated (for example on a scale of 0 to 10), and which can also make it possible to determine more global parameters, such as:

• a degree of interest: reading / attention to content (scrolling text, view images, view videos);

• a degree of interactivity with the content: partial or complete content scrolling, fast or slow, searches, navigation, ...

In this description, "degree of intent" refers to the likelihood of clicking links to other content or advertisements presented.

More particularly, according to a first aspect, the invention relates to a method for broadcasting a personalized content, in a communication network, to a navigation module, during a session established between the navigation module and a content server module, said current session. The content server module may be a website, or a data source adapted for use by the navigation module. The navigation module can be in particular a web browser, or an application adapted to be executed by a terminal and configured to use data provided by the content server module. The method comprises the following steps:

A first step of collecting information relating to at least one interaction of a user, by means of the navigation module, with at least one page of content developed at least partially from data transmitted by the content server module; A second step of determining at least one context variable relating to at least one point of interest of the user, as a function of the information relating to the said at least one user interaction collected during the first step; ;

A third step of determining at least one user commitment variable with said at least one content page, based on information relating to said at least one user interaction collected during the first step ;

A fourth step of characterizing the current session according to a historical set of data of anonymous previous sessions:

context according to said at least one context variable relating to at least one center of interest of the user; and o said at least one measure of the user's commitment or at least one measure of the degree of interest of the user, according to said at least one commitment variable determined during the third step ;

A fifth step of broadcasting to the navigation module, a personalized content according to the characterization of the context and the characterization of said at least one measure of the degree of commitment or said at least one measure of the degree of interest, determined during the fourth stage.

During the fifth step, the personalized content for the user can be determined and / or selected by a content delivery module, according to the characterization of the context and the characterization of said at least one measurement of the center. of interest, and said at least one measure of the degree of commitment determined during the fourth step. The content delivery module may belong to the system according to the fifth aspect, or be managed by a third party.

During the first step, the collection of information relating to at least one interaction of a user can be reduced to the collection of basic information, for example the collection of a single keyword, during a first search for content. In particular, said at least one content page may be a web page transmitted to the navigation module during the session. The invention makes it possible to take into account information going beyond a first characterization of the center of interest obtained as a result of the user entering one or more initial keywords or a semantic summary. one page. The invention makes it possible to take into account several successive pages of content or several iterations of searches based on keywords, and any preferences, in positive or negative, expressed by the user on content presented successively. The invention therefore makes it possible to take into account the progression of the search undertaken by the user, in particular the information entered or selected, but also the information consulted during a succession of search queries having generated multiple lists of results. . It is then possible to optimally characterize the interest of the user, which is not reduced just to the consideration of the last request, or interaction of the user, and therefore provide content relevant custom related to the entire current session of the user and until his last query or interaction included.

The steps of the method according to the invention may also be repeated, for example periodically or on detection of a particular event, such as a new interaction of the user with the navigation module.

The navigation module can be implemented from a personal computer, or from an application on mobile equipment. In particular, the first step may be executed periodically, or after detecting a new interaction of the user on a content page transmitted by the content server module. Thus, it is possible to determine and improve with each new interaction from the user, the personalized content offered to the navigation module.

The method of broadcasting personalized content according to the first aspect is adapted in particular to be implemented by a system for selling advertising space by refining. The method according to the first aspect thus makes it possible to refine, by iteration, on the one hand, the current interest of the user, in terms of current focus, and, on the other hand, the extent of his commitment. with respect to the content presented, in relation to its current focus, and / or the prediction of the degree of commitment, such as the potential purchase intent (in the case of commercial products or services). By collecting on one and the same website, or on the a content server supplying the application, the content of the pages viewed, the keywords as well as the interaction data or any other means for searching and accessing new information, and only for the current consultation session and / or search of the user, the method makes it possible not to use the user's own browsing history on the Internet, and does not require, in particular, the use of cookies for the identification of the user. user and access to his history. Indeed, the method according to the first aspect makes it possible to analyze the current activity of the user, that is to say his activity during the current session, and to deduce the signature of the current session, in the compared to multiple previously profiled and anonymized sessions, compared to the activity of previous users who accessed the same types of information on the same website and had similar behaviors and contexts.

The activity of the user can be characterized and refined iteratively by the information on the observed context - which can be keywords in the case of a search engine used to obtain the information visualized - by the center. real interest of the user obtained by deducting the current navigation of the user on the site currently browsed, and the level of interest of the user for the subjects of the pages viewed. The level of interest of the user for a subject is determined in particular by the way he interacts with the information of the site, by the way he searches for information (by links, by keywords, the sequence of interactions , etc.) and the different formats of content viewed (text pages, images, and videos, and their combinations). This level of interest can be compared with similar experiences from other users, and can be characterized by the degree of interest in the content or degree of intention in the potential purchase sense, when it comes to marketable products or services.

Said at least one interaction of the user with said at least one content page, for example a web page transmitted by the content server module, may relate to one or a combination of interactions from the following list, but not limiting: consultation of a page of content / web, consultation of a subsection of a page of content / web via a link or a menu, entry of a keyword, expression of an interest or a non-interest for a multimedia document, action and / or input of the user to perform content search, navigation action and / or user input to access new information within the content server, or the website. During the second step, said at least one context variable can be determined, depending on the content of said at least one page of content or web page transmitted to the navigation module during the session and / or non-current information. visible to the user attached to said content, such as metadata.

During the second step, said at least one context variable can be determined, if necessary, according to a selection and / or a combination, at the input, of criteria, and / or of actions and / or objects used by the user to formulate, by means of the navigation module, a search request for personalized content, intended for a search module associated with the content server module. In this context, the criteria considered may be based on the terms used by the user, either directly or indirectly, for example in the case of metadata associated with selected documents.

During the second step, the said at least one context variable can be determined, according to a selection and / or a combination, at the input, of at least one textual information and / or at least one object multimedia, for formulating the request for personalized content, for the search module associated with the content server module. The request can in particular be from a search engine. Said at least multimedia object may in particular be a multimedia object without limitation on the type of computer format of the object, for example a video, an image, a sound file, a pdf file, an E-Commerce product sheet ... The at least one multimedia object can be selected by the user, directly or indirectly. The request can be built by successive iterations so as to refine the personalized content during the same session. In particular, the request can be constructed by iteration of user selections of multimedia objects and / or textual entries, and by proposals, combinations and successive refinements of the results by the search engine. By way of example, the search module associated with the content server module may be implemented by the means described in the document WO 2014/191703.

During the second step, said at least one context variable can be determined, depending on the content of said at least one content page transmitted to the navigation module, or keywords collected during the first step, and according to at least one reference ontology relating to a plurality of concepts. The plurality of concepts of the reference ontology for example relates to the subjects consulted in the field, or that may be relevant as advertising targeting advertising segments.

During the third step, said at least one user commitment variable with said at least one content page transmitted by the internet content server module can be determined according to information relating to said at least one interaction of the Internet content server module. a user collected during the first step.

The historical data set of previous sessions comprises, for example, anonymous data collected during user sessions established between third-party navigation modules and the content server module. User sessions may in particular correspond to sessions of users having previously accessed the content on the content server module, without necessarily having been identified.

The characterization of the context may in particular be carried out in order to develop said at least one point of interest of the user. Also, during the fourth step, the characterization of the context can be determined:

• developing a signature of the session, according to said at least one context variable;

• comparing the signature of the session with the historical data set of anonymous user previous sessions, to determine earlier logged sessions having a signature having the greatest proximity to the signature of the session;

• Using a classification model, to determine the characteristics of the current session of the user, according to characteristics associated with the historized sessions having a signature having the greatest proximity, in terms of classification, with the signature of the current session .

To determine the similarity with respect to a classification model, the proximity of the current session can be calculated according to a distance function between sessions according to automatic learning techniques. This classification model can be constrained, for example to reflect a marketing micro-segmentation, using machine learning techniques or machine learning in order to correspond to a reference ontology of the domain. about the accessed, or an ontology suitable for an advertising targeting use, to define the marketing segment. In the same way, the iterative browsing of the content by the user, but also during a search of content by an iterative search engine, makes it possible to implement reinforcement learning techniques. allowing automatic learning. The purpose of this machine learning is to learn from the historized sessions and to propose, for any new session, customized content or targeted advertising in an optimized way according to predefined engagement criteria, such as, but not limited to , a click rate on the content, or a click rate on advertising.

In the fifth step, the custom content can be:

Selected from among a set of contents, according to the characterization of the context according to said at least one measure of the center of interest and the characterization of said at least one measure of the degree of commitment;

· Transmitted to the navigation module;

• displayed by the navigation module.

The selection from a set of contents can be made by a module of a website, on the content server, or by a module hosted on a third party server responsible for providing personalized content, or so-called contextual advertisements.

According to a second aspect, the invention relates to a computer program comprising instructions for performing the steps of the method according to the first aspect, when said program is executed by a processor. Each of these programs can use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, for example in a partially compiled form, or in any form what other form is desirable. In particular, it is possible to use the C / C ++ language, the Java ™ language, scripting languages, such as in particular such, JavaScript, python, Perl which allow a on-demand code generation and do not require significant overhead for their generation or modification.

According to a third aspect, the invention relates to a computer-readable recording medium on which is recorded a computer program comprising instructions for executing the steps of the method according to the first aspect. The information carrier may be any entity or any device capable of storing the program. For example, the medium may comprise storage means, such as a ROM, for example a CD-ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a diskette or a hard disk. On the other hand, the information medium may be a transmissible medium such as an electrical or optical signal, which may be conveyed by an electrical or optical cable, by radio or by other means. The program according to the invention can be in particular downloaded on an Internet or Intranet network. Alternatively, the information carrier may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method in question.

According to a fourth aspect, the invention relates to a module adapted to broadcast personalized content, in a communication network, to a navigation module, during a session established between the navigation module and a content server module. . The module includes:

A module configured to collect information relating to at least one interaction of a user, by means of the navigation module, with at least one page of content developed at least partially from data transmitted by the content server module;

A module configured to determine at least one context variable relating to at least one center of interest of the user, as a function of the information relating to said at least one user interaction;

A module configured to determine at least one commitment variable of the user with said at least one content page, based on the information relating to said at least one interaction of a user;

• a module configured to determine, based on a historical set of anonymized previous session data: o a characterization of the context according to said at least one context variable relating to at least one center of interest of the user; and

a characterization of said at least one measure of the user's commitment or at least one measure of the degree of interest of the user, according to said at least one commitment variable;

A module configured to broadcast, to the navigation module, personalized content according to the characterization of the context and the characterization of said at least one measure of the degree of commitment or said at least one measure of the degree of interest.

According to a fifth aspect, the invention relates to a system comprising a navigation module, a content server module and a module according to the fourth aspect.

Other features and advantages of the present invention will appear on reading the description which follows, made with reference to the appended drawings, in which:

FIG. 1 is a schematic representation of a system for implementing personalized information dissemination spaces intended for a user, according to one embodiment of the invention;

FIG. 2 is a block diagram of the steps of a method for predicting the behavior of the user, according to one embodiment of the invention;

FIG. 3 is a diagram, according to one embodiment, of a navigation module according to the invention.

In this description, the term "session" refers to the period during which a user produces a series of queries using his browser software during a visit to a website, or from time to time. an application on mobile equipment accessing the content server. Thus, in the context of access to a content server site, a session refers to a delimited period during which the same content server communicates and performs operations for the benefit of a user, connected by means of a module computer running content access software. To determine the period corresponding to a session, a session management mechanism, for example using, among other means, the session management mechanism associated with the user interface, may be used to determine for each new interaction of the session. user, if necessary, the start of a new session, the continuity of a session already started, and / or the end of the session, for the same internet browser on the same terminal, for the same application on the equipment mobile, based on computer criteria, such as the creation of a new session by the browser, or time criterion, following the expiration of the session on the basis of a time counter of inactivity. Similarly, the preceding criteria can be supplemented or combined with a thematic consistency criterion developed from the context, and based on the semantic proximity of the content consulted according to a reference ontology, within a session management module allowing force the end of the current session, in the case of too much divergence of subjects, and thus trigger the start of a new session.

Referring to Figure 1, which shows, according to one embodiment of the invention, a system for implementing custom information dissemination spaces for a user. More particularly, the system may be an advertising space implementation system with iterative characterization and refinement of a user's intent in terms of focus and engagement. The system is configured to collect and use, for this purpose, information relating to contextual and behavioral data collected during interactions of the user with the pages of a website, or with an application residing on mobile equipment. This user-specific information is typically obtained during the same browsing or search activity, possibly using a search engine, on the site currently browsed by the user. The collection of this information is limited in time, for example during a single session, most often characterized by a limited duration, and the same focus of interest on the part of the user.

The system comprises a navigation module 110 adapted to be in particular a browser software, or an application on a mobile device, such as a smartphone or tablet. The navigation software allows a user of the navigation module 110 to access the pages of a website or a website. content server, and to consult and / or interact with said pages or content. The navigation module 110 may in particular be embedded in a computer, in a mobile communication terminal, in a connected system, etc. The system comprises a content server module 120, coupled to the Internet and hosted by or on behalf of a content editor, adapted to store, generate and transmit web pages or content, to the navigation module 110, upon request of this last. The web server module 120 is coupled to a session profiling module 130. The session profiling module 130 uses one or more artificial intelligence, clustering, reinforcement learning techniques to determine the signature. of the current session, and recalculate it with each new iteration, taking into account the history of the entire session. The signature of the recalculated session during the last iteration is compared and classified against previous anonymized sessions of users. Classification profiling allows for the development of at least one point of interest, and at least one measure of the degree of commitment. The session profiling module 130 then transmits to the custom content server editor module 140 the focus of the session that can be the label, namely the name of the session cluster in which the session was assigned, and increased by additional attributes that may specify a finer classification. Similarly, the session profiling module 130 determines the commitment level and delivers one or more parameters qualifying the user's commitment, namely the degree of interest or intention with respect to the content presented.

The system includes a custom content server editor module 140, coupled to the session profiling module 130 and a content delivery module 150, which can also be operated by a third party. Said at least one point of interest and at least one degree of commitment are provided by the custom content server editor module 140 by an API to the content delivery module 150 which determines the content most suited to the content of the content. user according to the elements provided to him and an internal logic of decision. The content delivery module 150 may in particular be a suitable system for advertising sales, or recommendation of publisher content, for making available content provided by a plurality of advertisers, or third-party publishers, in from source data 162, 164. Thus the personalized content is determined by the content delivery module 150 outside the environment of the user without a connection indicator retrieved within the navigation module 110 ( for example of the type cookie, tag, ...) and without direct interaction with this environment. Likewise, the identity information (or any other means of user recognition) is not used in the determination of the personalized content, since only the information from the anonymized session provided by the server editor module custom contents 140, are used for decision making of the content delivery module 150.

The system includes a custom content server module 170, coupled to the content delivery module 150, which will implement the personalized content interacting with the navigation module 110.

The personalized content server 170 is also coupled to a tracking module of the display of the personalized information 190, as displayed on the navigation module 110, and interacting with the content server module 120 and / or the navigation module. 110, to measure the display of personalized content, thereby guaranteeing the proper performance of a display contract, it may be of commercial type with third parties.

The system may include a content distribution network 180, coupled to the navigation module 110 and the custom content server advertiser module 170, to optimize its efficiency in terms of response time and display.

FIG. 3 represents a detailed embodiment of the navigation module 110. The navigation module 110 comprises a user interface 310, typically a screen, and means of interaction with a user such as a keyboard, a touch surface, sensors, etc. The user interface is particularly adapted to display contents transmitted by a content server module 120.

The navigation module 110 comprises a collection module 320 of session data items. The collection module 320 is configured to obtain, by means of the navigation module 110, information relating to the interactions of the user with the content presented, including for example web pages, provided by the content server module 120. The information collected (logged in the Anglo-Saxon word) by the collection module 320 are for example related to the content pages viewed, keywords entered by the user, actions and / or inputs of the user to search and / or navigate to access new information within the same site - using links filters, all kinds of searches, for example by image, etc.

The navigation module 110 comprises an iterative classification module 330, operating in real time, coupled to storage means 340 for the current session and to a session history database 350. The iterative classification module 330 is notably configured to calculate the context variables of the current session, namely at least one interest-point variable and at least one commitment variable, for storing in the storage means the information collected by the collection module 320, and to provide the necessary information to the content delivery module 150.

The session history database 350 is coupled to a session profiler module 360 for classifying the historical session database 350, by constructing at least one interest-point variable, and at least one an engagement variable for each stored session.

The iterative classification module 330 is configured to transmit to the content delivery module 150 at least one context information qualified by a user's center of interest, determined according to:

The content of the pages viewed during a navigation session established by means of the navigation module 110; and or,

A selection, by means of a semantic or machine learning method, of the most significant terms characterizing its search query for content, whether used keywords or content selected by the user for formulating its request, using the navigation module 110, for a search module of the content server module 120.

The session history database 350, coupled to a session profiler module 360, stores a historical set of data from previous sessions, said session data typically being previously obtained. from a plurality of anonymous data collected during sessions, on the same website for many users. The session profiler module 360 is further configured to allow a signature for each session of the historical data set of previous sessions to be determined by constructing at least one interest-point variable and at least one a commitment variable, to thereby classify the sessions by degree of similarity (so-called "Clustering" method).

Referring to Figure 2, which are shown, according to one embodiment of the invention, the steps of a method of predicting the behavior of the user. The prediction of the user's behavior is expressed according to the interests of the user and the user's commitment to a content. The content is, for example, one or more web pages accessible locally, third-party content inserted in the content page, advertising messages. The recommended content, for its part, is typically generated by the content delivery module 150, from one of the data sources 162, 164 of a third party publisher or an advertiser, wishing to target the current context. of the user, but also more finely his real interest. The method of predicting the interest of the user is adapted to provide personalized content to the user by analyzing:

• the interest of the user qualifying the context, according to:

classified and characterized content of the viewed content, for example page views of a website, during a navigation session established by means of the navigation module 110; and or

a selection, by means of a semantic method, or learning and profiling by "Machine Learning" of the most important terms among terms used by the user during his session to formulate, by means of the module navigation 110, a content search request, for a search module of the content server module 120;

• the commitment of the user, in a behavioral way, according to the interactions on the pages transmitted by the content server module 120 to the navigation module 110. Commitment, of a behavioral nature, with content and in particular advertising engagement or commitment to advertising means being receptive to content and interacting with a content or advertising element. Reading an article down to the bottom of the article, or watching a video to the end, are examples of the user's commitment to the content. In the field of internet advertising, the advertising engagement is often the fact for a user to interact with an advertising element beyond and otherwise than by clicking. The concept of commitment is mainly used for so-called "rich media" formats in English, and is an indicator of efficiency complementary to click rates and indicators after-viewing (or "post-view" in English). For example, commitment can be achieved by:

• extend an expandable banner;

• activate the reading or the replay of an advertising video;

• play within a banner;

· To replay a so-called "rich media" creation;

• to mark with a "like" mention a content or advertisement.

Considering an interaction as a commitment is sometimes abusive, since a mouse hover can be totally unintentional. As a result, the measure of commitment is better when it is based on several acts of engagement, or can be derived from iterative actions. The engagement can be measured by one or more engagement rates of different natures, related for example to interactions relating to text, images, and / or videos. The method for predicting the behavior of the user comprises a step 210 of collecting, during a current session, information relating to new interactions of the user, by means of the navigation module 110, with pages provided. by the content server module 120. Advantageously, the step 210 is executed periodically, or after detection of a new interaction of the user on a content or a web page provided by the content server module 120 to the navigation module 110 The information relating to the new interactions is collected for one and the same Internet site, and is for example related to the content of the pages viewed, to the keywords entered by the user, to the actions and / or inputs of the user to perform search and / or access new information within the same site - links, filters, search by image, etc. The collection of information thus relates exclusively to data accessible during the current session of the user: no information relating to a browsing history of the user on the internet is collected. In addition, cookies are not used to identify the user.

The method for predicting the behavior of the user comprises a step 220 for accessing, during a current session, information relating to interactions of the user, by means of the navigation module 110, with pages provided. by the content server module 120. Typically, it is information relating to the visualized content and interactions that the user has generated, prior to the current session, updated until its last interaction. The information relating to interactions is obtained for a single Internet site, and is for example related to the content of the pages viewed, to the keywords entered by the user, to the actions and / or inputs of the user to perform a search and / or access new information within the same site - links, filters, search by image, etc. The collection of information thus relates only to the data accessible during the current session of the user: no information relating to a browsing history of the user on the internet is collected. In addition, cookies are not used to identify the user.

The method for predicting the behavior of the user comprises a step 230 of merging the information relating to the new user interactions obtained during step 210, with the information relating to the user's interactions obtained during the process. step 220, namely with the history of its current session.

The method for predicting the user's behavior comprises a step 240 of determining and refining the context making it possible to define and then to improve in precision the centers of interest of the user, based on the data merged in the course of time. step 230. At the end of step 240, at least one context and / or a point of interest of the user is / are determined and / or refined, by combining:

The content of the pages viewed during the current navigation session established by means of the navigation module 110; and or

A selection, by means of a semantic method, of the most significant terms associated with, or equivalent to, the user's content search requests, whatever the nature of his request, which can combine words, actions on the content, images, videos, by means of navigation module 110, for a search module of the content server module 120.

Thus, in addition to the information relating to the content of the pages consulted by the user (and the non-visible metadata that may be associated with them), to the keywords entered by the user, to the images and / or videos that have been used by the user for to formulate one or more successive search queries, the center (s) of interest of the user is / are determined (s), during step 240, according to one or more ontologies or thesauri of reference to a plurality of concepts illustrating diverse and varied centers of interest.

Similarly, the user behavior prediction method includes a step 250 of determining and refining the user's commitment, based on the merged behavioral data in step 230. More particularly, at the end of step 250, at least one measure of the commitment of the user is determined and / or refined, depending on the interactions of a behavioral nature on the pages transmitted by the content server module 120 to the module of navigation 110, obtained from the merged data in step 230.

The behavior of a user exploring a content on a website, or provided through an application, is characterized by all of the user's interactions with the human-machine interface. The behavior is elaborated from the set of events reflecting all the human actions positioned on a time axis. By way of nonlimiting examples it may be the following events:

• all types of Clicks on a page, a button, a content, ...

• running a page, launching a video and viewing time, ... · writing words in a search bar, dragging and dropping content, ...

• all actions on the keyboard, a touch screen, and more generally all interactions with the user interface.

Behavior variables will be deduced, such as, but not limited to:

• time between two actions of the user;

· Number of pages viewed, viewing time of a page;

• number of images viewed, and average time per image;

• number of videos viewed, time and percentage of viewing videos;

• number of actions depending on the type of content: text, image, video. The user engagement prediction method includes a step 260 characterizing the "center of interest" and the "commitment" of the user, based on the user's focus variables determined in step 240 and the user engagement variables determined in step 250.

At the end of step 260, a degree of user intent is obtained. The characterization of the current session of the user is obtained by:

Determining a signature of the current session, according to the context and / or the point of interest of the user determined during step 240 and the measurement of the user's commitment determined during the step 250;

Obtaining a historical set of data from previous sessions, said session data being typically previously obtained from a plurality of anonymous data collected during sessions, on a plurality of web sites, many users;

Determining, among the historical data set of previous sessions, the previous sessions having signatures having the greatest proximity to the signature of the current session;

Determining, from a classification model, the characteristics of the current session of the user, according to characteristics associated with a cluster of previous sessions having the signatures having the greatest proximities with the signature of the current session.

Thus, it is possible to analyze the current activity of the user and compare it against the activity of previous users who accessed the same types of information on the same website, from previously recorded sessions and anonymised. Machine Learning techniques can be used in particular to classify the interest and the commitment of the user in relation to a multitude of session groups, in order to propose to advertisers and publishers the purchase of advertising inserts or sponsored content in which they will offer personalized content closest to the user's interest and knowing in advance its current level of engagement with the content presented.

The user behavior prediction method includes a step 270 of selecting a personalized content from a set of content, based on the focus and commitment degree determined in step 260, then is transmitted to the navigation module 110 for display.

Claims

1. Method for broadcasting a personalized content in a communication network to a navigation module (110) during a session established between the navigation module and a content server module (120), called the current session , the method being characterized in that it comprises the following steps:

A first step (210, 220, 230) of collecting information relating to at least one interaction of a user, by means of the navigation module (110), with at least one page of content developed at least partially from data transmitted by the content server module (120);

A second step (240) of determining at least one context variable relating to at least one center of interest of the user, as a function of the information relating to said at least one user interaction collected during the the first step ;

A third step (250) of determining at least one user commitment variable with said at least one content page, based on information relating to said at least one user interaction collected during the the first step ;

A fourth step (260) characterizing the current session, based on a historical set of data from anonymous previous sessions;

context according to said at least one context variable relating to at least one of the user's interests; and said at least one measure of user engagement or at least one measure of the user's degree of interest, based on said at least one commitment variable determined in the third step;

A fifth step (270) of broadcasting personalized content to the navigation module (110), according to the characterization of the context and the characterization of the said at least one measure of the degree of commitment or the said at least one measurement the degree of interest determined in the fourth step.

The method of claim 1, wherein, in the fifth step (270), the personalized content for the user is determined and / or selected by a content delivery module (150), as a function of the characterization of context and characterization of said at least one measure of the degree of engagement or said at least one measure of the degree of interest determined during the fourth step.

3. Method according to any one of the preceding claims, wherein the first step is performed periodically, or after detecting a new interaction of the user on a content page transmitted by the content server module.

A method according to any one of the preceding claims, wherein said at least one user interaction with said at least one content page transmitted by the content server module is relative to one or a combination of interactions among the following list: consultation of a page of content / web, consultation of a sub-section of a web page via a link or a menu, entry of a keyword, expression of an interest or a non-interest for a multimedia document, action and / or input of the user to perform a content search, navigation action and / or input of the user to access new information within the content server, or the website.

A method according to any one of the preceding claims, wherein, in the second step (240), said at least one context variable is determined, depending on the content of said at least one content page and / or information not visible to the user attached to said content.

The method according to any one of the preceding claims, wherein in the second step (240) said at least one context variable is determined, based on a selection and / or combination, in input, criteria, and / or actions, and / or objects used by the user to formulate, by means of the navigation module (110), a personalized content request, for a search module associated with the content server module (120).

The method according to claim 6, wherein, in the second step (240), said at least one context variable is determined, based on a selection and / or combination, as input, of at least one textual information and / or at least one multimedia object for formulating the personalized content request for the search module associated with the content server module (120).

The method according to any one of the preceding claims, wherein, during the second step (240), said at least one context variable is determined, according to the content of said at least one content page transmitted to the module. of navigation, and according to at least one reference ontology relating to a plurality of concepts.

A method as claimed in any one of the preceding claims, wherein the historical data set of previous sessions includes anonymous data collected during user sessions, established between third party browsing modules and the content server module.

The method of any one of the preceding claims, wherein in the fourth step the characterization of the context is determined by:

· Developing a signature of the session, based on said at least one context variable;

Comparing the signature of the current session with the historical data set of anonymized past sessions of users, to determine earlier logged sessions having a signature having the greatest proximity to the signature of the session;

• using a classification model, to determine the characteristics of the current session of the user, based on characteristics associated with previous historized sessions having the signature having the greatest proximity, in terms of classification, with the signature of the current session .

The method of claim 2 and any of claims 3 to 10 wherein in the fifth step (270), the personalized content is:

Selected from among a set of contents, according to the characterization of the context and the characterization of said at least one measure of the degree of commitment or said at least one measure of the degree of interest;

Transmitted to the navigation module (110);

• displayed by the navigation module.

A computer program comprising instructions for executing the steps of one of the methods of any one of claims 1 to 11 when said program is executed by a processor.

A computer-readable recording medium on which a computer program is recorded including instructions for executing the steps of the methods according to any one of claims 1 to 11.

A module adapted to broadcast personalized content in a communication network, to a navigation module (110), during a session established between the navigation module and a content server module (120), characterized in that 'it comprises :

• a module configured to determine, based on a historical set of anonymized previous session data:

o a characterization of the context according to said at least one context variable relating to at least one center of interest of the user; and

A module configured to broadcast, to the navigation module (110), personalized content according to the characterization of the context and the characterization of said at least one measure of the degree of engagement or said at least one measure of the degree of interest.

System comprising a navigation module (110), a content server module (120) and a module according to claim 14.