FR3110727A1 - Method, device and computer program for identifying and ordering user selectable events - Google Patents

Abstract

The present invention relates to the identification and determination of an order of a subset of events of a set of events, each event being selectable by a user and being associated with a time, a category and an elements, according to at least a history of selections of events. After having obtained a tree associating events of which at least one instant belongs to a given period of time, the events of the tree being associated with each other by their category and / or their elements according to paths comprising at least one formed node events, categories or elements, and obtained a selection history comprising previously selected events, a subset of events linked to an event belonging to the selection history via a path is identified, the instants of the events identified being subsequent to the events in the selection history. An order of the events of the subset is then determined.

Description

Method, device and computer program for identifying and ordering user selectable events

The invention relates to the identification and determination of an order of events of a set of events, to allow a user to easily and quickly access events likely to interest him, according to a history Selection. The invention applies in particular to the field of online games, in particular sports betting.

In recent years, games accessible via a communication network, through a man-machine interface, have been very successful. By way of illustration, the number of players making “online” bets on sports competitions has more than tripled over the past five years. Thus, for example, a player can, using his computer, his tablet or his smartphone, access, via the Internet, a portal of a service provider offering online games, select a match on which he wants to bet. For these purposes, he may have an account managed by the service provider who can then accept or refuse bets and manage winnings and losses.

However, despite this success, many players are reluctant to use such services due to the difficulties encountered in identifying the competitions likely to interest them. Indeed, the catalogs of competitions that can be the subject of bets are generally immense. They can include one or several hundred daily competitions. It is observed that there are not only competitions by sport but also local, regional, national or international competitions, competitions for amateurs or professionals, competitions according to predetermined classes, etc. Thus, even by selecting a particular sport, it is often very difficult to quickly access the competitions sought. The player thus loses opportunities for games and winnings.

In addition, players increasingly use mobile devices such as smartphones, equipped with small screens and limited means of interaction, in particular generally not including a keyboard. Finding competitions with such devices often proves particularly difficult.

There is therefore a real need for a system for analyzing and identifying events likely to interest a user, for example sports competitions on which a user can place bets remotely, using a computer , so that this user can quickly identify and select, using a simple interface, the events that interest him and, if necessary, can initiate an action, for example a bet.

The present invention aims in particular to solve these problems.

To this end, the invention proposes a mechanism for identifying and ordering events in order to allow a user to simply and quickly select the events that interest him.

• l’obtention d’au moins un arbre associant des événements dudit ensemble d’événements dont au moins un instant appartient à une période de temps donnée, les événements de l’au moins un arbre étant associés les uns aux autres par leur catégorie et/ou leur au moins un élément selon des chemins comprenant au moins un nœud formé d’au moins un événement, au moins une catégorie et/ou au moins un élément ;
• l’obtention d’au moins un historique de sélections comprenant au moins un événement préalablement sélectionné ;
• l’identification, dans l’au moins un arbre obtenu, d’un sous-ensemble d’au moins un événement relié à au moins un événement appartenant à l’historique de sélections via un chemin comprenant au moins un nœud formé d’au moins un événement, au moins une catégorie et/ou au moins un élément, les instants des événements identifiés étant postérieurs aux événements de l’au moins un historique de sélections ; et
• la détermination d’un ordre des événements du sous-ensemble d’au moins un événement.

There is thus proposed a computer method for identifying and ordering a subset of at least one event from a set of events, each event being selectable by a user and being associated with at least one instant, at least one category and to at least one element, the identification and the determination of an order of events being carried out according to at least a history of event selections, the method comprising:

• obtaining at least one tree associating events of said set of events of which at least one instant belongs to a given period of time, the events of the at least one tree being associated with each other by their category and/ or their at least one element along paths comprising at least one node formed of at least one event, at least one category and/or at least one element;
• obtaining at least one selection history comprising at least one previously selected event;
• the identification, in the at least one tree obtained, of a subset of at least one event linked to at least one event belonging to the selection history via a path comprising at least one node formed of at at least one event, at least one category and/or at least one element, the instants of the identified events being subsequent to the events of the at least one selection history; And
• determining an order of the events of the subset of at least one event.

The method according to the invention thus makes it possible to sort events in order to identify a limited number of events among a very large number of events, allowing a user to directly access events likely to interest him, authorizing him to quickly select events from a device provided with limited resources, for example a small screen and limited means of selection.

According to a particular embodiment, the determination of an order of the events of the subset is carried out according to an identification frequency of the events of the subset of at least one event in the at least one tree obtained.

Still according to a particular embodiment, the identification, in the at least one tree obtained, of a subset of at least one event linked to at least one event belonging to the selection history via a path is performed for all events in the selection history.

Still according to a particular embodiment, the identification, in the at least one tree obtained, of a subset of at least one event linked to at least one event belonging to the selection history via a path is performed for all paths resulting from at least one event belonging to the selection history or for all paths resulting from at least one event belonging to the selection history comprising a number of nodes less than a given threshold.

Still according to a particular embodiment, the method further comprises an update of the at least one tree obtained, the update comprising the deletion of at least one instant associated with an event, the at least one instant deleted being prior to a first instant, or the addition of at least one instant associated with an event, the at least one added instant being prior to a second instant subsequent to the first instant.

Still according to a particular embodiment, the method further comprises the selection of at least one category, the events of the subset of at least one event being linked to the at least one category selected by a path not comprising intermediate node formed by a category.

Still according to a particular embodiment, the at least one category is selected according to the distribution of the categories of the events of the selection history.

Still according to a particular embodiment, the at least one category is selected if the proportion of events of the selection history linked to the at least one category is greater than a given threshold.

Still according to a particular embodiment, the at least one category is selected randomly if the proportion of events of the selection history linked to the at least one category is below a given threshold.

Still according to a particular embodiment, the at least one category is selected from preferences of a user determined from the history of event selections.

Still according to a particular embodiment, a plurality of events of the set of events belongs to at least one group, events of the at least one tree being associated with each other by their category, their at least one group and their at least one element along paths comprising at least one node formed of at least one event, at least one category, at least one group or at least one element.

Still according to a particular embodiment, a plurality of groups belong to at least one class, the events of the at least one tree being associated with each other by their category, their groups, the at least one class and their at least an element along paths comprising at least one node formed of at least one event, at least one category, at least one group, at least one class or at least one element.

Still according to a particular embodiment, the method further comprises a step of filtering the subset of at least one event.

A computer program, implementing all or part of the method described above, installed on pre-existing equipment, is in itself advantageous, since it makes it possible to easily and quickly access and select an event likely to be of interest an user.

Thus, the present invention also relates to a computer program comprising instructions for implementing the method described above, when this program is executed by a processor.

This program can use any programming language (for example, an object language or other) and be in the form of interpretable source code, partially compiled code or fully compiled code.

Another aspect relates to a non-transitory storage medium of a computer-executable program, comprising a set of data representing one or more programs, said one or more programs comprising instructions for, when executing said one or more programs by a computer comprising a processing unit operationally coupled to memory means and to an input/output interface module, to execute all or part of the method described above.

Other characteristics, details and advantages of the invention will appear on reading the detailed description below. This is purely illustrative and should be read in conjunction with the attached drawings, in which:

Fig. 1

illustrates an example of an environment in which the invention can be implemented according to particular embodiments;

Fig. 2

schematically illustrates the logical architecture of an engine for identifying and ordering events likely to interest a user;

Fig. 3

illustrates an example of engine steps for identifying and ordering events shown in Figure 2;

Fig. 4

illustrates an example of steps for selecting categories from which a search for events likely to interest a user can be carried out;

Fig. 5

illustrates an example of steps for obtaining up-to-date event trees, as described with reference to Figure 3;

Fig. 6

illustrates an example of steps for adding events to a sports betting event tree;

Fig. 7

illustrates an example of steps for browsing an event tree and selecting events likely to interest a user, as described with reference to Figure 3;

Fig. 8

represents an example of a tree of events linked to sports betting linking teams (elements) to matches (events) themselves linked to types of competitions (groups) which, in turn, are linked to countries (classes ) themselves related to sports (categories);

Fig. 9

represents an extract from the event tree of FIG. 8;

Fig. 10

, comprising Figures 10a to 10d, illustrates an example of a search for future events that can be offered to a user, in the event tree illustrated in Figure 9, from a past event already selected by the user ; And

Fig. 11

illustrates an example of a device that can be used to implement, at least partially, an embodiment of the invention, in particular the steps described with reference to Figures 3 to 7.

detailed description

According to embodiments of the invention, user-selectable events are organized into one or more event trees spanning a past and future time period. These trees are filtered according to criteria based on a user's selection history and the results obtained are ordered in order to present the user with future events likely to interest him, classified in order of relevance. Events, which are based on elements, can be organized into groups, each group can belong to one or more classes which can themselves be organized into categories.

By way of illustration, the following description is based on an example of sports betting, the invention making it possible to select and order the competitions likely to interest a player. According to this example, events can be matches, elements can be teams, groups can be competition types, classes can be countries and categories can be sports. However, it should be understood that the invention is not limited to this particular example.

Figure 1 illustrates an example of an environment in which the invention can be implemented according to particular embodiments. As shown, a user equipped with a personal computer 100, a tablet 105 or a smartphone 110 can access, via a communication network 115, for example the Internet, a portal hosted on a server 120 of a supplier Services. This can be, for example, a standard web portal. The service provider here manages games, for example bets on competitions whose characteristics are stored in a database of a storage server 125, local or remote.

The device used by the user to access the service provider's portal can use wired communication means to access the communication network 115, for example ADSL type communication means (for Asymmetric Digital Subscriber Line in English terminology ) or wireless communication means, for example 3G, 4G or 5G type telecommunication means.

An example of the architecture of the device 120 is illustrated in figure 11.

According to particular embodiments of the invention, when a user accesses the portal of the service provider, an identifier is requested from him to allow him to access his account. The user is then preferably authenticated, for example using a password or biometric means such as means based on face, iris or fingerprint recognition. Still according to particular embodiments of the invention, a user account makes it possible to manage the user's personal data (e.g. last name, first name, place of residence, date of birth, etc.), a profile (e.g. preferences regarding games, sports, types of competition, etc.) and/or an electronic purse that the user can fund and use to play or bet and to withdraw money won. Preferably, a history of the user's actions (e.g. the list of games or bets in which he participated during an elapsed period, for example a period of 3 months, with information on these games or bets) is in further associated with his account.

From the service provider's portal and their account, a user can search for upcoming games and/or competitions to select one or more and play or bet.

As described below, embodiments of the invention aim to identify and order games or bets, or more generally events, that can be selected by a user, in order to simplify and accelerate the search for such events.

Figure 2 schematically illustrates the logical architecture of an engine for identifying and ordering events likely to interest a user.

As illustrated and according to particular embodiments of the invention, the engine for identifying and ordering events 200 can use, on the one hand, a history of selections 205 of a user as well as preferences 210 of this user and, on the other hand, one or more event trees 215. Examples of event trees are described with reference to FIGS. 8, 9 and 10. Optionally, a general recommendations module can be used to identify and order events when the user's history and/or his preferences do not make it possible to identify events corresponding to the user's expectations, in a relevant way.

The engine for identifying and ordering events 200 generates a list of ordered relevant events allowing a user to quickly and simply identify and select events likely to be of interest to him, in particular using a device provided with limited resources .

FIG. 3 illustrates an example of steps of the engine for identifying and ordering events 200 illustrated in FIG. 2. The steps illustrated in FIG. 3 can in particular be implemented in a device of the service provider, for example in a module of the server implementing the access portal.

As illustrated, a first step has the object of obtaining a user identifier (step 300), denoted i . This identifier allows access to data specific to the user, in particular to the preferences of the user (step 305) and to a history of the actions that he has taken in a past period (step 310).

By way of illustration, considering the field of sports betting, the user's preferences may include a type of sport, for example football, a type of competitions, for example league 1 matches and European cup matches. Europe as well as a country, for example matches played in France and Germany.

The history of actions performed by a user includes, for example, the list of bets made by this user. An example of such a list is given in the appendix (Table 1). It preferably includes the characteristics of the actions, for example, in the field of sports betting, the match, the teams involved, the type of competition, the country where the match took place and the sport. The period covered can cover all the actions carried out or can be limited to avoid too old data which would have little or no longer be of interest. Thus, for example, only the data corresponding to a predetermined period, denoted Δ t _P , from the present instant, denoted t ₀ , can be taken into consideration. This period is preferably set by the service provider. Still by way of example, it can be 3 months.

Such a list may include additional information, for example the name of the winning team, the amount bet by the player, the winnings obtained, etc.

In a next step, one or more categories are selected (step 315). Such a category is, for example, a type of sport. As described below, particularly with reference to Figure 4, one or more categories may be selected from a user's preferences and history.

In parallel, before or after, one or more up-to-date event trees are obtained (step 320). This step is described in more detail below, with reference to FIGS. 5 and 6. A selection or a filtering of the tree or trees obtained is preferably carried out to select the event trees (or tree portions, considered by the sequence as trees) relevant (step 325). This selection or this filtering may consist, for example, in deleting the trees or portions of event trees which are not associated with the selected categories.

Alternatively, step 320 can be integrated with step 325 to obtain only the trees corresponding to the desired categories.

The selected event tree or trees are then analyzed, in particular with regard to the user's history, to identify relevant events likely to be of interest to the user (step 330). These events are ordered, from the most relevant to the least relevant, to facilitate possible selection by the user. An example of steps to identify relevant events, likely to interest a user is described with reference to Figure 7.

The results obtained can be filtered (step 335), in particular with respect to the preferences of the user, to rule out events potentially likely to interest him but which he does not intend to select, for example because of the amount of the wager or level of risk.

The results are then transmitted to a user device to be displayed, thus allowing a simple and fast selection of one or more events among a very large number of events. The selection(s) made by the user are then transmitted to the service provider, if necessary with other information such as a bet amount. The service provider may perform related actions, for example placing bets on the selected events. Such a combination of display and selection allows a user to use without difficulty a device having limited resources, in particular a small-sized screen and limited means of selection.

Figure 4 illustrates an example of steps to select categories from which a search for events that may be of interest to a user can be made.

As illustrated, a first step has the object of selecting the set of categories selected by the user i in his preferences, denoted C_S(i, j) , where j represents an index of the categories (step 400). Alternatively, only some of the categories selected by user i in his preferences can be selected. By way of example, 10% of the categories selected by user i in his preferences can be selected. They can be selected randomly.

In a next step, the set of categories of events selected by user i during a past time period Δ t _P , denoted C_H(i Δ t _P , j) , where j represents an index of categories, is obtained ( step 405). Considering the example given in table 1 in the appendix and starting from the assumption that the dates are increasing, that the date date_1 is outside the period considered (ie date_1 < t ₀ – Δ t _P ) and that the date date_2 is in the period considered, the set of categories obtained includes categories cat_1 and cat_7 .

In a next step, the most significant categories of the categories of the events selected by the user i in the past period considered are selected (step 410). For these purposes, the proportion of each category obtained, relative to all the events selected, is compared with a threshold, denoted θ. This is, for example, each category representing more than 15% of the user's bets over the past period considered. Again, taking the example of table 1 given in the appendix, only the category cat_1 , representing 89% of the bets, is selected (out of the 9 bets committed over the period considered, 8 bets relate to the category cat_1 , i.e. 8 / 9 x 100 = 89). The cat_7 category, representing only 11% of the bets (a single bet out of the 9 committed), is not selected.

In a next, optional step, one or more categories can be selected from among the categories of events selected by user i in the past period considered which, not counting in the most significant categories, have not been selected (step 415). By way of illustration, it may be the drawing of lots for a single category from among the categories obtained at step 405 which have not been selected at step 410 once a significant part (for example more than 30%) of the categories obtained at step 405 has not been selected at step 410.

This random selection can also be made from the categories which do not count in the most significant categories (i.e. which were not selected at step 410) but whose proportion, relative to all the selected events , is greater than a threshold, denoted η. This is for example a random draw on the categories each of which represents less than 15% of the user's bets over the past period considered but more than 8%.

The set of categories selected at steps 400, 410 and 415 is denoted C_P .

Figure 5 illustrates an example of steps for obtaining up-to-date event trees, as described with reference to Figure 3.

As illustrated, a first step has the object of obtaining one or more event trees (step 500). It may be, for example, a single event tree, comprising all the events that may have been the subject of a selection in the past period considered and the set of future events that may be selected by an user. It can also be the set of event trees comprising the set of events that may have been the subject of a selection in the past period considered and the set of future events that may be selected by a user. Finally, it may be one or more event trees corresponding to one or more particular criteria, for example the event tree or trees corresponding to one or more particular categories (for example the categories of events to be offered to an user).

In a next step, it is determined whether the tree or trees are up to date (step 505). This determination is, for example, based on a comparison of the date of update of the tree(s) with that of the day. If the update date of all event trees is today's, the algorithm terminates. Otherwise, each event tree whose update date is earlier than today's date is updated. Such an update preferably comprises the deletion of old events, for example events not belonging to the considered past period ( Δ t _P ), and the addition of future events that can be selected by a user. The period considered for the events to be added can be limited to the current date or can include the current date and the following days. This period to which belong the future events that can be selected by a user is denoted Δ t _F .

In a following step, all the events whose dates are prior to the past period considered ( t ₀ -Δ t _P ) are selected from the tree of events being updated (step 510) in order to be deleted (step 515). According to certain particular embodiments, the deletion comprises the simple deletion of the old date associated with the event. According to other embodiments, the deletion includes the deletion of the event in the event tree (provided that this event is not also associated with a date belonging to the past period considered or to the future period considered ).

Then, the events to be added, whose dates belong to the future period considered ( t ₀ -Δ t _F ) are obtained with the associated information (step 520), for example elements, a group, a class, a category and a date. These events can then be added to the event tree (step 525). If the event already exists in the event tree, only the new date is added. Otherwise, an event is created.

Figure 6 illustrates an example of steps for adding events to a sports betting event tree.

A first step aims to select the first event to add (step 600), here a match, bearing the index m ( m varying from 1 to the number of events to add) and to obtain the associated information, for example the date, the category (sport), the class (country where the match takes place), the group (type of competition) and the elements (teams).

After obtaining this data, a first test is performed to determine if the match (event) exists in the event tree (step 605). Indeed, some matches are played regularly, for example every year. In this case and according to certain embodiments, a new event is not created as such but a new date is added to an existing event. Thus, if the selected match exists in the event tree, the new date of the match is added to this event (step 610).

Otherwise, if the selected match does not exist in the event tree, it is created.

For these purposes and if the event tree to be updated is a single event tree, a second test is performed to determine whether the sport (category) corresponding to the match (event) selected exists in the tree (step 615). If the sport corresponding to the selected match does not exist, it is created (step 620).

If several event trees are used, for example one event tree per sport, and if there is no event tree associated with the sport corresponding to the selected event, such a tree is created.

In a next step, a third test is performed to determine if the country (class) in which the selected match (event) takes place exists in the tree (step 625). If the country corresponding to the selected match does not exist, it is created (step 630).

In a following step, a fourth test is carried out to determine whether the competition (group) to which the selected match (event) belongs exists in the tree (step 635). If the competition corresponding to the selected match does not exist, it is created (step 640). It is noted here that according to particular embodiments a match may count in different competitions. In such a case and as illustrated with a dotted arrow, steps 635 and 640 are repeated as many times as there are different competitions to which the match belongs (the index n is the index of the competitions, it varies from 1 to the number of competitions to which the match belongs).

In a next step, a fifth test is performed to determine if the teams (elements) competing in the selected match (event) exist in the tree (step 645). As illustrated with a dotted arrow, each team is processed one after the other (the index p is the index of the teams, it varies from 1 to the number of teams playing in the match). If the team considered does not exist, it is created (step 650).

Finally, the selected match is added to the event tree and the links between the match, the elements and the competition types are created (step 655). If they do not exist, the links between the types of competition, the country and/or the sport are also created. The date of the match is added (step 610).

The steps illustrated in Figure 6 are repeated for each event to be created and/or updated.

Figure 7 illustrates an example of steps for traversing an event tree and selecting events of interest to a user, as described with reference to Figure 3.

According to particular embodiments of the invention, all the events selected by a user in a considered past period (e.g. all the matches on which a player has placed a bet during the last three months) are used as reference events, in an event tree, to identify future events that may be of interest to the user. A score that may correspond to the number of identifications of the future event can be used as a relevance indicator.

Still according to particular embodiments and to limit the processing, the future events likely to interest the user can be pre-filtered to keep only those which are, a priori, the most relevant. For these purposes, a distance expressed in number of nodes between two events of an event tree can be used. All future events, located at a distance less than a given threshold, from a reference event are identified and a score, corresponding to the number of identifications of the future event, is used. The threshold can be predetermined, for example by the service provider, or can be determined dynamically, for example to control the number of events likely to interest an identified user.

As illustrated in FIG. 7, a first step aims to obtain a list of all the events on which a user i has initiated an action, denoted évén_sélect(i,q) , belonging to a selected category ( C_P ), in a past period considered (included between the present instant t ₀ and the instant t ₀ −Δ t _P ), where q is an index on the events (step 700). Considering the example presented in table 1 in the appendix, this list includes the events match_3 , match_C , match_B , match_5 , match_C , match_7 , match_11 and match_X (the event match_A is not part of the list because it does not belong not in the time period considered and the event match_k is not part of the list because its category was not selected).

A degree of relevance, or score, is determined for each of the future events (identified as being events to be proposed). According to one embodiment, this determination is made by exploring all the paths starting from each event of the list of events established in step 700 (reference event) and by identifying all the future events located on these paths (step 705 ). These events are denoted évén_prop(q,l), where l is index on the paths. The score of each identified future event is updated (step 710) in order to represent the number of identifications of this event (ie the number of paths leading to this one).

According to a particular embodiment, only the future events located at a distance less than a given threshold, for example a threshold equal to 4, events from the list of events established in step 700 are identified as being events at to propose. Thus the scores can be determined by exploring all the paths starting from the events considered (reference events) having a length less than or equal to the given threshold, by identifying all the future events located on these paths (step 705) and by updating the score of each identified future event (step 710) in order to represent the number of identifications of this event (i.e. the number of paths leading to this one).

A first test is then carried out to verify that all the paths starting from an event in the list of events established in step 700 have been analyzed (step 715).

When all the paths starting from an event in the list of events established in step 700 have been analyzed, a second test is carried out to verify that all the events in the list of events established in step 700 have been used as reference events to identify future events (step 720).

When all the events in the list of events established in step 700 have been used as reference events to identify future events and when all the paths starting from each of these events have been analyzed, the identified future events are ordered according to their score, from highest to lowest, and, if applicable, according to other criteria (step 725).

This list of events (or part of it if a filter is used) can then be transmitted to a user who can thus quickly select the events that interest him without having to perform a long and tedious navigation in all the selectable future events. As described previously, the filtering may consist, for example, in deleting the events which would be associated with stakes greater than a certain amount or which would be associated with an excessively high level of risk. It can also consist in deleting events that would be too close, in time or in space, to each other.

Figure 8 shows an example of a tree of events linked to sports betting linking teams (elements) to matches (events) themselves linked to types of competitions (groups) which, in turn, are linked to countries (classes) themselves linked to sports (categories).

Figure 9 shows an excerpt from the event tree of Figure 8.

As shown, a first M1 match pits teams E1 and E2 against each other. This match, belonging to the competition C1 , is played in the country P1 and concerns the sport S1 . Similarly, a second match M2 opposes the teams E3 and E4 . This match, also belonging to the competition C1 , is also played in the country P1 and also concerns the sport S1 . Similarly again, a third match M3 opposes the teams E4 and E5 . The competition, country and sport are the same as for M1 and M2 matches.

Figure 10, comprising Figures 10a to 10d, illustrates an example of a search for future events that can be offered to a user, in the event tree illustrated in Figure 9, from a past event already selected by the user.

By way of illustration, the past event already selected by the user is the match M3 (or match_3 appearing in table 1 given in the appendix, the oldest match in the past period considered) on which the user had committed a bet. This reference event is indicated in Figures 10a to 10d in solid bold lines. Still by way of illustration, it could be considered that the maximum distance between the past reference event and a future event to be proposed is for example equal to 2.

As illustrated in FIG. 10a, starting from event M3 , it is possible, due to the link between event M3 and element E5 , to establish a path between event M3 and element E5 . The length of this path is equal to 1. Since the E5 element is not an event, it cannot be selected to be offered to the user. The only link connecting the E5 element being the link between the M3 event and the E5 element, the path continues towards the M3 event, as illustrated with the arrows. The event M3 being an event, it can be offered to the user if, in addition to the past date on which the user placed a bet, a future date is associated with this event. By way of illustration, it is assumed here that a future date date_11 is associated with this event. The latter is then considered as a potential choice. It is stored with its date, its score and, preferably, the length of the path which made it possible to wait for it, for example in a table such as table 2 given in the appendix.

The event M3 can thus be selected in a similar way by three distinct paths (ie the paths M3-E5-M3 , M3-E4-M3 and M3-C1-M3 ). The associated score is therefore 3. The average path length is equal to 2 (2 + 2 + 2) / 3), as shown in Table 2.

As illustrated in figure 10b, starting from the event M3 , it is also possible, because of the link between the event M3 and the element E4 , to establish a path between the event M3 and the element E4 then between the latter and the event M2 which can thus be selected to be proposed to the user.

Similarly, as illustrated in Figure 10c, starting from the event M3 , it is also possible, due to the link between the event M3 and the category C1 , to establish a path between the event M3 and the category C1 then between the latter and the event M2 which can thus be selected again to be offered to the user.

The event M2 can thus be selected by two distinct paths (ie the paths M3-E4-M2 and M3-C1-M2 ). The associated score is therefore 2. The average path length is equal to 2 (2 + 2) / 2), as shown in Table 2.

Similarly, as illustrated in Figure 10d, starting from event M3 , it is also possible, due to the link between event M3 and category C1 , to establish a path between event M3 and category C1 then between the latter and the event M1 which can thus be selected to be proposed to the user.

The event M1 can thus be selected by a single path (ie the path M3-C1-M1 ). The associated score is therefore 1. The length of the path is equal to 2, as shown in Table 2.

The results obtained are then ordered so that they can be transmitted to the user so that the latter has quick access to the events likely to interest him. According to embodiments, the events are classified according to their score then, for equal scores, by the average length of the paths then, for equal average lengths, by increasing dates. Naturally, other criteria can be used.

According to the example illustrated in FIGS. 10a to 10d, the events offered to the user are, in order: M3 , M2 and M1 .

It is observed here that if a systematic search for the events to be offered to a user can be carried out on the basis of all the events that he has previously selected during the past period of time considered, by exploring all the possible paths, others approaches are possible. Thus, for example, it is possible to use only a subset of the events previously selected by the user during the past period of time considered, which can for example be determined randomly, and/or only explore certain paths, for example randomly.

Figure 11 illustrates an example of a device that can be used to implement, at least partially, an embodiment of the invention, in particular the steps described with reference to Figures 3 to 7.

The device 1100 is for example a server, a computer or a terminal.

The device 1100 preferably comprises a communication bus 1102 to which are connected:

a central processing unit or microprocessor 1104 (CPU, acronym for Central Processing Unit in English terminology);

a read only memory 1106 (ROM, acronym for Read Only Memory in Anglo-Saxon terminology) which can include the operating system and programs such as "Prog";

a random access memory or cache memory 1108 (RAM, acronym for Random Access Memory in English terminology) comprising registers suitable for recording variables and parameters created and modified during the execution of the aforementioned programs; And

a communication interface 1126 connected to a distributed communication network 1128, for example a wireless communication network and/or a local communication network, the interface being capable of transmitting and receiving data, in particular to and from a device of a user.

Optionally, the device 1100 can also have the following elements:

a hard disk 1120 which can include the aforementioned "Prog" programs and data processed or to be processed according to the invention;

a keyboard 1122 and a mouse 1124 or any other pointing device such as a light pen, a touch screen or a remote control enabling the user to interact with the programs according to the invention;

a reader 1110 of removable storage medium 1112 such as a memory card or a disc, for example a DVD disc; And

a graphics card 1114 connected to a screen 1116.

The communication bus allows communication and interoperability between the different elements included in the device 1100 or connected to it. The representation of the bus is not limiting and, in particular, the central unit is capable of communicating instructions to any element of the device 1100 directly or via another element of the device 1100.

The executable code of each program allowing the programmable device to implement the processes according to the invention can be stored, for example, in the hard disk 1120 or in ROM 1106.

According to a variant, the executable code of the programs could be received via the communication network 1128, via the interface 1126, to be stored in a manner identical to that described above.

More generally, the program or programs may be loaded into one of the storage means of the device 1100 before being executed.

The central unit 1104 will control and direct the execution of the instructions or portions of software code of the program or programs according to the invention, instructions which are stored in the hard disk 1120 or in the ROM 1106 or else in the other elements of aforementioned storage. When the power is turned on, the program or programs which are stored in a non-volatile memory, for example the hard disk 1120 or the ROM 1106, are transferred to the random access memory 1108 which then contains the executable code of the program or programs according to the invention, as well as registers for storing the variables and parameters necessary for the implementation of the invention.

Depending on the chosen embodiment, certain acts, actions, events or functions of each of the methods described in this document may be performed or occur in a different order from that in which they were described, or may be added, merged or else not be effected or not occur, as the case may be. Further, in some embodiments, certain acts, actions or events are performed or occur concurrently and not sequentially.

Although described through a certain number of detailed exemplary embodiments, the proposed method and the equipment for the implementation of the method include various variants, modifications and improvements which will appear obvious to those skilled in the art, it being understood that these various variants, modifications and improvements form part of the scope of the invention, as defined by the following claims. Additionally, various aspects and features described above may be implemented together, or separately, or substituted for each other, and all of the various combinations and sub-combinations of aspects and features are within the scope of the 'invention. Additionally, some systems and equipment described above may not incorporate all of the modules and functions described for the preferred embodiments.

Annex

event date element 1 element 2 band class category match_A date_1 team_4 team_2 competitive_1 class_1 cat_2 match_3 date_2 team_4 team_5 competitive_1 class_1 cat_1 match_C date_3 team_1 team_2 competitive_1 class_1 cat_1 match_B date_4 team_1 team_4 competitive_1 class_1 cat_1 match_K date_5 team_2 team_15 competitive_5 class_1 cat_7 match_5 date_6 team_2 team_5 competitive_1 class_1 cat_1 match_C date_7 team_1 team_2 competitive_1 class_3 cat_1 match_7 date_8 team_6 team_3 competitive_1 class_1 cat_1 match_11 date_9 team_5 team_1 competitive_2 class_2 cat_1 match_X date_10 team_3 team_9 competitive_1 class_1 cat_1

event date score medium length match_3 date_11 3 2 match_2 date_12 2 2 match_1 date_13 1 2

Claims (15)

Computer method for identifying and ordering a subset of at least one event from a set of events, each event being selectable by a user and being associated with at least one instant, at least one category and at least an element, the identification and the determination of an order of events being carried out according to at least a history of selections of events, the method comprising:

• obtaining at least one tree associating events of said set of events of which at least one instant belongs to a given period of time, the events of the at least one tree being associated with each other by their category and/ or their at least one element along paths comprising at least one node formed of at least one event, at least one category and/or at least one element;
• obtaining at least one selection history comprising at least one previously selected event;
• the identification, in the at least one tree obtained, of a subset of at least one event linked to at least one event belonging to the selection history via a path comprising at least one node formed of at at least one event, at least one category and/or at least one element, the instants of the identified events being subsequent to the events of the at least one selection history; And
• determining an order of the events of the subset of at least one event.

Method according to claim 1, according to which the determination of an order of the events of the subset is carried out according to an identification frequency of the events of the subset of at least one event in the at least one tree obtained. Method according to Claim 1 or Claim 2, according to which the identification, in the at least one tree obtained, of a subset of at least one event linked to at least one event belonging to the history of selections via a path is performed for all events in the selection history. Method according to any one of Claims 1 to 3, according to which the identification, in the at least one tree obtained, of a subset of at least one event linked to at least one event belonging to the history of selections via a path is carried out for all the paths resulting from at least one event belonging to the history of selections or for all the paths resulting from at least one event belonging to the history of selections comprising a number of nodes lower than a threshold given. Method according to any one of Claims 1 to 4, further comprising an update of the at least one tree obtained, the update comprising the deletion of at least one instant associated with an event, the at least one deleted instant being prior to a first instant, or the addition of at least one instant associated with an event, the at least one added instant being prior to a second instant subsequent to the first instant. Method according to any one of Claims 1 to 5, further comprising the selection of at least one category, the events of the subset of at least one event being linked to the at least one category selected by a path not comprising intermediate node formed by a category. Method according to Claim 6, in which the at least one category is selected according to the distribution of the categories of the events of the selection history. Method according to claim 7, according to which the at least one category is selected if the proportion of the events of the selection history linked to the at least one category is greater than a given threshold. Method according to Claim 7 or Claim 8, according to which the at least one category is selected randomly if the proportion of the events of the selection history linked to the at least one category is below a given threshold. A method according to any of claims 6 to 9 wherein the at least one category is selected from user preferences determined from the event selection history. Method according to any one of claims 1 to 10, in which a plurality of events of the set of events belong to at least one group, events of the at least one tree being associated with each other by their category , their at least one group and their at least one element along paths comprising at least one node formed of at least one event, at least one category, at least one group or at least one element. Method according to Claim 11, in which a plurality of groups belong to at least one class, the events of the at least one tree being associated with each other by their category, their groups, the at least one class and their at least one element along paths comprising at least one node formed of at least one event, at least one category, at least one group, at least one class or at least one element. Method according to any one of claims 1 to 12 further comprising a step of filtering the subset of at least one event. Computer program comprising instructions for implementing each of the steps of the method according to one of Claims 1 to 13, when this program is executed by a processor. Device comprising a processing unit configured to execute each of the steps of the method according to one of Claims 1 to 13.