Method and device for managing a list of actions to be performed by a terminal user

Abstract

The invention relates to a method for managing a list of actions and / or non-actions to be performed by a user of a terminal, said method being implemented by a management device and characterized in that it comprises the following steps: Adding, to said list, at least one action and / or at least one non-action determined according to the result of a corpus analysis; determining the execution of said at least one action and / or said at least one non-action according to the result of an analysis of the activity of said terminal over a predetermined period of time; and, according to the determination, deleting said at least one action and / or said at least one non-action from said list. Figure 1

Description

Title of the invention: Method and device for managing a list of actions to be carried out by a user of a terminal

[0001] 1. Field of the invention

[0002] The invention relates to the field of telecommunications and more particularly concerns decision-making and / or project management support services executed on a personal computer.

[0003] 2. Prior Art

[0004] There are many tools for creating, assigning, and tracking tasks / actions within a company. Post-it notes, for example, are simple and practical tools for managing actions. They are often used to note down tasks, reminders, or ideas, and can be stuck on boards, walls, or calendars for easy viewing. However, post-it notes have some limitations. They can be easily lost or damaged, which can lead to the loss of important information. In addition, the information written on post-its is by definition localized and visible, which can lead to sharing and confidentiality issues. In addition, when the actions to be managed become numerous or complex, post-it management can become inefficient and impractical.In such cases, using digital tools or project management software such as Microsoft Project®, Trello® or lira® may be preferable for more systematic organization and monitoring. Indeed, these tools offer features such as centralizing information, visualizing deadlines, online collaboration and reminders to ensure more efficient and organized management of actions.

[0005] It should be noted that the choice of tools depends on the specific needs of the company, its size, its sector of activity and its budget. Some companies may also opt for customized or integrated solutions to enable its employees to manage their actions more efficiently.

[0006] It is clear that feeding these tools with tasks / actions and their associated management remains the responsibility of the user. The user must manually create a list of tasks and indicate the tasks as completed when they are completed.

[0007] 3. Statement of the invention

[0008] The invention improves the state of the art and proposes for this purpose a method for managing a list of actions and / or non-actions to be carried out by a user of an electronic terminal, said method being implemented by a management device and characterized in that it comprises the following steps: - addition, in the said list, of at least one action and / or at least one non-action determined based on the result of an analysis of a corpus; - determination of the performance of said at least one action and / or said at least one non-action based on the result of an analysis of the activity of said terminal over a predetermined period of time;

[0009] and depending on the determination - deletion of said at least one action and / or said at least one non-action from said list.

[0010] Advantageously, this method makes it possible to automatically add an action / task to a list of actions / tasks of a user of an electronic terminal. To do this, the method analyzes a corpus in order to determine / extract one or more actions (or non-actions) to be carried out and then adds them to the user's list of actions. This analysis phase can, for example, rely on learning techniques known from the state of the art such as machine learning.

[0011] The method then analyzes the activity of the user and / or the terminal in order to evaluate whether the action(s) (and / or non-action(s)) have been carried out. Concretely, the method scans a set of events internal and / or external to the management device which make it possible to determine whether an action (respectively a non-action) has been completed.

[0012] When the result of the analysis indicates that an action is performed (positive determination), the method deletes the action from the user's action list.

[0013] A corpus is understood to mean a finite set of textual data (e.g. words, texts, terms of interest) chosen as the basis for a study / analysis.

[0014] According to a particular embodiment of the invention, a method as described above is characterized in that the corpus is obtained from said terminal and / or from a device capable of communicating with said terminal.

[0015] This embodiment allows, for example, obtaining the corpus from: - a RAM or ROM of the user's terminal; - a peripheral (keyboard, camera, microphone, mouse, etc.) integrated or connected to the user's terminal; - a database / file system stored / hosted on the user's terminal and / or on a server located in a communications network; - a content distribution server (in English streaming); - OCR of a file; - etc.

[0016] When all or part of the corpus is obtained from an optical sensor type device (for example a camera), an optical character recognition step can be previously applied to one or more images captured by the camera in order to to obtain a text corresponding to all or part of the corpus analyzed by the process. Alternatively or cumulatively, a step of transcription of a sequence in sign language captured by the camera can be carried out in order to obtain a text corresponding to all or part of the corpus analyzed by the process.

[0017] Similarly, when all or part of the corpus is obtained from a content distribution server (audio and / or video and / or text type) an optical character recognition step can be previously applied to one or more images received from the content server in order to obtain a text corresponding to all or part of the corpus analyzed by the method.

[0018] Also, when all or part of the corpus is obtained from a microphone-type device, an automatic speech recognition step can be previously applied to one or more audio sequences captured by the microphone in order to obtain a text corresponding to all or part of the corpus analyzed by the method.

[0019] All or part of the corpus may also correspond to all or part of one or more electronic text-type files (Word®, Excel®, OneNote®, email, instant message, etc.).

[0020] All or part of the corpus may also correspond to text obtained via an optical character recognition process carried out on one or more images / screenshots (recording of an image displayed on a screen / display device in the form of an electronic graphic file).

[0021] (R3) According to a particular embodiment of the invention, a method such as described above is characterized in that the determination of the completion of said at least one action corresponds to the detection of all or part of at least one sequence of events or of at least one event representative of the completion of said at least one action.

[0022] This embodiment makes it possible to determine that one or more actions in the user's action list have been performed during a predetermined period of time. Concretely, the action(s) are obtained by the method as a result of the analysis of the corpus. For example, when the corpus includes an instant message sent by a third party to the user such as: "please call me as soon as possible", the method detects via a semantic analysis of the sentence that an action / task relating to the call request must be added to the user's action list.

[0023] The determination of the performance of the action can then correspond to the detection of a sequence of events capable of allowing the establishment of the call to the third party such as: - launching of a telephony software by the user; - composition of the third party's number; - triggering the call.

[0024] Note that the completion / completion of an action / task may be determined when all of the events in a sequence of events have been detected. Alternatively, the completion of an action may be determined when a subset (e.g., 80%) of the events in the sequence of events have been detected.

[0025] Also note that the time period can correspond to minutes, hours, days, weeks or years.

[0026] According to a particular embodiment of the invention, a method as described above is characterized in that the determination of the completion of said at least one non-action corresponds to the absence of detection of all or part of at least one sequence of events or of at least one event representative of the completion of at least one action antinomic to said at least one non-action.

[0027] This embodiment makes it possible to determine that one or more non-actions from the user's list of actions have been complied with during a predetermined period of time. Concretely, the non-action(s) (user request(s) for inaction) are obtained by the method as a result of the analysis of the corpus.

[0028] For example, the method can extract from the corpus a sentence such as: "please do not reply to Alice's email attached". In this case, after a semantic analysis of the extracted sentence, the method adds to the user's list of actions an action / task relating to the absence of a response, i.e. the express request for no response, to the identified Alice's email. Determining whether the non-action has been carried out can then correspond to detecting the absence of an email sent to Alice with the same subject / object during the predetermined period of time.

[0029] Also note that the time period can correspond to minutes, hours, days, weeks or years.

[0030] According to a particular embodiment of the invention, a method as described above is characterized in that said at least one event or said at least one sequence of events comprises at least one event obtained via a step of processing a textual representation of said at least one action or said at least one non-action.

[0031] This embodiment makes it possible to determine, as a function of textual data corresponding to the textual representation / transcription of an action (or non-action) obtained by the method as a result of the analysis of the corpus, the event(s) representative of the completion of the action (or non-action).

[0032] For example, when the corpus comprises a text transcription of a voice message from Alice obtained via an automatic speech recognition process and the transcription comprises a sentence such as; "please call me as soon as possible quickly”, the process, after a semantic analysis of the sentence, adds to the user's list of actions an action / task relating to the call request formulated by Alice.

[0033] The method then processes this sentence in order to determine the event(s) capable of determining the completion of the action. In this case, the method can associate the term "call" with the user's use of a telephony software but also the term "me" with the identity of the person who left the voice message (i.e. Alice). Thus, the method considers that when the user uses his telephony software and triggers a call to Alice which is successful (in other words, Alice took the call), the action / task relating to the call request made by Alice is completed.

[0034] Another example, when the transcription includes a sentence such as; "please contact me as soon as possible", the method adds to the user's list of actions an action / task relating to the contact request by Alice. The method then processes this sentence in order to determine the event(s) capable of determining the completion of the action. In this case, the method can associate the term "contact" with the use by the user of software allowing the establishment of contact such as telephony software, instant messaging, emails, etc. Thus, the method considers that when the user uses his telephony software and triggers a call to Alice which is successful (in other words, Alice has taken the call), or when the user sends an email or an instant message to Alice the action / task relating to the contact request formulated by Alice is completed.

[0035] According to a particular embodiment of the invention, a method as described above is characterized in that said at least one event or all or part of said at least one sequence of events comprises at least one event obtained via a human-machine interface.

[0036] This embodiment allows the user to indicate to the method the event(s) representative of the completion of an action (or of a non-action).

[0037] According to a particular embodiment of the invention, a method as described above is characterized in that said at least one event or all or part of said at least one sequence of events comprises at least one event obtained via automatic learning.

[0038] This embodiment makes it possible to determine, based on automatic learning (in English “machine learning”) carried out by the method, the event(s) representative of the completion of an action (or of a non-action).

[0039] Concretely, when the method detects in the corpus an unknown type of task / action, that is to say an action that the method cannot associate with a known embodiment, in other words with one or more events representative of the completion of the action, the method triggers a learning phase in order to determine the event(s) representative of the completion of the action (or of the non-action). This learning phase can be carried out automatically via an artificial intelligence module which can exploit data obtained from software probes capable of observing on the user's terminal: - user activity (mouse movement, clicks, position, gestures, etc.); - the type of activity being carried out by the user (call, instant messaging, emails, etc.); - the active application windows running and their position on a display device (background position, foreground position, etc.). - running processes; - etc.

[0040] According to a particular embodiment of the invention, a method as described above is characterized in that the deletion step is conditioned by the result of a validation request returned to the user via a human-machine interface.

[0041] This embodiment allows the user to explicitly validate that an action has been performed and that this action can be deleted from the user's list of actions.

[0042] Note that the restitution of the request and / or the validation of the deletion can be done vocally and / or graphically.

[0043] According to a particular embodiment of the invention, a method as described above is characterized in that, when the step of determining the performance of said at least one action and / or said at least one non-action is negative, a notification is sent to said terminal of said user.

[0044] This embodiment allows, for the actions, to remind the user of the tasks / actions not carried out during the predetermined period of time. The user can then, if he wishes, delete the action(s) from the list of actions, if they are obsolete, or keep the action in the list of actions. In this case, the time period is reset.

[0045] Alternatively or cumulatively, the notification may correspond to a request for consultation, by the user, of the user's list of actions.

[0046] In the case of non-action, this embodiment makes it possible to alert the user to the fact that the non-action (request for inaction) has not been respected by the user. The user can then take the necessary steps to remedy this.

[0047] The invention also relates to a device for managing a list of actions and / or non-actions to be carried out by a user of an electronic terminal, said device management being characterized in that it includes: - a module for adding, to said list, at least one action and / or at least one non-action determined according to the result of an analysis of a corpus; - a module for determining the performance of said at least one action and / or said at least one non-action based on the result of an analysis of the activity of said terminal over a predetermined period of time; - a module for deleting said at least one action and / or said at least one non-action from said list.

[0048] The term module can correspond to a software component as well as to a hardware component or a set of hardware and software components, a software component itself corresponding to one or more computer programs or sub-programs or more generally to any element of a program capable of implementing a function or a set of functions as described for the modules concerned. In the same way, a hardware component corresponds to any element of a hardware assembly capable of implementing a function or a set of functions for the module concerned (integrated circuit, smart card, memory card, etc.).

[0049] The invention also relates to a computer program comprising instructions for implementing the above method according to any of the particular embodiments described above, when said program is executed by a processor. The method can be implemented in various ways, in particular in hard-wired form or in software form. This program can use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0050] The invention also relates to a recording medium or information medium readable by a computer, and comprising instructions of a computer program as mentioned above. The recording media mentioned above can be any entity or device capable of storing the program. For example, the medium can comprise a storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a hard disk. Furthermore, the recording media can correspond to a transmissible medium such as an electrical or optical signal, which can be conveyed via an electrical or optical cable, by radio or by other means. The programs according to the invention can in particular be downloaded from a network such as the Internet.

[0051] Alternatively, the recording media may correspond to a circuit integrated circuit in which the program is incorporated, the circuit being adapted to carry out or to be used in carrying out the method in question.

[0052] This management device and this computer program have characteristics and advantages similar to those described previously in relation to the management method.

[0053] 4. List of figures

[0054] Other characteristics and advantages of the invention will appear more clearly on reading the following description of particular embodiments, given as simple illustrative and non-limiting examples, and the appended drawings, among which:

[0055] [Fig-1] [Fig. 1] illustrates an example of an environment for implementing the method management according to a particular embodiment of the invention;

[0056] [Fig.2] [Fig.2] represents the hardware architecture of a management device according to a particular embodiment of the invention;

[0057] [Fig.3] [Fig.3] represents in the form of an organizational chart the main stages of the management method according to a particular embodiment of the invention.

[0058] 5. Description of an embodiment of the invention

[0059] [Fig.l] illustrates an example of an implementation environment of the invention according to a particular embodiment of the invention. The implementation environment comprises a terminal 101 of a user UT capable of executing one or more computer applications such as a personal computer, a server, an internet gateway, a mobile terminal or a connected object. The terminal 101 is connected to a display device 103 such as a screen, a touch screen or a video projector making it possible to display the graphical interfaces of the executed applications. The connection between the display device 103 and the terminal 101 can be wired, for example via an HDMI cable, or wireless, for example via a Wi-Fi® or Bluetooth® connection.The terminal 101 is also capable of interpreting data from an input / output device connected or integrated into the terminal 101 such as a pointing system (for example a mouse), a keyboard, a camera, a microphone (data which may be a voice command) allowing a user to interact with the applications and their graphical interfaces displayed on the display device 103.

[0060] According to a particular embodiment of the invention, the terminal 101 and the display device 103 are one and the same terminal (for example the mobile terminal 102).

[0061] The terminal 101 is capable of communicating with another terminal, for example, via point-to-point communication (WiFi®, Bluetooth®, etc.) or via a local LAN communication network. This network can be managed by a residential / professional / mobile PAS gateway connected to a data communication network of a R_EXT operator which hosts one or more SA application servers.

[0062] We will now describe an example of hardware architecture of a obtaining device according to a particular embodiment of the invention in connection with [Fig.2].

[0063] The device DISP has the classic architecture of a computer, and notably comprises a memory MEM, a processing unit UT1, equipped for example with a processor PROC, and controlled by the computer program PG stored in memory MEM. The computer program PG comprises instructions for implementing the steps of the management method as described later in support of [Fig. 3], when the program is executed by the processor PROC.

[0064] At initialization, the code instructions of the computer program PG are, for example, loaded into a memory before being executed by the processor PROC. The processor PROC of the processing unit UT1 notably implements the steps of the management method according to any one of the particular embodiments described in relation to [Fig. 3] and according to the instructions of the computer program PG.

[0065] The DISP device comprises an ADD module capable of adding one or more actions / non-actions to a user's list of actions based on the result of an analysis of a corpus.

[0066] The DISP device further comprises a DETER module capable of determining whether an action or non-action from a list of actions of a user has been carried out during a given period of time. An action can be judged as carried out when the result of the analysis of the activity of the terminal and / or the user indicates it. Concretely, particular events indicating the end of a task are sought within the data representative of the activity of the terminal and / or the user.

[0067] The DISP device also comprises a SUPPR module capable of deleting a task / action (alternatively or cumulatively a non-action) from a list of tasks / actions of a user. The deletion can occur when the result of the analysis of the activity of the terminal and / or the user indicates that the task (respectively the non-action) has been carried out.

[0068] According to a particular embodiment, the DISP device comprises a module (not shown) for rendering a notification. The rendering module may be a human-machine interface such as a screen (touch or not), a video projector, a television, a speaker, or any other device capable of fulfilling the same objective.

[0069] According to a particular embodiment, the DISP device comprises a first human-machine interface module (not shown) capable of obtaining the description / transcription / designation of an event or a sequence of events representative of the completion of an action (or of a non-action). The human-machine interface machine can be a microphone, a keyboard, a camera, etc.

[0070] According to a particular embodiment, the DISP device comprises a second human-machine interface module (not shown) capable of obtaining a response to a request to validate the deletion of an action from a list of user actions. The human-machine interface may be a microphone, a keyboard, a camera, etc.

[0071] According to a particular embodiment, the first and second human-machine interface modules can be one and the same module.

[0072] [Fig. 3] illustrates steps of the management method according to a particular embodiment of the invention. In this exemplary implementation, the management method is executed by the terminal 101 which may correspond to a personal computer of the user UT.

[0073] According to a particular embodiment of the invention, the management method is executed on a server located in the R_EXT network.

[0074] According to a particular embodiment of the invention, the management method is executed in a distributed manner between the terminal 101 and one or more terminals (for example the terminal 102, the PAS gateway or the SA server).

[0075] In step 300 the method obtains / accesses a corpus. The corpus is for example obtained from a RAM or ROM of the terminal 101 and / or a peripheral integrated or connected to the terminal 101 and / or a database stored / hosted by the terminal 101 and / or from a content distribution server (“streaming” server).

[0076] When all or part of the corpus is obtained from a content distribution server (audio and / or video and / or text type) an optical character recognition step can be previously applied to one or more images received from the content server in order to obtain a text corresponding to all or part of the corpus analyzed by the method.

[0077] Also, when the corpus is obtained from a microphone-type device, an automatic speech recognition step (transcription in the form of a text of the words / sentences spoken) can be previously applied to one or more audio sequences captured by the microphone in order to obtain a text corresponding to all or part of the corpus analyzed by the method.

[0078] When all or part of the corpus is obtained from a camera-type device, an optical character recognition step may be previously applied to one or more images captured by the camera in order to obtain a text corresponding to all or part of the corpus analyzed by the method. Alternatively or cumulatively, a step of transcription of a sequence in sign language captured by the camera may be carried out in order to obtain a text corresponding to all or part of the corpus analyzed by the process.

[0079] Note that the corpus may also correspond to all or part of one or more electronic text-type files (Word®, Excel®, OneNote®, email, instant message, etc.).

[0080] It should also be noted that the corpus can be completed / amended directly by the user via a human-machine interface of the terminal 101 such as a keyboard, a mouse, a touch screen, a microphone, a camera, etc. For example, the user may wish to add to the corpus acronyms and / or terms specific to his profession and / or his organization.

[0081] Once the corpus is obtained, the method analyzes it in order to determine / extract one or more actions (or non-actions) to be performed by the user UT such as questions requiring a response from the user and therefore an action on their part. This analysis phase can rely on learning techniques known from the state of the art such as machine learning. In a particular embodiment, the method can perform a semantic analysis of the corpus in order to identify within the corpus the action verbs and the questions posed to the user UT (for example via punctuation and more particularly the question mark). The method then isolates the associated sentences, each sentence (or part of a sentence) being able to correspond to an action to be performed by the user UT.For example, when the corpus includes the sentence "please call me as soon as possible", the method detects the action verb "call", extracts the sentence from the corpus and then adds an action related to the sentence in the user UT's list of actions. The action can correspond to the extracted sentence or to the result of processing the extracted sentence. For example, the method can take into account the entire message (title / subject, content, etc.) and / or messages previously exchanged between the sender and the user UT in order to determine the context and the name of the sender of the message. Thus, when the extracted sentence corresponds to all or part of an email sent by Alice, the method transforms the extracted sentence by substituting the "m'" into "Alice". The method can also add context elements such as the subject of the email (for example: "planning of project X") or the priority associated with the email and therefore with the action.The priority level can, for example, be detected / estimated via the presence in the email of a prefix / word such as "urgent" or a red exclamation mark in the Outlook® interface indicating high importance of the email. The priority can also be estimated based on the wording of the sentence.

[0082] The action entered by the method in the user UT's action list can then correspond to: "Call Alice about the planning meeting for project X".

[0083] Note that an action may also correspond to a request for inaction on the part of the user UT (non-action within the meaning of the invention). For example, the method can, after analysis, extract from the corpus: "please do not contact this customer directly". In this case, a task can be added to the user's list of actions corresponding to a request for inaction, i.e. not to contact the identified customer, for example, via an identifier and / or a name obtained from the application that fed the corpus with the sentence "please do not contact this customer directly" (for example, an email application, process management, etc.).

[0084] During step 301 the method determines whether an action from the list of actions of the user UT has been performed. To do this the method analyzes / observes the activity of the terminal 101 and / or of the user UT. Concretely, the method scans a set of events internal and / or external to the terminal 101 which make it possible to determine whether an action (respectively a non-action) is completed.

[0085] An event may correspond to a voice command, to one or more words spoken by the user UT captured by the microphone of the terminal 101, to one or more signs from the user UT captured by a camera of the terminal 101, to an event managed by the operating system of the terminal 101 (mouse click, pressing a key on a keyboard and / or on a touch screen, etc.), to an event managed by an application executed by the terminal 101 (launch / closing of an application, rendering of a particular human-machine interface, etc.), to a sequence of characters / words detected within a text, to a message received from a third-party terminal, etc.

[0086] Note that the message received from a third-party terminal (for example, a connected watch or a health sensor) may correspond to a message relating to the user's physical activity (geolocation data, user movement, cardiac activity, etc.). Thus, when the action extracted from the corpus corresponds to "please contact me as soon as possible", the method can obtain from a third-party terminal geolocation information indicating that the user UT has physically moved to the office of the person who wished to be contacted. The geolocation information may, for example, correspond to an identifier of a Beacon type beacon installed in the office of the person to be contacted. Obviously, this geolocation information may be associated with information relating to a presence indicator in the office of the person to be contacted.

[0087] The event or sequence of events representative of the completion of an action scrutinized by the method can be obtained as a result of a step of processing a textual representation of the action (semantic analysis of the textual representation of the action and / or the associated elements / words / sequence of words). Concretely, the method processes and transforms each action into a sequence of keywords and / or a portion of a sentence which can correspond to: - an action verb; - an entity representing the one who must perform the action; - an entity representing the person requesting the action; - context element(s).

[0088] For example: No. Action(s) / question(s) detected in the corpus Verb action Who must do the action For whom context element(s) 1 fill in the project monitoring table X Fill in Me NA (Not Applicable) project monitoring table X 2 Call Alice Call Me Alice NA 3 Send document Z to Paul Send Me Paul Document Z 4 drafting of the project Y report Draft Me NA (Not Applicable) Document “project Y report”.

[0089] In the case of the action “send document Z to Paul”, the method can deduce that the determination of the performance of the action corresponds to the detection of a sequence of events capable of allowing the sending of a message to Paul such as: - launching of a messaging software (instant, email, etc.); - drafting a message containing document Z as an attachment; - sending the message to Paul.

[0090] Alternatively or cumulatively, the event or sequence of events representative of the completion of an action can be entered / declared by the user UT via a human-machine interface of the terminal 101 (microphone, keyboard, mouse, camera, etc.), obtained from a database or obtained via machine learning carried out by the method.

[0091] For the latter case, when the method detects a new type of task, it can launch a learning / acquisition phase in order to determine the event or sequence of events to be examined representative of the performance of the action. The learning / acquisition phase can be triggered: - manually, when the user starts and / or stops the learning phase himself, for example, via a human-machine interface of the terminal 101, i.e. the recording of the process of carrying out the action and the associated events; - and / or automatically via the search by the process of one or more words keys (name of processes executed on the terminal 101, file name, date and title of an email, etc.) in data obtained from software probes for observing the activity of the terminal 101 and / or the user UT. These probes make it possible, for example, to know at a given time T information concerning a given application window (name, identifier, software, associated computer process, title, etc.).

[0092] A probe may, for example, provide the method with data with the following content:

[0093] "1": {

[0094] "isForeground": false,

[0095] "isVisible": true,

[0096] "islconic": false,

[0097] "title": "SNCF, timetable, train, info",

[0098] "processID": "11824",

[0099] "processName": "msedge",

[0100] "startTime": "2022-10-12T08:52:35.0044929+02:00",

[0101] },

[0102] In this example, it is possible for the method to automatically detect an event representative of the start of a task corresponding to the purchase of a train ticket for Paris via the SNCF (French train operator) website. Indeed, the value associated with the “processName” corresponds to “msedge”, i.e. an identifier of an internet browser, and the value of the “title” of the internet page consulted by the browser corresponds to an online sales site, in this case that of the SNCF.

[0103] Thus, the method can detect in this data a first event corresponding to the opening of a browser and a second event corresponding to the consultation by the user UT of the website allowing the purchase of train tickets. These two events can be representative of the start of an execution of the task “take a train ticket for Paris”.

[0104] Once the start of the task / action has been detected, the method can once again rely on the software probes of the terminal 101 (keyboard probes, mouse, content of active windows, etc.) to record the event(s) enabling the task / action to be carried out. For example, the method can take screenshots of the internet page displayed on the screen 103 of the terminal 101 and then execute an optical character recognition process in order to obtain the text displayed to the user. The method then searches for keywords in the text obtained and checks, for example, that the words “destination” and “Paris” are present in the text. If this is the case, the method considers that the action “take a train ticket to Paris" is still in progress by user UT.

[0105] As for the event representing the completion of the action, this can be obtained by the method via: • a voice and / or graphical human-machine interface of the terminal 101 (for example in a response to an explicit request sent to the user UT); • the software probes executed by the terminal 101 allowing, for example, the detection of the closing of the active window, a particular application window with a confirmation message (in English a pop-up) of the successful completion of the action (the message is for example obtained via an optical character recognition method applied to a screenshot of the application window), etc.

[0106] Note that the event or sequence of events representative of the completion of the action can be recorded by the method in an internal and / or external memory of the terminal 101 (for example within a database / knowledge base) in association with the action. Thus, when the same action is requested again from the user UT, the method will be able, thanks to the wording of the action, to find in the knowledge base the event or sequence of events representative of the completion of the action. Obviously, when the event or sequence of events representative of the completion of the action changes, the knowledge base is updated (for example via the learning / acquisition method).

[0107] It should also be noted that the learning phase may be more suited to tasks / actions for which the use of business applications is necessary to carry out the action. Indeed, it is not uncommon for business applications to use particular jargon, making it more difficult, for example, for the process to interpret the texts displayed to the user and the associated events.

[0108] Obviously, the methods for determining the event or sequence of events representative of the performance of an action implemented during the learning phase can be the same for determining the performance of the action as such.

[0109] According to a particular embodiment of the invention, the completion of an action can be determined when a subset (for example 80%) of the events of the sequence of events representative of the completion of the action has been detected. This embodiment makes it possible to take into account variations in the performance of the action, for example following an update of the working environment or the practices of the user UT, etc. In such a case, the method can update in the knowledge base, the sequence of events representative of the completion of the action associated with the action.

[0110] According to a particular embodiment of the invention, when the action corresponds to a non-action, that is to say to a request for inaction on the part of the user UT, the determination of the realization of the non-action can then correspond to the detection of the absence of an event or of a sequence of events representative of the completion of an action antinomic to the non-action, that is to say of an action negated by the non-action. [YES] For example:

[0112] The non-action can be: "please do not call Alice", and the antinomic action: "call Alice".

[0113] In step 302, when the result of the analysis indicates that an action (or a non-action) has been carried out (positive determination), that is to say that the event or the sequence of events representative of the completion of the action has been detected by the method during a given period of time (respectively that the event or the sequence of events representative of the completion of the action antinomic to the non-action has been detected by the method), the method deletes the action from the list of actions of the user UT.

[0114] According to a particular embodiment of the invention, the time period is determined as a function of the action. For example, when the action entered by the method in the user UT's action list corresponds to: "Call Alice about the planning meeting for project X", the time period given for carrying out the action can be determined as a function of the date of the planning meeting for project X, for example obtained from the Outlook® software calendar.

[0115] Alternatively, the time period may be determined based on the nature of the action. For example, the method may arbitrarily assign a time period (time period within the meaning of the invention) of 5 days to the completion of a task corresponding to drafting a document and 2 days to the completion of a task corresponding to responding to an email.

[0116] Alternatively, the time period may be the same for all actions / tasks in the UT user's action list.

[0117] According to a particular embodiment of the invention, the step of deleting an action may be conditioned by the response to a validation request returned to the user UT. The restitution of the validation request may for example be done vocally and / or graphically by the terminal 101.

[0118] According to a particular embodiment of the invention, when the result of the analysis indicates that an action (or a non-action) has not been carried out / respected (negative determination) a notification of reminder or request for deletion of the action can be sent to the user UT.

Claims

Claims

1. Method for managing a list of actions and / or non-actions to be performed by a user of an electronic terminal, said method being implemented by a management device and characterized in that it comprises the following steps: - adding (300), in said list, at least one action and / or at least one non-action determined as a function of the result of an analysis of a corpus; - determining (301) the performance of said at least one action and / or said at least one non-action as a function of the result of an analysis of the activity of said terminal over a predetermined period of time; and as a function of the determination - deleting (302) said at least one action and / or said at least one non-action from said list.

2. Method according to claim 1 in which the corpus is obtained from said terminal and / or from a device capable of communicating with said terminal.

3. Method according to any one of the preceding claims in which the determination of the completion of said at least one action corresponds to the detection of all or part of at least one sequence of events or of at least one event representative of the completion of said at least one action.

4. Method according to any one of claims 1 to 2 in which the determination of the completion of said at least one non-action corresponds to the absence of detection of all or part of at least one sequence of events or of at least one event representative of the completion of at least one action antinomic to said at least one non-action.

5. Method according to any one of claims 3 to 4 wherein said at least one event or said at least one sequence of events comprises at least one event obtained via a step of processing a textual representation of said at least one action or said at least one non-action.

6. A method according to any one of claims 3 to 4 wherein said at least one event or all or part of said at least one sequence of events comprises at least one event obtained via a human-machine interface.

7. Method according to any one of claims 3 to 4 wherein said at least one event or all or part of said at least one sequence of events comprises at least one event obtained via machine learning.

8. Method according to any one of the preceding claims in which the deletion step is conditioned by the result of a validation request returned to the user via a human-machine interface.

9. Method according to any one of the preceding claims in which when the step of determining the performance of said at least one action and / or said at least one non-action is negative, a notification is sent to said terminal of said user.

10. Device for managing a list of actions and / or non-actions to be carried out by a user of an electronic terminal, said management device being characterized in that it comprises: - a module for adding (ADD), to said list, at least one action and / or at least one non-action determined according to the result of an analysis of a corpus; - a module for determining (DETER) the carrying out of said at least one action and / or said at least one non-action according to the result of an analysis of the activity of said terminal over a predetermined period of time; - a module for deleting (SUPPR) said at least one action and / or said at least one non-action from said list.

11. Computer or mobile terminal characterized in that it comprises a management device according to claim 10.

12. A computer program comprising instructions for implementing the method according to any one of claims 1 to 9, when the program is executed by a processor.

13. Information medium readable by a management device, and comprising instructions of a computer program according to claim 12.

Images