Traitement en cours

Veuillez attendre...

Paramétrages

Paramétrages

Aller à Demande

1. WO2020018367 - PROCÉDÉS ET SYSTÈMES DE SUGGESTION D’ENTRÉE

Note: Texte fondé sur des processus automatiques de reconnaissance optique de caractères. Seule la version PDF a une valeur juridique

[ EN ]

WHAT IS CLAIMED IS:

1. A computer-implemented method comprising:

receiving, by one or more computing devices and from a first application executed by the one or more computing devices, data indicating information that has been one or more of presented by or input into the first application;

generating, by the one or more computing devices, based at least in part on the data indicating the information that has been one or more of presented by or input into the first application, and without receiving user input via the first application indicating a desire to be able to utilize data based at least in part on the information in a second application executed by the one or more computing devices, one or more suggested candidate inputs for the second application, the second application being different and distinct from the first application;

providing, by the one or more computing devices and in association with the second application, an interface comprising one or more options to select at least one suggested candidate input of the one or more suggested candidate inputs; and

responsive to receiving data indicating a selection of a particular suggested candidate input of the one or more suggested candidate inputs via the interface,

communicating, by the one or more computing devices and to the second application, data indicating the particular suggested candidate input.

2. The computer-implemented method of claim 1, wherein generating the one or more suggested candidate inputs comprises generating the one or more suggested candidate inputs based at least in part on data indicating information that has been one or more of presented by or input into the second application.

3. The computer-implemented method of claim 1, wherein receiving the data indicating the information comprises receiving the data indicating the information via a device accessibility application programming interface (API) of the one or more computing devices.

4. The computer-implemented method of claim 1, wherein receiving the data indicating the information comprises:

communicating, by the first application, the data indicating the information via an application programming interface (API) of the one or more computing devices configured

to manage a file-based application-indexing log stored on the one or more computing devices and comprising data indicating one or more application actions or events; and

retrieving the data indicating the information from the file-based application indexing log via the API.

5. The computer-implemented method of claim 1, comprising:

utilizing, by the one or more computing devices, a machine learning (ML) model to determine, for each suggested candidate input of the one or more suggested candidate inputs, a predicted likelihood that a user will select the suggested candidate input via the interface; and

determining, by the one or more computing devices, for each suggested candidate input of the at least one suggested candidate input, and based at least in part on the predicted likelihood that the user will select the suggested candidate input via the interface, to include an option in the interface for selecting the suggested candidate input.

6. The computer-implemented method of claim 5, comprising, for at least one suggested candidate input of the one or more suggested candidate inputs:

annotating, by the one or more computing devices, the suggested candidate input with metadata determined based at least in part on one or more of:

a portion of the information,

factual knowledge about the user, or

factual knowledge about a physical environment in which the one or more computing devices are located; and

determining, by the one or more computing devices and based at least in part on the metadata, the predicted likelihood that the user will select the suggested candidate input via the interface.

7. The computer-implemented method of claim 5, comprising, for at least one suggested candidate input of the one or more suggested candidate inputs:

generating, by the one or more computing devices, one or more numeric descriptions of one or more features of the suggested candidate input distinct from its content; and

determining, by the one or more computing devices and based at least in part on the one or more numeric descriptions, the predicted likelihood that the user will select the suggested candidate input via the interface.

8. The computer-implemented method of claim 7, wherein, for at least one suggested candidate input of the one or more suggested candidate inputs, the one or more features comprise one or more of how recently or how frequently the content of the suggested candidate input was presented by the one or more computing devices.

9. The computer-implemented method of claim 7, wherein, for at least one suggested candidate input of the one or more suggested candidate inputs, the one or more features comprise a type of societal entity associated with the suggested candidate input.

10. The computer-implemented method of claim 5, comprising determining, by the one or more computing devices and for at least one suggested candidate input of the one or more suggested candidate inputs, the predicted likelihood that the user will select the suggested candidate input based at least in part on data identifying one or more of the second application or a selected input field of the second application.

11. The computer-implemented method of claim 5, comprising updating, by the one or more computing devices, the ML model based at least in part on data generated by a plurality of different computing devices via a federated-learning paradigm.

12. The computer-implemented method of claim 5, comprising generating, by the one or more computing devices and for at least one suggested candidate input of the one or more suggested candidate inputs, a training record for the ML model based at least in part on one or more numeric descriptions of one or more features of the suggested candidate input distinct from its content.

13. The computer-implemented method of claim 5, comprising generating, by the one or more computing devices and for at least one suggested candidate input of the one or more suggested candidate inputs, a training record for the ML model based at least in part on a determination that the user manually input data corresponding to content of the suggested candidate input in lieu of selecting the suggested candidate input via the interface.

14. The computer-implemented method of claim 1, wherein:

providing the interface comprises generating a graphical user interface (GUI) comprising one or more elements corresponding to the one or more options to select the at least one suggested candidate input; and

the method comprises:

determining, by the one or more computing devices, a predicted likelihood that a user will select the particular suggested candidate input via the interface; and determining, by the one or more computing devices and based at least in part on the predicted likelihood, one or more of:

a size of an element, of the one or more elements,

corresponding to an option, of the one or more options, to select the particular suggested candidate input; or

a number of the one or more elements, corresponding to one or more of the one or more options to select one or more other suggested candidate inputs of the at least one suggested candidate input, to be presented alongside the element corresponding to the option to select the particular suggested candidate input.

15. A system comprising:

one or more processors; and

a memory storing instructions that when executed by the one or more processors cause the system to perform operations comprising:

receiving data indicating information that has been presented by a first application executed by the system;

receiving data indicating information that has been input into a second application executed by the system, the second application being different and distinct from the first application;

generating, based at least in part on the data indicating the information that has been presented by the first application and the data indicating the information that has been input into the second application, one or more suggested candidate inputs for the second application;

providing, in association with the second application, an interface comprising one or more options to select at least one suggested candidate input of the one or more suggested candidate inputs; and

responsive to receiving data indicating a selection of a particular suggested candidate input of the one or more suggested candidate inputs via the interface, communicating, to the second application, data indicating the particular suggested candidate input.

16. The system of claim 15, wherein the operations comprise:

utilizing a machine learning (ML) model to determine, for each suggested candidate input of the one or more suggested candidate inputs, a predicted likelihood that a user will select the suggested candidate input via the interface; and

determining, for each suggested candidate input of the at least one suggested candidate input and based at least in part on the predicted likelihood that the user will select the suggested candidate input via the interface, to include an option in the interface for selecting the suggested candidate input.

17. The system of claim 16, wherein the operations comprise, for each suggested candidate input of the one or more suggested candidate inputs:

generating one or more numeric descriptions of one or more features of the suggested candidate input distinct from its content; and

determining, based at least in part on the one or more numeric descriptions, the predicted likelihood that the user will select the suggested candidate input via the interface.

18. One or more non-transitory computer-readable media comprising instructions that when executed by one or more computing devices cause the one or more computing devices to perform operations comprising:

receiving data indicating information that has been one or more of presented by or input into an application executed by the one or more computing devices;

generating, based at least in part on the data indicating the information that has been one or more of presented by or input into the application, one or more suggested candidate inputs for the application including a suggested candidate input comprising

metadata determined based at least in part on an entity associated with the suggested candidate input; and

providing, in association with the application executed by the one or more computing devices, an interface comprising one or more options to select at least one suggested candidate input of the one or more suggested candidate inputs including the suggested candidate input comprising the metadata determined based at least in part on the entity associated with the suggested candidate input.

19. The one or more non-transitory computer-readable media of claim 18, wherein the suggested candidate input comprising the metadata comprises data configured to cause a different and distinct application to output content related to the information that has been one or more of presented by or input into the application executed by the one or more computing devices.

20. The one or more non-transitory computer-readable media of claim 18, wherein:

providing the interface comprises generating a graphical user interface (GUI) comprising one or more elements corresponding to the one or more options to select the at least one suggested candidate input; and

the operations comprise:

determining a predicted likelihood that a user will select the suggested candidate input comprising the metadata via the interface; and

determining, based at least in part on the predicted likelihood, one or more of:

a size of an element, of the one or more elements,

corresponding to an option, of the one or more options, to select the suggested candidate input comprising the metadata; or

a number of the one or more elements, corresponding to one or more of the one or more options to select one or more other suggested candidate inputs of the at least one suggested candidate input, to be presented alongside the element corresponding to the option to select the suggested candidate input comprising the metadata.