Controlling a graphical user interface using a presentation of messages based upon predicted interactions with the messages

1. A method, comprising:

receiving a plurality of messages, associated with a messaging account, comprising a first message and a second message;

tracking interactions with the plurality of messages to generate a first set of message interactions for the first message and a second set of message interactions for the second message;

analyzing the plurality of messages to identify a first set of attributes for the first message and a second set of attributes for the second message;

generating an expected action model, customized for the messaging account, based upon a combination of at least three expectation models determined to be applicable to the messaging account including:

(i) a user-specific expectation model generated for the messaging account using the first set of message interactions, the second set of message interactions, the first set of attributes and the second set of attributes;

(ii) a company-specific subset expectation model generated for a subset of user accounts associated with a company based upon a subset of message interactions associated with the company, wherein a user of the messaging account is associated with the company, wherein the subset of message interactions comprise message interactions of a second user of a second messaging account based upon the second user being associated with the company but wherein the subset of message interactions do not comprise message interactions of a third user of a third messaging account based upon the third user not being associated with the company; and

(iii) a global expectation model generated for a plurality of user accounts based upon a set of message interactions associated with at least some user accounts of the plurality of user accounts, wherein the set of message interactions comprise the message interactions of the second user of the second messaging account and the message interactions of the third user of the third messaging account of the plurality of user accounts not included in the subset of message interactions used to generate the company-specific subset expectation model;

analyzing a set of messages, comprising a third message and a fourth message, associated with the messaging account based upon the expected action model that was generated based upon the user-specific expectation model, the company-specific subset expectation model and the global expectation model;

based upon the analyzing the set of messages, predicting one or more interactions, associated with one or more contexts, corresponding to a plurality of potential presentations of the set of messages comprising the third message and the fourth message, wherein the one or more interactions comprise (i) one or more first predicted interactions corresponding to a first potential presentation of the set of messages comprising the third message and the fourth message in a first context of a client device accessing the messaging account and (ii) one or more second predicted interactions corresponding to a second potential presentation of the set of messages comprising the third message and the fourth message in the first context of the client device accessing the messaging account;

based upon a current context of the client device accessing the messaging account and an evaluation of both the first potential presentation of the set of messages in the first context and the second potential presentation of the set of messages in the first context, selecting a presentation from the plurality of potential presentations, of the set of messages, comprising the first potential presentation of the set of messages and the second potential presentation of the set of messages, wherein the selecting the presentation is based upon a determination that the presentation is associated with at least one of a higher level of interaction or a higher proportion of messages interacted with than one or more other presentations in the plurality of potential presentations;

responsive to selecting the presentation from the plurality of potential presentations:

generating, based upon at least one of one or more message notification times or one or more message delivery times associated with the selected presentation, a feedback message comprising instructions to modify one or more rules associated with future message delivery times; and

transmitting the feedback message to a sender of at least one of the third message or the first message; and

controlling a graphical user interface of the client device using the presentation.

2. The method of claim 1 , wherein:

the plurality of potential presentations correspond to one or more potential ordered arrangements of the set of messages;

the presentation corresponds to an ordered arrangement, of the set of messages, associating each message of the set of messages with a position; and

the controlling the graphical user interface comprises:

presenting a fifth message of the set of messages in a first position in accordance with the ordered arrangement; and

presenting a sixth message of the set of messages in a second position below the first position in accordance with the ordered arrangement.

3. The method of claim 1 , wherein the controlling the graphical user interface using the presentation comprises:

displaying, in the graphical user interface, a list of four or more selectable messages; and

responsive to determining that one selectable message of the selectable messages is associated with a higher relevance than three or more other selectable messages of the other selectable messages, visually distinguishing the one selectable message from the three or more other selectable messages by modifying at least one of a font, a size or formatting associated with the one selectable message.

4. The method of claim 1 , wherein:

the plurality of potential presentations correspond to one or more potential notification settings for the set of messages;

the presentation corresponds to a notification setting, for the set of messages, associating each message of the set of messages with a notification time; and

the controlling the graphical user interface comprises:

presenting a notification for a fifth message of the set of messages at a first notification time in accordance with the notification setting; and

presenting a notification for a sixth message of the set of messages at a second notification time in accordance with the notification setting.

5. The method of claim 1 , wherein:

the predicting comprises predicting one or more interactions with a first bundle of messages of the set of messages by a first time threshold;

the selecting is performed responsive to determining that the first bundle of messages have not been interacted with by the first time threshold; and

the controlling the graphical user interface comprises:

presenting a notification for the first bundle of messages.

6. The method of claim 1 , wherein:

the plurality of potential presentations correspond to one or more potential delivery settings for the set of messages;

the presentation corresponds to a delivery setting, for the set of messages, associating each message of the set of messages with a delivery time; and

the method comprises:

delaying delivery of the third message of the set of messages until a first delivery time in accordance with the delivery setting; and

delaying delivery of the fourth message of the set of messages until a second delivery time in accordance with the delivery setting.

7. The method of claim 6 , wherein the third message is delivered to a mailbox of the messaging account at the first delivery time.

8. The method of claim 7 , wherein concurrently with the delaying delivery of the third message, the third message is delivered to a secondary mailbox of the messaging account prior to the first delivery time.

9. The method of claim 1 , comprising:

analyzing the set of messages based upon the expected action model to predict one or more other interactions, associated with one or more other contexts, corresponding to two or more potential presentations of the set of messages, wherein the one or more other interactions comprise (i) one or more third predicted interactions corresponding to a third potential presentation of the set of messages in a second context of the client device accessing the messaging account and (ii) one or more fourth predicted interactions corresponding to a fourth potential presentation of the set of messages in the second context of the client device accessing the messaging account;

based upon an updated context of the client device accessing the messaging account and an evaluation of the two or more potential presentations of the set of messages, selecting an updated presentation from the two or more potential presentations, wherein the selecting the updated presentation is based upon a determination that the updated presentation is associated with at least one of a higher level of interaction or a higher proportion of messages interacted with than one or more other presentations in the two or more potential presentations; and

controlling the graphical user interface of the client device using the updated presentation.

10. The method of claim 1 , wherein:

the interactions comprise at least one of opening of a message, a replying to a message or a forwarding of a message; and

attributes of at least one of the first set of attributes or the second set of attributes comprise at least one of a sender of a message, a subject of a message, a body of a message, a time of an interaction with a message, a time of receipt of a message or a domain associated with a message.

11. A computing device, comprising:

a processor; and

memory comprising processor-executable instructions that when executed by the processor cause performance of operations, the operations comprising:

receiving a plurality of messages, associated with a messaging account, comprising a first message and a second message;

at least one of:

tracking interactions with the plurality of messages to generate a first set of message interactions for the first message and a second set of message interactions for the second message; or

analyzing the plurality of messages to identify a first set of attributes for the first message and a second set of attributes for the second message;

generating an expected action model, customized for the messaging account, based upon a combination of at least three expectation models determined to be applicable to the messaging account including:

(i) a user-specific expectation model generated for the messaging account using at least one of:

the first set of message interactions and the second set of message interactions; or

the first set of attributes and the second set of attributes;

(ii) a company-specific subset expectation model generated for a subset of user accounts associated with the subset of user accounts based upon a subset of message interactions associated with the subset of user accounts, wherein a user of the messaging account is associated with the subset of user accounts, wherein the subset of message interactions comprise message interactions of a second user of a second messaging account based upon the second user being associated with the subset of user accounts but wherein the subset of message interactions do not comprise message interactions of a third user of a third messaging account based upon the third user not being associated with the subset of user accounts; and

(iii) a global expectation model generated fora plurality of user accounts based upon a set of message interactions associated with at least some user accounts of the plurality of user accounts, wherein the set of message interactions comprise the message interactions of the second user of the second messaging account and the message interactions of the third user of the third messaging account of the plurality of user accounts not included in the subset of message interactions used to generate the company-specific subset expectation model;

analyzing a set of messages, comprising a third message and a fourth message, associated with the messaging account based upon the expected action model that was generated based upon the user-specific expectation model, the company-specific subset expectation model and the global expectation model;

based upon the analyzing the set of messages, predicting one or more interactions, associated with one or more contexts, corresponding to a plurality of potential presentations of the set of messages comprising the third message and the fourth message, wherein the one or more interactions comprise (i) one or more first predicted interactions corresponding to a first potential presentation of the set of messages comprising the third message and the fourth message in a first context of a client device accessing the messaging account and (ii) one or more second predicted interactions corresponding to a second potential presentation of the set of messages comprising the third message and the fourth message in the first context of the client device accessing the messaging account;

based upon a current context of the client device accessing the messaging account and an evaluation of both the first potential presentation of the set of messages in the first context and the second potential presentation of the set of messages in the first context, selecting a presentation from the plurality of potential presentations, of the set of messages, comprising the first potential presentation of the set of messages and the second potential presentation of the set of messages; and

responsive to selecting the presentation from the plurality of potential presentations:

generating, based upon at least one of one or more message notification times or one or more message delivery times associated with the selected presentation, a feedback message comprising instructions to modify one or more rules associated with future message delivery times; and

transmitting the feedback message to a sender of at least one of the third message or the first message.

12. The computing device of claim 11 , wherein:

the plurality of potential presentations correspond to one or more potential ordered arrangements of the set of messages;

the presentation corresponds to an ordered arrangement, of the set of messages, associating each message of the set of messages with a position; and

the operations comprise:

presenting a fifth message of the set of messages in a first position in accordance with the ordered arrangement; and

presenting a sixth message of the set of messages in a second position below the first position in accordance with the ordered arrangement.

13. The computing device of claim 11 , wherein:

the plurality of potential presentations correspond to one or more potential notification settings for the set of messages;

the presentation corresponds to a notification setting, for the set of messages, associating each message of the set of messages with a notification time; and

the operations comprise:

presenting a notification for a fifth message of the set of messages at a first notification time in accordance with the notification setting; and

presenting a notification for a sixth message of the set of messages at a second notification time in accordance with the notification setting.

14. The computing device of claim 11 , wherein:

the predicting comprises predicting one or more interactions with a first bundle of messages of the set of messages by a first time threshold;

the selecting is performed responsive to determining that the first bundle of messages have not been interacted with by the first time threshold; and

the operations comprise:

presenting a notification for the first bundle of messages.

15. The computing device of claim 11 , wherein:

the plurality of potential presentations correspond to one or more potential delivery settings for the set of messages;

the presentation corresponds to a delivery setting, for the set of messages, associating each message of the set of messages with a delivery time; and

the operations comprise:

delaying delivery of the third message of the set of messages until a first delivery time in accordance with the delivery setting; and

delaying delivery of the fourth message of the set of messages until a second delivery time in accordance with the delivery setting.

16. A non-transitory machine readable medium having stored thereon processor-executable instructions that when executed cause performance of operations, the operations comprising:

accessing an expected action model, customized for a messaging account, indicative of associations between one or more message interactions and one or more message attributes, wherein the expected action model is based upon a combination of at least two expectation models determined to be applicable to the messaging account including at least two of:

(i) a user-specific expectation model generated for the messaging account using the one or more message interactions and the one or more message attributes;

(ii) a company-specific subset expectation model generated for a subset of user accounts based upon a subset of message interactions associated with the subset of user accounts, wherein a user of the messaging account is associated with the subset of user accounts, wherein the subset of message interactions comprise message interactions of a second user of a second messaging account based upon the second user being associated with the subset of user accounts but wherein the subset of message interactions do not comprise message interactions of a third user of a third messaging account based upon the third user not being associated with the subset of user accounts; or

(iii) a global expectation model generated for a plurality of user accounts based upon a set of message interactions associated with at least some user accounts of the plurality of user accounts, wherein the set of message interactions comprise the message interactions of the second user of the second messaging account and the message interactions of the third user of the third messaging account of the plurality of user accounts not included in the subset of message interactions used to generate the company-specific subset expectation model:

analyzing a set of messages, comprising a first message and a second message, associated with the messaging account based upon the expected action model that was generated based upon at least two of the user-specific expectation model, the company-specific subset expectation model or the global expectation model;

based upon the analyzing the set of messages, predicting one or more interactions corresponding to a plurality of potential presentations of the set of messages comprising the first message and the second message, wherein the one or more interactions comprise (i) one or more first predicted interactions corresponding to a first potential presentation of the set of messages comprising the first message and the second message in a first context of a client device accessing the messaging account and (ii) one or more second predicted interactions corresponding to a second potential presentation of the set of messages comprising the first message and the second message in the first context of the client device accessing the messaging account;

based upon an evaluation of both the first potential presentation of the set of messages and the second potential presentation of the set of messages, selecting a presentation from the plurality of potential presentations, of the set of messages, comprising the first potential presentation of the set of messages in the first context and the second potential presentation of the set of messages in the first context; and

responsive to selecting the presentation from the plurality of potential presentations:

generating, based upon at least one of one or more message notification times or one or more message delivery times associated with the selected presentation, a feedback message comprising instructions to modify one or more rules associated with future message delivery times; and

transmitting the feedback message to a sender of at least one of the second message or the first message.

17. The non-transitory machine readable medium of claim 16 , wherein:

the plurality of potential presentations correspond to one or more potential ordered arrangements of the set of messages;

the presentation corresponds to an ordered arrangement, of the set of messages, associating each message of the set of messages with a position; and

the operations comprise:

presenting a third message of the set of messages in a first position in accordance with the ordered arrangement; and

presenting a fourth message of the set of messages in a second position below the first position in accordance with the ordered arrangement.

18. The non-transitory machine readable medium of claim 16 , wherein:

the plurality of potential presentations correspond to one or more potential notification settings for the set of messages;

the presentation corresponds to a notification setting, for the set of messages, associating each message of the set of messages with a notification time; and

the operations comprise:

presenting a notification for a third message of the set of messages at a first notification time in accordance with the notification setting; and

presenting a notification for a fourth message of the set of messages at a second notification time in accordance with the notification setting.

19. The non-transitory machine readable medium of claim 16 , wherein:

the predicting comprises predicting one or more interactions with a first bundle of messages of the set of messages by a first time threshold;

the selecting is performed responsive to determining that the first bundle of messages have not been interacted with by the first time threshold; and

the operations comprise:

presenting a notification for the first bundle of messages.

20. The non-transitory machine readable medium of claim 16 , wherein:

the plurality of potential presentations correspond to one or more potential delivery settings for the set of messages;

the presentation corresponds to a delivery setting, for the set of messages, associating each message of the set of messages with a delivery time; and

the operations comprise:

delaying delivery of a third message of the set of messages until a first delivery time in accordance with the delivery setting; and

delaying delivery of a fourth message of the set of messages until a second delivery time in accordance with the delivery setting.