Recall service for productivity applications

A system comprising:

• at least one processor;
• and
• a memory operatively connected with the at least one processor storing computer-executable instructions that, when executed by the at least one processor, causes the at least one processor to execute a method that comprises:
• identifying an operational context associated with a user, wherein the identifying of the operational context comprises:
• detecting, through a productivity service, user access to an electronic document;
• identifying a content interaction that modifies content within the electronic document during at least one previous instance of the user access to the electronic document based on an identification of the electronic document;
• identifying one or more other productivity services that concurrently accessed digital content through the one or more other productivity services during the at least one previous instance of the user access to the electronic document;
• identifying digital artifacts that are each configured to recall, for a digital document that is associated with the one or more other productivity services, a displayed user state that is a state of display of the digital document showing user action with the digital content of the digital document during the content interaction within the electronic document;
• generating, using trained machine learning modeling configured to determine a relevance between the displayed user state of each of the digital artifacts and the content interaction within the electronic document, confidence scoring that ranks relevance between metadata associated with the displayed user state of each of the digital artifacts and metadata identifying user actions that modify the content of the electronic document during the at least one previous instance of user access to the electronic document;
• and surfacing a representation of one or more of the digital artifacts in a user interface based on a threshold evaluation of the confidence scoring.

A computer storage device storing computer-executable instructions that, when executed by at least one processor, causes the at least one processor to execute a method comprising:

• identifying an operational context associated with a user, wherein the identifying of the operational context comprises:
• detecting, through a productivity service, user access to an electronic document;
• identifying a content interaction that modifies content within the electronic document during at least one previous instance of the user access to the electronic document based on an identification of the electronic document;
• identifying one or more other productivity services that concurrently accessed digital content through the one or more other productivity services during the at least one previous instance of the user access to the electronic document;
• identifying digital artifacts that are each configured to recall, for a digital document that is associated with the one or more other productivity services, a displayed user state that is a state of display of the digital document showing user action with the digital content of the digital document during the content interaction within the electronic document;
• generating, using trained machine learning modeling configured to determine a relevance between the displayed user state of each of the digital artifacts and the content interaction within the electronic document, confidence scoring that ranks relevance between metadata associated with the displayed user state of each of the digital artifacts and metadata identifying user actions that modify the content of the electronic document during the at least one previous instance of user access to the electronic document;
• and surfacing a representation of one or more of the digital artifacts in a user interface based on a threshold evaluation of the confidence scoring.