Calendar-aware resource retrieval

Abstract

Generally discussed herein are devices, systems, and methods for resource retrieval. A method may include determining that a calendar event is scheduled to occur in a specified period of time, responsive to the determination, extracting content of a calendar event on a calendar of the messaging interface, generating a list of resources accessible by the user and related to the extracted content of the calendar event, ranking the resources by a comparison of the extracted content of the calendar event and the content of resources of the list of resources, and causing respective summaries of a specified number of the respective resources with higher respective ranks to be output on the display.

1 . (canceled) 
     
     
         2 . A system comprising:
 a user interface that provides a user access to functionality of an electronic mail (email) application;   a processor; and   a memory device coupled to the processor, the memory device including a program stored thereon, wherein the program, when executed, causes the processor to perform operations comprising:   determining that a calendar event is scheduled to occur;   responsive to the determination, extracting or identifying features of a calendar event of the email application;   generating, by a query of a database, a list of resources accessible by the user and related to the calendar event;   ranking, by a machine learning (ML) model, the resources by a feature vector that indicates overlap between the extracted features of the calendar event and resource feature vectors representing content of respective resources of the list of resources; and   causing respective summaries of a specified number of the respective resources with higher respective ranks to be output on the user interface.   
     
     
         3 . The system of  claim 2 , wherein the operations further comprise:
 receiving, by the user interface, a first indication that the user has selected a summary of the respective summaries;   retrieving the resource associated with the selected summary; and   causing the resource to be output by the display.   
     
     
         4 . The system of  claim 2 , wherein the content of the calendar event includes a subject, one or more names of people attending or invited, a name of an organizer, a date and time, and text in a body, of the calendar event and the comparison of the feature vector and the calendar event includes at least two of: (a) content similarity of respective resources and one or more of the subject and the text in the body; (b) the names of people attending or invited and one or more names of people in the respective resources; (c) a temporal recency of the respective resources; and (d) a relationship of the respective resources to the calendar event in a resource-relation graph. 
     
     
         5 . The system of  claim 4 , wherein the resource feature vector indicates the amount of content in the resource relative to the one or more other resources of the resources and the operations further comprise determining the number of words in the body and title of the resource, combined, and converting the number of words to a quantile based on the amount of content in the resource relative to the one or more other resources. 
     
     
         6 . The system of  claim 4 , wherein the resource feature vector indicates whether the resource includes one or more attachments. 
     
     
         7 . The system of  claim 4 , wherein the resource feature vector indicates a content similarity between the resource and the content of the calendar event and the operations further comprise determining a similarity between content of the resource and the subject of the calendar event. 
     
     
         8 . The system of  claim 4 , wherein the resource feature vector indicates a content similarity between the resource and the content of the calendar event and the operations further comprise determining a similarity between content of the resource and the subject and body of the calendar event. 
     
     
         9 . The system of  claim 4 , wherein the resource feature vector indicates a person match between the name of one or more people attending, invited to, or organizer of the calendar event and the name of one or more people in the resource and the operations further comprise determining a Jaccard index or cosine similarity based on the name of one or more people attending, invited to, or organizer of the calendar event and the name of one or more people in the resource, wherein the name of one or more people in the resource includes one or more of an author or co-author, a sender, a receiver, a copy contact, a person that last modified the resource, or a person mentioned in a body of the resource. 
     
     
         10 . The system of  claim 4 , wherein the resource feature vector indicates whether the organizer is an author or co-author, a sender, a receiver, a copy contact, a person that last modified the resource, or a person mentioned in a body of the resource. 
     
     
         11 . The system of  claim 4 , wherein the resource feature vector indicates whether an attendee or invitee of the calendar event is an author or co-author, a sender, a receiver, a copy contact, a person that last modified the resource, or a person mentioned in a body of the resource. 
     
     
         12 . The system of  claim 2 , wherein:
 the resource feature vector includes data indicating three or more of: (a) an amount of content in the resource relative to one or more other resources of the resources, (b) a temporal recency of the resource relative to one or more other resources of the resources, (c) a content similarity between the resource and the content of the calendar event, and (d) a person match between the name of one or more people attending, invited to, or organizer of the calendar event and the name of one or more people in the resource.   
     
     
         13 . The system of  claim 12 , wherein the operations further comprise:
 accessing a resource-relation graph with resources and calendar events as nodes of the connected graph and respective labelled edges between nodes of the connected graph, wherein labels of the labelled edges indicate the relationship between the nodes on each end of an edge of the edges; and   wherein the resource feature vector indicates a number of edges from the calendar event and the respective resources in the resource-relation graph in a meta-path between the calendar event and the respective resources.   
     
     
         14 . The system of  claim 13 , wherein ranking the resources includes ranking the resources by a comparison of a relationship of the respective resources to the calendar event in the resource-relation graph includes assigning a score to respective labels and combining one or more scores, using the Mt model, to determine the rank. 
     
     
         15 . The system of  claim 14 , wherein the resources include one or more of an email, an attachment to the email, and one or more other documents accessible either stored locally or remotely to the user and accessible by the user, the one or more other documents including one or more documents authored by an entity attending the calendar event or documents authored by an organizer of the calendar event. 
     
     
         16 . At least one machine-readable storage medium including instructions for execution by processing circuitry to perform operations comprising:
 extracting content of a calendar event on an electronic mail (email) application, the content including a subject, one or more names of people attending or invited, a name of an organizer, a body, and a date and time of the calendar event;   identifying, by a query of a database, a list of resources accessible by the user and related to the extracted content of the calendar event;   ranking, by a machine learning (ML) model, the resources by a feature vector that indicates: (a) content similarity of respective resources and one or more of the subject and text in the body; (b) the names of people attending or invited and one or more names of people in the respective resources or metadata of the respective resources; (c) a temporal recency of the respective resources; and (d) a relationship of the respective resources to the calendar event in a resource-relation graph; and   causing respective summaries of a specified number of the resources with higher respective ranks to be displayed on a display.   
     
     
         17 . The at least one machine-readable storage medium of  claim 16 , wherein the operations further comprise:
 receiving, by a user interface, a first indication that the user has selected a summary of the respective summaries;   retrieving the resource associated with the selected summary; and   causing the retrieved resource to be displayed on the display.   
     
     
         18 . The at least one machine-readable storage medium of  claim 16 , wherein the operations further comprise:
 receiving, by a user interface, a second indication of one of: (a) the user has selected a software control to retrieve related items, (b) a specified date and time has elapsed, (c) a specified period of time has elapsed since a last update, and (d) a calendar event is scheduled to begin in a second specified period of time; and   wherein extracting the content of the calendar event occurs in response to receiving the second indication.   
     
     
         19 . The at least one machine-readable storage medium of  claim 16 , wherein:
 the feature vector includes data indicating three or more of: (a) an amount of content in the resource relative to one or more other resources of the resources, (b) a temporal recency of the resource relative to one or more other resources of the resources, (c) a content similarity between the resource and the content of the calendar event, and (d) a person match between the name of one or more people attending, invited to, or organizer of the calendar event and the name of one or more people in the resource.   
     
     
         20 . A method, performed by at least one processor of a computing system, the method comprising:
 extracting features of a calendar event on an electronic mail (email) application, the content of the calendar event including: a subject, one or more names of people attending or invited, a name of an organizer, a date and time, and text in a body, of the calendar event;   generating, by a query of a database, a list of resources accessible by the user and related to the extracted content of the calendar event;   ranking, by a machine learning (ML) technique, the resources by a feature vector that indicates overlap between the extracted features and at least two of: (a) content similarity of respective resources and one or more of the subject and the text in the body; (b) the names of people attending or invited and one or more names of people in the respective resources; (c) a temporal recency of the respective resources; and (d) a relationship of the respective resources to the calendar event in a resource-relation graph; and   causing respective summaries of a specified number of the resources with higher respective ranks to be displayed on the user interface.   
     
     
         21 . The method of  claim 20 , wherein:
 the feature vector includes data indicating one or more of: (a) an amount of content in the resource relative to one or more other resources of the resources, (b) a temporal recency of the resource relative to one or more other resources of the resources, (c) a content similarity between the resource and the content of the calendar event, and (d) a person match between the name of one or more people attending, invited to, or organizer of the calendar event and the name of one or more people in the resource.