Abstract: The technology described herein protects the privacy of data stored in a knowledge graph (“graph”) by enforcing privacy policies when returning information in response to a query or other attempt to extract information from the graph and/or about the graph. In one aspect, unauthorized information is trimmed from the information output in response to a query. In other words, the privacy policy for a knowledge graph is enforced during the information extraction process. This is in contrast to other methods that attempt to enforce privacy policies at the information ingestion process or through some other service-level process after information is output from the graph. The privacy policies may be stored in a node of the graph.

Abstract: Automatically extracting and completing tasks associated with a communication is provided. In some examples, an application, such as a task management application, may generate nodes based on properties extracted from a communication, generate tasks based on the nodes, and construct a graph based on the nodes and the tasks. The task management application may also perform a prioritization of the tasks within the graph based on the properties of the nodes, order the tasks within the graph based on the prioritization of the tasks, and provide a task management interface to present the tasks within the graph. The task management application may also transmit the tasks within the graph to a task extraction engine of the task management application in response to a modification input executed on the graph.

Abstract: Automatically extracting and completing tasks associated with a communication is provided. In some examples, an application, such as a task management application, may generate nodes based on properties extracted from a communication, generate tasks based on the nodes, and construct a graph based on the nodes and the tasks. The task management application may also perform a prioritization of the tasks within the graph based on the properties of the nodes, order the tasks within the graph based on the prioritization of the tasks, and provide a task management interface to present the tasks within the graph. The task management application may also transmit the tasks within the graph to a task extraction engine of the task management application in response to a modification input executed on the graph.