Abstract: Content items are rearranged within a digital collaboration space (e.g., a digital whiteboard) based on category metadata determined for the content items using one or more learning models. The content items are added to the digital collaboration space by one or more users. For each of the content items, category metadata is determined using a learning model that processes information associated with the content item. A rearrangement of the content items is determined based on the category metadata determined for each of the content items. The content items, rearranged according to the rearrangement, are then output to a layer of the digital collaboration space. Other rearrangements of the content items can be determined based on non-category metadata associated with the content items, and the content items, rearranged according to those other rearrangements, can be output to other layers of the digital collaboration space.

Abstract: Content items are rearranged within a digital collaboration space (e.g., a digital whiteboard) based on category metadata determined for the content items using one or more learning models. The content items are added to the digital collaboration space by one or more users. For each of the content items, category metadata is determined using a learning model that processes information associated with the content item. A rearrangement of the content items is determined based on the category metadata determined for each of the content items. The content items, rearranged according to the rearrangement, are then output to a layer of the digital collaboration space. Other rearrangements of the content items can be determined based on non-category metadata associated with the content items, and the content items, rearranged according to those other rearrangements, can be output to other layers of the digital collaboration space.

Abstract: Content items are rearranged within a digital collaboration space (e.g., a digital whiteboard) based on category metadata determined for the content items using one or more learning models. The content items are added to the digital collaboration space by one or more users. For each of the content items, category metadata is determined using a learning model that processes information associated with the content item. A rearrangement of the content items is determined based on the category metadata determined for each of the content items. The content items, rearranged according to the rearrangement, are then output to a layer of the digital collaboration space. Other rearrangements of the content items can be determined based on non-category metadata associated with the content items, and the content items, rearranged according to those other rearrangements, can be output to other layers of the digital collaboration space.

Abstract: One example method for integrating software application content into video conferences includes receiving, by a video conferencing application executed by a client device from a video conference provider, one or more participant video streams, each participant video stream corresponding to a participant in a video conference; receiving, by the video conferencing application from a software application, software application content; receiving, by the video conferencing application from the software application, display layout information; generating, by the video conferencing application, a graphical representation of the video conference according to the display layout information, the graphical representation comprising at least a portion of the software application content and at least one of the one or more participant video streams; and displaying, by the video conferencing application, the graphical representation of the video conference.

Abstract: A combined search and graph interface is provided for facilitating investigative searches. In an embodiment, a set search results for a first search is concurrently displayed with a graph that represents the first search. User input is received that selects a first search result from the set of search results for the first search. In response, a node is added to the graph that represents the first search result and one or more child nodes that each correspond to a named entity extracted from a data source that corresponds to the first search result are added below the node in the graph. In another embodiment, a second search is initiated in response to receiving user input that selects a node from the graph. The graph and a second set of search results generated by the second search are concurrently displayed in the combined search and graph interface.