Abstract: Embodiments are directed to managing data visualizations. A primary visualization that is associated with a data model may be provided such that the primary visualization may be displayed in a display panel. Insight items may be generated based on the primary visualization and the data model such that the insight items may correspond to one or more visualizations that may share one or more portions of the data model and such that the insight items may be displayed in an insight panel. In response to an insight item being selected from the insight panel, additional actions may be performed, including: generating a visualization based on the insight item that is displayed in the display panel instead of the primary visualization; and generating a scratch item that includes a thumbnail view of the primary visualization such that the thumbnail view is displayed in a scratch panel.

Abstract: A method generates data analysis templates. A computing device receives, via a graphical user interface, a natural language input, which includes multiple terms. The terms include one or more data fields and/or data values from a data source. The device designates the natural language input as trigger text for a data analysis template. The trigger text includes the multiple terms. A user defines one or more rules for a subset of the terms, each rule specifying respective criteria to match a respective term in the subset. A user specifies multiple data visualizations corresponding to the trigger text. Each of the data visualizations (i) uses a respective one or more data fields, from the data source, corresponding to terms from the trigger text and (ii) has a respective data visualization type. The device generates the data analysis template according to the trigger text, the user definition, and the user specification.

Abstract: Embodiments are directed to managing data visualizations. Candidate data fields from a data source may be determined based on a search expression. The candidate data fields may be displayed in the model panel. A working data model may be generated based on a portion of the candidate data fields such that the portion of the candidate data fields may be included in the working data model. Visualizations may be determined based on recommendation models and the working data model such that a portion of the visualizations may be determined based on shared data fields that may be included in the working data model and the visualizations. A working visualization may be determined based on a visualization listed in the display panel and the working data model such that data fields included in the working data model may be associated with the working visualization.

Abstract: Embodiments are directed to intent driven dashboard recommendations. A plurality of collection specifications that declare visualization declarations may be provided such that each collection specification is associated with an author intent. And attributes-of-interest selected from a data source may be provided. A plurality of collections may be generated based on the plurality of collection specifications, the attributes-of-interest, or data from the data source such that each collection may include visualizations that may be based on the visualization declarations. A preference score may be generated for each collection based on the author intent, visualizations in each collection, or the attributes-of-interest. The plurality of collections may be classified based on the preference scores associated with the classified collections such that the classified collections may be displayed based on the preference scores.

Abstract: Embodiments are directed to managing data using a network computers. A hierarchy model that includes edges that represent dependency relationships between fields may be generated. In response to a query for determining values of metadata associated with an anchor field performing further actions, including: traversing the hierarchy model upwards from the anchor field based on the query, and in response to visiting field nodes in the hierarchy model, collecting the values of the metadata that correspond to the visited field node and terminating the traversal based on a type of the query; in response to visiting a calculation node, terminating the traversal of the hierarchy model associated with the visited calculation node; and in response to visiting a flow node that is dependent on two or more other nodes, terminating the traversal of the hierarchy model. The collected metadata may be provided in a response to the query.

Abstract: A method uses natural language processing for visual analysis of a dataset. The method includes receiving a first natural language (NL) input directed to a data source, from a first client. The method also includes parsing the first NL input into tokens based on a grammar and the data source. The method also includes generating and outputting an intermediate NL response, to a second client, based on the tokens and output modality of the second client. In response to receiving a user input to provide missing information: the method also includes generating an input query based on the user input; and querying the data source using the input query, to obtain a result set. The method also includes generating and outputting, to the second client, a first NL output and a snapshot of a data visualization, based on the result set and the output modality of the second client.

Abstract: A computing device generates and displays a data visualization in a data visualization user interface. The data visualization includes a plurality of visual data marks representing data from a data source. The computing device detects a first user input to select a visual data mark of the plurality of visual data marks. In response to detecting the first user input, the computing device identifies a first aggregated data value, for the visual data mark, corresponding to a first data field in the data source. The computing device retrieves a first disaggregated set of data rows corresponding to the first aggregated data value, from a first table, in the data source, containing the first data field. The computing device displays, in the data visualization user interface, a plurality of data rows from the first disaggregated set of data rows.

Abstract: A method visualizes object models for data sources. The device displays, in an object model visualization region, a first visualization of a tree of data object icons, each data object icon representing a logical combination of one or more tables. The device also displays recommendations for one or more data sources to add objects to the tree. While displaying the first visualization in the object model visualization region, the device detects a first input on a first data object icon of the tree of data object icons. In response to detecting the first input on the first data object icon, the device displays a second visualization of the tree of the data object icons in a first portion of the object model visualization region and displays a third visualization of information related to the first data object icon in a second portion of the object model visualization region.

Abstract: A computer device receives a user input to specify a natural language command. In response to receiving the user input, the computer device generates a semantic interpretation for the natural language command using a trained word similarity model, based on semantic annotations for a published data source. The computer device queries the published data source based on the semantic interpretation, thereby retrieving a dataset. The computer device generates and displays a data visualization based on the retrieved dataset.

Abstract: A computing device receives user selection of a data source. In accordance with the user selection, the device generates one or more initial natural language utterances according to metrics of data fields in the data source and/or previous user interaction with the data source. Each of the initial natural language utterances corresponds to a respective suggestion to guide visual analysis of the data source. The device displays the initial utterances in a graphical user interface. The device receives user selection of a first initial utterance of the initial utterances. In response to the user selection, the device generates and displays a first data visualization in accordance with data fields and/or analytical operations specified in the first initial utterance. The device also generates updated natural language utterances in accordance with the first initial utterance and the first data visualization, and displays the updated utterances with the first data visualization.

Abstract: A method for generating schemas to compress storage of denormalized data includes receiving user selection of an initial data table and constructing a snowflake schema to decompose the initial data table into a plurality of tables. Tables in the plurality of tables that are related to one another are connected by foreign keys. The snowflake schema is constructed according to a cost model that balances the cost of storing the plurality of tables and the cost of storing the foreign keys. The method also includes transforming data in the initial data table into the plurality of tables according to the snowflake schema.

Abstract: A computing device receives, in a graphical user interface, a first natural language query. The computing device parses the first natural language query, thereby identifying one or more keywords in the query. The computing device identifies one or more data sources that are relevant to the query. The computing device also identifies one or more data fields and/or data values from the data sources in the query. The computing device compares the keywords to respective trigger text for each of a plurality of data analysis templates, thereby selecting a first data analysis template. The first data analysis template includes a plurality of predefined data visualizations. The computing device generates a dashboard that includes the data visualizations using the identified data fields and/or data values, and displays the dashboard in the graphical user interface.

Abstract: Embodiments are directed to presenting visualizations over a network. A presentation specification. An overlay may be generated that includes visualizations of presentation objects based on the presentation specification. A video stream of a view of a presenter captured by a presenter station may be provided. A composited video stream may be generated based on the overlay and the video stream. The composited video stream may be provided to a video conference service that distributes the composited video stream to the presenter station and to audience stations. Movements of the presenter may be tracked based on the video stream captured by the presenter station. Gestures may be determined based on the tracked movements. In response to a determined gesture being associated with an interactive feature, activating the interactive feature to update the overlay such that the updated overlay may be included in a recomposited video stream.

Abstract: Embodiments are directed to managing data. A data model that includes data type nodes and relationship edges may be provided. Other data types and other data type relationships may be provided and included in the data model. If a portion of the nodes in the data model may be downstream of leaf nodes in the graphlet: the data model may be traversed to visit the downstream nodes; shortcut edges may be generated to each downstream node associated with shortcut nodes. If a second portion of the nodes in the data model may be upstream of the leaf nodes: the data model may be traversed upwards from the leaf nodes; other shortcut edges may be generated to each node visited in the upwards traversal associated with shortcut nodes.

Abstract: A method selects a federation engine to process a query. The system receives a query that specifies retrieval of data from multiple data sources at distinct physical locations. The system generates multiple federated query plans. Each federated query plan specifies execution of the query using a respective data source as the federation engine. For each federated query plan, the system forms a feature vector including two or more of: (1) a remote cost estimate for executing query operators on data sources distinct from the respective data source; (2) a local cost estimate for executing query operators on the respective data source; and (3) a data transfer cost estimate for transferring partial results from external data sources to the respective data source. The system predicts runtime for each federated query plan, selects a preferred federated query plan using the predicted runtimes, and the preferred federated query plan specifies the federation engine.

Abstract: A computing device receives user selection of a data source and a lens that specifies supplemental metadata for the data source. The device receives user input to specify a natural language command directed to the data source. In response to the user input, the device determines one or more data fields and/or data values from the data source for interpreting the natural language command in accordance with the metadata from the lens. The device translates the natural language command into one or more executable database queries referencing data fields and/or data values specified in the natural language command. The device executes the one or more database queries to retrieve data from the data source. The device also generates and displays a data visualization using the retrieved data.

Abstract: A method is performed at a computing device having a display, processors, and memory. A user inputs a natural language expression, directed to a selected data source, in a first region of a graphical user interface. The device parses the natural language expression and determines that the natural language expression specifies a data table from the data source either (i) directly by identifying the data table specified in the natural language expression or (ii) indirectly by identifying a data value specified in the natural language expression. The data value is a data value of a data field within the data table. In response, the device generates and displays, in the graphical user interface, a count of data rows of the data table, and a rendering of data rows and data columns of the data table.

Abstract: Embodiments are directed to generating metrics based on visualizations. A dashboard that may be associated with source visualizations display a current value of metrics from source visualization models. A classifier may automatically use characteristics from the source visualizations to determine metrics for the source visualization. The source visualization models may be sample to provide values of the metrics across time, at a sampling rate determined by a metric profile. The sampled values may be stored with time values in a metric data store such that the time values may correspond to when the values sampled from the visualization. Metric visualizations may be generated based on the values and the time values such that the metric visualizations display previously sampled values of the metrics.

Abstract: A method for generating data visualizations includes receiving user selection of a data source, and receiving user input to define a dynamic set according to: (i) user selection of visual marks in a visualization region and/or (ii) user specification of a parameter corresponding to data fields in the data source. The method also includes receiving user input to define a calculation to compare data values for a data field from the data source to data values in the dynamic set. The method further includes identifying a set of rows from the data source whose data values are included in the dynamic set. The method also includes receiving user input to place a data field from the data source in a shelf region, and displaying a data visualization that includes a plurality of visual marks corresponding to data values, of the data field, in the set of rows.

Abstract: A computer receives a first user input linking a first output of a first task to a second task. The first task is scheduled for execution at a specific time. The computer receives a second user input selecting a third task from a list of additional tasks that are downstream from the second task. In response to the second user input, the computer links the third task to a second output of the second task. At the specific time, the computer automatically executes the first task and updates the first output of the first task. After updating the first output of the first task, the computer automatically executes the second task that is linked to the first output of the first task, including updating the second output of the second task. After updating the second output of the second task, the computer automatically executes the third task.