Abstract: A computer system obtains a data model that encodes a data source as a tree of logical tables. Each logical table includes respective data fields of the data source. For each data field in the logical tables, the system automatically assigns a semantic role to the data field based on a concept obtained from a concept graph, a data type of the data field, and data values of the data field. The system determines hierarchies between the data field and other data fields in the data source based on their respective semantic roles. The system validates the data field based on the assigned semantic role and the determined hierarchies. The system displays transformations to clean the data field based on the assigned semantic role. The system transforms data values for the data field according to a user input and updates the logical tables according to the transforming.

Abstract: A computer system obtains a data model that encodes a data source as a tree of logical tables. Each logical table includes respective data fields of the data source. For each data field in the logical tables, the system automatically assigns a semantic role to the data field based on a concept obtained from a concept graph. The system determines hierarchies between the data field and other data fields in the first data source based on a respective semantic role of the data field and semantic roles of the other data fields. The system validates the data field based on the assigned semantic role and the determined hierarchies. The system displays transformations to clean the data field based on the assigned semantic role. The system transforms data values for the data field according to a user input and updates the logical tables according to the transformed data values.

Abstract: A computing device has a display, one or more processors, and memory. The memory stores one or more programs configured for execution by the one or more processors. The computing device receives, via a first region of a graphical user interface, a first natural language input related to a data source. The first natural language input includes one or more terms. In response to receiving the first natural language input, computing device generates one or more interpretations for the first natural language input, including: determining a first term of the one or more terms that is excluded from the one or more interpretations, and visually de-emphasizing the first term in the first region.

Abstract: A computer system receives from a messaging system a first input commencing an analytic conversation between the computer system and an electronic device. The messaging system is communicatively connected with the electronic device and executes a messaging application on the electronic device. The computer system identifies a data source corresponding to the analytic conversation. The computing system receives from the messaging system a second input specifying a natural language command related to the data source. The computer system generates an interpretation corresponding to the natural language command and queries the data source according to the interpretation. The computer system generates a first data visualization using a first subset of data fields from the data source and sends a first representation of the first data visualization to the messaging system, for display on the electronic device.

Abstract: A computing device receives from a user a partial natural language input related to a data source. The computing device receives an additional keystroke corresponding to the partial natural language input. The partial natural language input and the additional keystroke comprise a character string. In response to the additional keystroke, the computing device generates one or more interpretations corresponding to entities in the data source. The computing device displays the interpretations. In some implementation, the character string comprises a sequence of terms, and the device displays the interpretations in a dropdown menu adjacent to the most recently entered term in the sequence. In some implementations, the dropdown menu includes a plurality of rows, each row displaying a respective data value and a respective data field corresponding to the respective data value. Some implementations display a statistical distribution of data values for a data field (displayed adjacent to the first interpretation).

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: Embodiments are directed to a display engine that provides a graphical user interface (GUI) that includes a graph panel and a visualization panel may be generated. The graph panel may receive graph related inputs or interactions. Graph information based on the graph related inputs or interactions and a data source may be generated. A graph model based on the graph information and the data source may be generated. Elements that correspond to objects in the graph model or relationships in the graph model may be displayed in the graph panel. The graph information and the graph model may be employed to execute a query that provides a data model based on results of the query. The data model may be employed to provide a visualization that is displayed in the display panel.

Abstract: A computer system receives from a messaging system a first input commencing an analytic conversation between the computer system and an electronic device. The messaging system is communicatively connected with the electronic device and executes a messaging application on the electronic device. The computer system identifies a data source corresponding to the analytic conversation. The computing system receives from the messaging system a second input specifying a natural language command related to the data source. The computer system generates an interpretation corresponding to the natural language command and queries the data source according to the interpretation. The computer system generates a first data visualization using a first subset of data fields from the data source and sends a first representation of the first data visualization to the messaging system, for display on the electronic device.

Abstract: A computing system receives from a messaging system a first input identifying a data source. The computing system receives from the messaging system a second input specifying a natural language command related to the data source. The system translates the second input into a set of parameters for building a first data visualization from the data source. The system calls one or more API functions provided by a data visualization server, and the computing system receives from the data visualization server a first data visualization that includes a first subset of data fields from the data source. The system generates a representation of the first data visualization and sends the representation to the messaging system, which in turn sends the representation to an electronic device for display on the device.

Abstract: A method generates a data visualization dashboard using a natural language interface. The method is performed at a computing device having a display, one or more processors, and memory. A user specifies a natural language input directed to a data source, using a first region of a graphical user interface. In response, the computing device parses the natural language input to identify one or more database entities from the data source. For each of the database entities, the computing device determines one or more respective analytical operations for the respective database entity. The device then generates a dashboard, which includes a plurality of data visualizations corresponding to distinct combinations of the analytical operations. The first dashboard is displayed in a second region of the graphical user interface, distinct from the first region.

Abstract: A method processes natural language commands (e.g., for data visualization). The method includes obtaining word embeddings (e.g., word vectors) for a set of words (or phrases) of a natural language. In some implementations, the set of words includes high-frequency or common words of the natural language. The method also includes training a word similarity model to identify similar words based on the word embeddings and a synonym database (a large lexical database, such as a thesaurus). The method also includes generating semantic annotations for a published data source using the trained word similarity model, based on the synonym database, and the word embeddings. The method also includes generating a semantic interpretation for a natural language command based on the semantic annotations for the published data source. The method also includes querying the published data source based on the sematic interpretation, thereby retrieving a dataset.

Abstract: A computer system obtains a data model that encodes a data source as a tree of logical tables. Each logical table includes respective data fields of the data source. For each data field in the logical tables, the system automatically assigns a semantic role to the data field based on a concept obtained from a concept graph. The system determines hierarchies between the data field and other data fields in the first data source based on a respective semantic role of the data field and semantic roles of the other data fields. The system validates the data field based on the assigned semantic role and the determined hierarchies. The system displays transformations to clean the data field based on the assigned semantic role. The system transforms data values for the data field according to a user input and updates the logical tables according to the transformed data values.

Abstract: A computing device displays a first data visualization that includes a first plurality of data marks. Each of the data marks corresponds to a respective distinct data value of a first data field from a dataset. In response to user selection of a first data mark that corresponds to a first data value of the first data field, the device displays a first data widget that includes one or more user-selectable affordances. In response to user selection of a first affordance of the affordances, the device displays a first drill down widget. The device receives user selection of a second data field from the dataset in the drill down widget. In response to the user selection, the device generates a second data visualization that includes a second plurality of data marks and displays the second data visualization.

Abstract: A computing device has a display, one or more processors, and memory. The memory stores one or more programs configured for execution by the one or more processors. The computing device receives, via a first region of a graphical user interface, a first natural language input related to a data source. The first natural language input includes one or more terms. In response to receiving the first natural language input, computing device generates one or more interpretations for the first natural language input, including: determining a first term of the one or more terms that is excluded from the one or more interpretations, and visually de-emphasizing the first term in the first region.

Abstract: A method prepares data for subsequent analysis. The method obtains a data model that encodes a data sources as a tree of logical tables. Each logical table has its own physical representation and includes logical fields. Each logical field corresponds to either a data field or a calculation that spans logical tables. The method associates each logical table in the data model with a corresponding concept in a concept graph. The concept graph embodies hierarchical inheritance of semantics. For each logical field, the method assigns a semantic role to the logical field based on a concept corresponding to a logical table. The method also validates the logical field based on its semantic role. The method also displays transformations to clean the logical field based on validating the logical field. The method transforms the logical field according to a user selection, and updates the logical table.

Abstract: A computing device has a display, one or more processors, and memory. The memory stores one or more programs configured for execution by the one or more processors. The computing device receives from a user a partial natural language input related to a data source. The partial natural language input includes a most recently entered first term. In response to receiving the first term, the computing device generates a first token that includes the first term. The computing device maps the first token to one or more analytical concepts in a lexicon of the data source. The computing device determines a first interpretation corresponding to the first token. The computing device also displays a first phrase corresponding to the first interpretation. The first interpretation includes the first term.

Abstract: A computing system receives from a messaging system a first input identifying a data source. The computing system receives from the messaging system a second input specifying a natural language command related to the data source. The system translates the second input into a set of parameters for building a first data visualization from the data source. The system calls one or more API functions provided by a data visualization server, and the computing system receives from the data visualization server a first data visualization that includes a first subset of data fields from the data source. The system generates a representation of the first data visualization and sends the representation to the messaging system, which in turn sends the representation to an electronic device for display on the device.

Abstract: A method processes natural language commands (e.g., for data visualization). The method includes obtaining word embeddings (e.g., word vectors) for a set of words (or phrases) of a natural language. In some implementations, the set of words includes high-frequency or common words of the natural language. The method also includes training a word similarity model to identify similar words based on the word embeddings and a synonym database (a large lexical database, such as a thesaurus). The method also includes generating semantic annotations for a published data source using the trained word similarity model, based on the synonym database, and the word embeddings. The method also includes generating a semantic interpretation for a natural language command based on the semantic annotations for the published data source. The method also includes querying the published data source based on the sematic interpretation, thereby retrieving a dataset.

Abstract: A computing device displays a first data visualization that includes a first plurality of data marks. Each of the data marks corresponds to a respective distinct data value of a first data field from a dataset. In response to user selection of a first data mark that corresponds to a first data value of the first data field, the device displays a first data widget that includes one or more user-selectable affordances. In response to user selection of a first affordance of the affordances, the device displays a first drill down widget. The device receives user selection of a second data field from the dataset in the drill down widget. In response to the user selection, the device generates a second data visualization that includes a second plurality of data marks and displays the second data visualization.

Abstract: A computing device displays a first data visualization based on a dataset retrieved from a database. The dataset includes a first data field. The first data visualization includes a first plurality of data marks. The device receives user selection of a first data mark that corresponds to a first data value of the first data field. In response to the user selection, the device displays a first data widget that includes one or more user-selectable affordances. A user selects a first user-selectable affordance. In response to the user selection, the device displays a first drill down widget. User input in the first drill down widget selects a second data field from the dataset. In response to the user input, the device generates a second data visualization that includes a second plurality of data marks and displays the second data visualization.