ThoughtSpot, Inc.

Automatic target visualization adaptation includes obtaining a natural language string expressing a request for data, obtaining large language model generated structured query language data expressing the natural language string, obtaining, by transforming the large language model generated structured query language data, first resolved request data, obtaining target-parameters data for a target visualization type, obtaining target-parameters deficit data, obtaining a co-occurring parameter, obtaining, by transforming a combination of the first resolved request data and the co-occurring parameter, an automatically generated data query, obtaining from a database, results data generated by execution of the automatically generated data query by the database, and outputting data for presenting the results data in accordance with the target visualization type.

Automatic data modeling in includes identifying an analytical object in response to first data expressing usage intent, generating an analytical model generation data query for the analytical object, obtaining a trained analytical model generated in accordance with the analytical model generation query and trained using results data obtained in accordance with the analytical object, generating a resolved request representing second data expressing usage intent and indicating a request for results data obtained using the trained analytical model, generating an analytical model results data query for obtaining the results data in accordance with the trained analytical model and the analytical object, and outputting data for presenting a visualization of the results data obtained by executing the analytical model results data query, wherein a first portion of the results data corresponds with the analytical object and a second portion of the results data corresponds with the trained analytical model.

Interaction data integrity protection in a data access and analysis system includes obtaining user account data including a user account identifier and a public key, receiving, from a user device, first system interaction data indicating a request to interact with the data access and analysis system, sending, to the user device, first interaction response data including interaction challenge data, wherein the interaction challenge data includes a system interaction identifier that uniquely identifies the request to interact with the data access and analysis system, receiving, from the user device, second system interaction data signed with a private key of the asymmetric key pair, wherein the second system interaction data includes the system interaction identifier, validating the second system interaction data using the public key, and recording validated system interaction data obtained by the validating, the validated system interaction data including the second system interaction data signed with the private key.

Systems and methods for mapping natural language to queries using a query grammar are described. For example, methods may include generating, based on a string, a set of tokens of a database syntax; generating a query graph for the set of tokens from a finite state machine representing a query grammar, wherein nodes of the finite state machine represent token types, directed edges of the finite state machine represent valid transitions between token types in the query grammar, vertices of the query graph correspond to respective tokens of the set of tokens, and directed edges of the query graph represent a transition between two tokens in a sequencing of the tokens; determining, based on a tour of the query graph, a sequence of the tokens in the set of tokens, forming a database query; and invoking a search of a database using a query based on the database query to obtain search results.

Automatic data-analysis formula phrase generation includes obtaining a natural language string expressing a request for an automatically generated data-analysis formula phrase, obtaining first prompt data including a first proper subset of defined data-analysis formula phrases previously defined in a data access and analysis system in accordance with a defined data-analysis-formula grammar and associated with defined data-analysis formula phrase categories, obtaining first large language model generate data responsive to the first prompt data, obtaining a proper subset of the defined data-analysis formula phrase categories, obtaining second prompt data including a second proper subset of the defined data-analysis formula phrases identified in accordance with the proper subset of the defined data-analysis formula phrase categories, and obtaining second large language model generated data responsive to the second large language model input data that includes the automatically generated data-analysis formula phrase.

Semantic analysis to resolve ambiguous user input data with respect to a request for data includes identifying tokens based on the text string, identifying a tables in a database, wherein a token indicates a column from a table and a token indicates a column from another table, identifying candidate join paths for joining tables, obtaining respective candidate results using the candidate join paths, outputting combined candidate results including values from the respective candidate results, obtaining second user input data indicating a selected value, identifying the request for data as unambiguous and identifying a selected join path based on the selected value, and, in response to identifying the request for data as unambiguous, outputting data responsive to the request for data using the selected join path.

Generating object morphisms during object search includes obtaining object-search request data, wherein the object-search request data includes object-search terms, obtaining resolved-request data representing the object-search terms, determining that a first analytical object partially consistent with the resolved-request data is available, wherein the first analytical object is consistent with a first portion of the resolved-request data, generating candidate object-morphism data with respect to the first analytical object in accordance with a second portion of the resolved-request data, outputting object-search response data including the candidate object-morphism data for presentation to a user, obtaining data indicating a selected object morphism from the candidate object-morphism data, generating a second analytical object in accordance with the first analytical object and the selected object morphism, wherein the second analytical object differs from the first analytical object, and outputting response data including the second analytical object for presentation to the user.

Systems and methods for query generation based on a logical data model with one-to-one joins are described. For example, methods may include accessing a join graph representing tables in a database; receiving a first query; selecting a connected subgraph of the join graph that includes the two or more tables referenced in the first query; accessing an indication that a directed edge of the connected subgraph corresponds to a one-to-one join; modifying the connected subgraph based on the indication to obtain a modified subgraph; generating one or more leaf queries based on the modified subgraph; generating a query graph that specifies joining of results from queries based on the one or more leaf queries; invoking a transformed query on the database that is based on the query graph and the queries based on the one or more leaf queries.

Systems and methods for mapping natural language to queries using a query grammar are described. For example, methods may include generating, based on a string, a set of tokens of a database syntax; generating a query graph for the set of tokens from a finite state machine representing a query grammar, wherein nodes of the finite state machine represent token types, directed edges of the finite state machine represent valid transitions between token types in the query grammar, vertices of the query graph correspond to respective tokens of the set of tokens, and directed edges of the query graph represent a transition between two tokens in a sequencing of the tokens; determining, based on a tour of the query graph, a sequence of the tokens in the set of tokens, forming a database query; and invoking a search of a database using a query based on the database query to obtain search results.

Operating a low-latency database analysis system with phrase translation may include obtaining a locale-specific phrase localization rule and a canonical phrase localization rule for a phrase, generating a locale-specific index and a locale-specific finite state machine for the locale using the localization definition data and a canonical finite state machine, generating a resolved-request by obtaining a locale-specific token representing locale-specific input data by traversing the locale-specific index, obtaining a canonical token associated with locale-specific token, obtaining a locale-specific phrase by traversing the locale-specific finite state machine, obtaining a canonical phrase corresponding to the locale-specific phrase, the canonical phrase including the canonical token, generate a data-query based on the canonical phrase, obtaining results data responsive to the data expressing the usage intent by executing a query corresponding to the data-query by an in-memory database of the low-latency database analysis system, and outputting the results data for presentation to a user.