Servicenow, Inc.

Aliaksei Vertsel

Mikhail Rumiantsau

Alexey Zenovich

Alyaksandr Zaytsav

Seth Tucker Stafford

Dmitry Korolev

A system may include: a database table, a grammar containing metadata-referencing non-terminals, and metadata mapping the metadata-referencing non-terminals into alternative terms for the table and a column therein. The system may also include a server device configured to: (i) receive a natural language query, the grammar, and the metadata, wherein the natural language query contains a first alternative term for the table and a second alternative term for the column; (ii) parse, using the grammar, the natural language query into a series of terminals and non-terminals including a first metadata-referencing non-terminal that the metadata maps to the first alternative term and a second metadata-referencing non-terminal that the metadata maps to the second alternative term; (iii) identify, based on the first metadata-referencing non-terminal, the table; (iv) identify, based on the second metadata-referencing non-terminal, the column; and (v) generate, based on the table and the column as identified, a database language query.

Metadata-based translation of natural language queries into database queries

An example embodiment may involve a software application executable on computing devices of a remote network management platform and a computation instance associated with a managed network. The computational instance may contain a database storing data of the managed network. The software application may receive, from a client device of the managed network, a natural language query (NLQ), and retrieve Backus-Naur form (BNF) rules and a set of metadata associated with the BNF rules. The metadata may include a text-based description of a schema of the database and abbreviations associated with the BNF rules. The NLQ may be parsed using the BNF rules together with the metadata by applying the metadata during parsing to extend the BNF rules. A query object based on the parsed query may be generated, and the database searched using the query object. A result of the search may be transmitted to the client device.

Enhanced natural language processing with semantic shortcuts

An example embodiment may involve a software application executable on a computing device of a computational instance of a remote network management platform associated with a managed network. The software application may receive, from a client device of the managed network, a natural language query, determine that a first term of the query does not match any term in a list of predetermined terms, and revise the query by (i) applying a set of linguistic rules that modify the first term to generate a replacement term, (ii) determining whether the replacement term matches at least one term in the list, and (iii) after determining that the replacement term matches at least one term in the list, replacing the first term in the query with the replacement term. The software application may then parse the revised query, search the database, and transmit a result to the client device.

Parsing of user queries in a remote network management platform using linguistic matching

An example embodiment may involve a software application executable on a computing device of a computational instance of a remote network management platform associated with a managed network. The computational instance may contain a database from which information associated with the managed network can be delivered to the client device. The software application may receive, from a client device of the managed network, a natural language query, parse the query using context-free grammar rules in Backus-Naur form (BNF), generate a search sentence based on the parsed query, search the database using the search sentence as a search query, and transmit, to the client device, a result of the search query. The context-free grammar rules may include an extended context-free grammar rule that defines a non-terminal symbol and comprises a Boolean NOT operator that eliminates one or more potential definitions of the non-terminal symbol from consideration when parsing the query.

Parsing of user queries in a remote network management platform using extended context-free grammar rules

Translating a natural language search query into a query language includes receiving a natural language query for a database, processing the natural language query to generate a modified text input, generating an entity tree based on the modified text input, including assigning one or more semantic markers to one or more words or one or more groups of words within the modified text input, wherein each semantic tag denotes a semantic class for each respective word or group or words, and converting the entity tree into the query language associated with the first database.

Growing community of inventors

Mountain View, CA, United States of America

Aliaksei Vertsel