Bridge from natural language processing engine to database engine

A computer-implemented method comprising:

• receiving a data structure representing a natural language query from a natural language processing (NLP) engine, the data structure comprising, for respective fields of the data structure, an intent and one or more query parameters;
• receiving a ruleset comprising a plurality of rules, each rule respectively comprising a field specification and identifying a function;
• for each of the plurality of rules, determining whether the data structure matches the respective field specification, wherein the data structure is determined to match the respective field specification for at least first, second, and third rules of the plurality of rules;
• for the first rule, executing the respective function to add a sub-intent value to the data structure;
• for the second rule, executing the respective function to update the query parameters in the data structure;
• for the third rule, executing the respective function to construct a filter string from one or more of the received or updated query parameters;
• constructing a query string comprising a base record locator corresponding to the sub-intent and the filter string;
• and
• transmitting the query string to a web service of a database engine for generation of a result.

A system comprising:

• one or more hardware processors with memory coupled thereto;
• and
• computer-readable storage media storing instructions which, when executed by the one or more hardware processors, cause the one or more hardware processors to implement a natural language processing (NLP) engine and a bridge module;
• wherein the NLP engine is configured to receive a natural language query and process the natural language query by:
• preprocessing the natural language query to obtain a preprocessed query;
• applying an annotator, named entity recognition, and lexical rules in parallel to obtain a plurality of parse trees representing the natural language query;
• applying a feature function to obtain respective scores for the parse trees;
• selecting one of the parse trees based on the scores;
• and
• identifying an intent and input query parameters according to the selected parse tree;
• and
• storing the intent and the input query parameters in an intermediate data structure;
• wherein the bridge module is configured to generate a query string, corresponding to the natural language query, by:
• comparing a plurality of rules to attributes of the intermediate data structure to determine one or more matching rules of the plurality of rules;
• based on the one or more matching rules:
• identifying a sub-intent and a corresponding base record locator;
• and
• transforming a set of the input query parameters into a set of output query parameters;
• and
• constructing a query filter from the output query parameters;
• and
• constructing the query string from the base record locator and the query filter;
• and
• wherein the system is configured to obtain a result for the natural language query by:
• transmitting the query string to the database engine.