Natural language queries based on user defined attributes

A non-transitory computer readable medium storing instructions for causing a processor to perform steps comprising:

• storing a dataset comprising one or more attributes for data analysis;
• storing information describing a plurality of intents of natural language queries, each intent associated with:
• criteria for identifying natural language queries having the intent;
• and
• instructions for processing data of the dataset according to the intent;
• receiving information describing a user defined metric, the information associating a natural language phrase identifying the user defined metric with an expression, the natural language phrase comprising a name in natural language, the expression associating the user defined metric with a column of the dataset and specifying a user-selected subset of named entities in the column, the expression enumerating each of the named entities in the user-selected subset, the user defined metric editable to add or remove a named entity enumerated in the expression;
• receiving a plurality of natural language queries, each using the natural language phrase identifying the user defined metric;
• identifying, for each natural language query, an intent of the natural language query responsive to the natural language query satisfying the criteria associated with the identified intent;
• processing the natural language query for each natural language query, the processing comprising:
• generating a database query for retrieving data of the dataset as requested by the natural language query, wherein the database query applies the expression of the user defined metric to select records that are associated with the user-selected subset of named entities enumerated in the expression;
• and
• executing the database query to process the data of the dataset;
• and
• storing a plurality of documents, each document corresponding to one of the plurality of natural language queries, each document storing a result of processing the one of the plurality of the natural language queries and storing an association between the result and the one of the plurality of natural language queries that identifies the user defined metric;
• receiving an edit to the expression of the user defined metric, the edit adding or removing a named entity from the user-selected subset;
• and
• responsive to the edit, updating the plurality of documents based on the association stored in each document, the updating of the plurality of documents comprising re-evaluating the plurality of natural language queries based on the edit to the user define metric.

A computer system comprising:

• a computer processor;
• and
• a non-transitory computer readable medium storing instructions executable by the processor, the instructions for:
• storing a dataset comprising one or more attributes for data analysis;
• storing information describing a plurality of intents of natural language queries, each intent associated with:
• criteria for identifying natural language queries having the intent;
• and
• instructions for processing data of the dataset according to the intent;
• receiving information describing a user defined metric, the information associating a natural language phrase identifying the user defined metric with an expression, the natural language phrase comprising a name in natural language, the expression associating the user defined metric with a column of the dataset and specifying a user-selected subset of named entities in the column, the expression enumerating each of the named entities in the user-selected subset, the user defined metric editable to add or remove a named entity enumerated in the expression;
• receiving a plurality of natural language queries, each using the natural language phrase identifying the user defined metric;
• identifying, for each natural language query, an intent of the natural language query responsive to the natural language query satisfying the criteria associated with the identified intent;
• processing the natural language query for each natural language query, the processing comprising:
• generating a database query for retrieving data of the dataset as requested by the natural language query, wherein the database query applies the expression of the user defined metric to select records that are associated with the user-selected subset of named entities enumerated in the expression;
• and
• executing the database query to process the data of the dataset;
• and
• storing a plurality of documents, each document corresponding to one of the plurality of natural language queries, each document storing a result of processing the one of the plurality of the natural language queries and storing an association between the result and the one of the plurality of natural language queries that identifies the user defined metric;
• receiving an edit to the expression of the user defined metric, the edit adding or removing a named entity from the user-selected subset;
• and
• responsive to the edit, updating the plurality of documents based on the association stored in each document, the updating of the plurality of documents comprising re-evaluating the plurality of natural language queries based on the edit to the user define metric.