Query Template Based Architecture For Processing Natural Language Queries For Data Analysis

Abstract

A data analysis system allows users to interact with distributed data structures stored in-memory using natural language queries. The data analysis system receives a prefix of a natural language query from the user. The data analysis system provides suggestions of terms to the user for adding to the prefix. Accordingly, the data analysis system iteratively receives longer and longer prefixes of the natural language queries until a complete natural language query is received. The data analysis system stores natural language query templates that represent natural language queries associated a particular intent. For example, a natural language query template may represent queries that compare two columns of a dataset. The data analysis system compares an input prefix of natural language with the natural language query templates to determine the suggestions. The data analysis system receives user defined metrics or attributes that can be used in the natural language queries.

We claim: 
     
         1 . A computer-implemented method for visualization of natural language queries, the computer-implemented method comprising:
 storing a plurality of natural language query templates;   receiving a natural language query requesting analysis of a dataset;   matching the natural language query against the plurality of natural language query templates, a matching query template comprising a query intent and one or more attributes of a dataset;   identifying the query intent and the one or more attributes based on a matching query template;   determining whether the natural language query specifies a pivot clause associated with a pivot attribute;   determining a chart type for visualizing the natural language query based on the query intent, the one or more attributes, and whether the natural language query specifies a pivot clause; and   rendering a chart of the identified chart type based on the data set and the natural language query and sending for presentation.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein the chart type is determined to be a first chart type if the natural language query specifies a pivot clause and the chart type is a second chart type if the natural language query does not specify a pivot clause. 
     
     
         3 . The computer-implemented method of  claim 1 , wherein if the query intent is to show relationship between a first attribute and a second attribute and the first attribute is of numeric type and the second attribute is of numeric type, the chart type is determined to be scatter plot chart if the natural language query does not specify a pivot clause. 
     
     
         4 . The computer-implemented method of  claim 3 , wherein the chart type is determined to be a hex binned scatter plot if the natural language query does not specify a pivot clause. 
     
     
         5 . The computer-implemented method of  claim 3 , wherein the chart type is determined to be a small multiples hex binned scatter plot if the natural language query specifies a pivot clause. 
     
     
         6 . The computer-implemented method of  claim 3 , wherein the chart type is determined to be a scatter plot with liner trend lines if the natural language query specifies a pivot clause. 
     
     
         7 . The computer-implemented method of  claim 1 , wherein if the query intent is to show relationship between a first attribute and a second attribute and the first attribute is of numeric type and the second attribute is of categorical type, the chart type is determined to be a bar chart if the natural language query does not specify a pivot clause. 
     
     
         8 . The computer-implemented method of  claim 7 , wherein the chart type is determined to be a grouped bar chart bar chart if the natural language query specifies a pivot clause. 
     
     
         9 . The computer-implemented method of  claim 1 , wherein determining the chart type is further based on a number of categories of one of the attributes. 
     
     
         10 . The computer-implemented method of  claim 1 , further comprising:
 suggesting one or more additional chart types for a user to select.   
     
     
         11 . A non-transitory computer-readable medium configured to store computer code comprising instructions, the instructions, when executed by one or more processors, cause the one or more processors to perform steps comprising:
 storing a plurality of natural language query templates;   receiving a natural language query requesting analysis of a dataset;   matching the natural language query against the plurality of natural language query templates, a matching query template comprising a query intent and one or more attributes of a dataset;   identifying the query intent and the one or more attributes based on a matching query template;   determining whether the natural language query specifies a pivot clause associated with a pivot attribute;   determining a chart type for visualizing the natural language query based on the query intent, the one or more attributes, and whether the natural language query specifies a pivot clause; and   rendering a chart of the identified chart type based on the data set and the natural language query and sending for presentation.   
     
     
         12 . The non-transitory computer-readable medium of  claim 11 , wherein the chart type is determined to be a first chart type if the natural language query specifies a pivot clause and the chart type is a second chart type if the natural language query does not specify a pivot clause. 
     
     
         13 . The non-transitory computer-readable medium of  claim 11 , wherein if the query intent is to show relationship between a first attribute and a second attribute and the first attribute is of numeric type and the second attribute is of numeric type, the chart type is determined to be scatter plot chart if the natural language query does not specify a pivot clause. 
     
     
         14 . The non-transitory computer-readable medium of  claim 13 , wherein the chart type is determined to be a hex binned scatter plot if the natural language query does not specify a pivot clause. 
     
     
         15 . The non-transitory computer-readable medium of  claim 13 , wherein the chart type is determined to be a small multiples hex binned scatter plot if the natural language query specifies a pivot clause. 
     
     
         16 . The non-transitory computer-readable medium of  claim 13 , wherein the chart type is determined to be a scatter plot with liner trend lines if the natural language query specifies a pivot clause. 
     
     
         17 . The non-transitory computer-readable medium of  claim 11 , wherein if the query intent is to show relationship between a first attribute and a second attribute and the first attribute is of numeric type and the second attribute is of categorical type, the chart type is determined to be a bar chart if the natural language query does not specify a pivot clause. 
     
     
         18 . The non-transitory computer-readable medium of  claim 17 , wherein the chart type is determined to be a grouped bar chart bar chart if the natural language query specifies a pivot clause. 
     
     
         19 . The non-transitory computer-readable medium of  claim 11 , wherein determining the chart type is further based on a number of categories of one of the attributes. 
     
     
         20 . The non-transitory computer-readable medium of  claim 11 , wherein the steps further comprise:
 suggesting one or more additional chart types for a user to select.