Natural Language Query Federation via Intermediate IQL2
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current business intelligence tools require skilled professionals for data analysis and manual ETL processes to extract insights from multiple data sources, leading to inefficiencies and increased administrative burdens, especially in enterprise environments with scattered data.
Innovation Solution
A system and method that converts natural language queries into intermediate query language 2 (IQL2), which is then parsed to generate database-specific queries, allowing data retrieval from multiple sources using a query engine selector, builder, and retriever modules, eliminating the need for manual data copying and ETL processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional ETL process is used to copy data from client's database to local database, then data can be extracted and stored locally, but it requires skilled data analysts and data modelers to create denormalized data models, increasing device complexity and loss of time
Solution Approach 1:
The patent introduces an intermediate query language (IQL2) as a mediator between natural language queries and database-specific query languages. The system automatically translates IQL2 queries into SQL or NoSQL queries based on the target database type, eliminating the need for manual data modeling and ETL script creation. This intermediary layer handles the complexity of data translation automatically.
Solution Approach 2:
The system enables self-service data extraction by allowing users to query data directly from multiple sources using natural language without requiring skilled data analysts. The automated query translation system performs the previously manual ETL process automatically, making the system self-sufficient and eliminating dependency on specialized professionals.
2Adaptability or versatility
If separate data modeling and ETL script is created for each data source in enterprise environment, then data can be extracted from multiple scattered sources, but it increases device complexity and requires multiple specialized skills
Solution Approach 1:
The patent creates a universal query interface that works across multiple database types (SQL and NoSQL) and multiple data sources. The intermediate query language (IQL2) serves as a universal abstraction layer that can be translated to any target database format, providing a single interface for accessing diverse data sources without requiring separate modeling for each source.
Solution Approach 2:
The IQL2 language acts as a universal intermediary that mediates between user queries and diverse data sources. The system automatically detects the target database type and translates the standardized IQL2 query into the appropriate database-specific query, eliminating the need for separate ETL scripts for each data source.
3Reliability
If skilled professionals are employed to create dashboards and reports, then analytical insights can be generated, but it increases loss of time for business users and creates administrative burden
Solution Approach 1:
The system enables business users to directly query data and generate analytical insights using natural language without requiring skilled data analysts or business intelligence developers. The automated query translation and execution system performs the previously manual analytical work automatically, eliminating the time loss and administrative burden associated with professional intermediaries.
Solution Approach 2:
The patent replaces the mechanical manual process of creating dashboards and reports with an automated computational system. Natural language processing algorithms automatically translate user queries into executable database queries, and the system automatically executes queries, processes data, and generates visualizations, substituting manual analytical work with automated computational processes.
Data Source
AI summary
A system and method for generating a database query from a natural language using a query engine selector module for identifying appropriate database query format, a query engine builder module to generate appropriate database query using an intermediate query language 2 (IQL2) as input, and a data retriever module for fetching data from single or multiple data sources.