Business Intelligence Search System for Natural Language Query Translation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Business intelligence systems lack user-friendly interfaces that allow non-sophisticated users to perform textual searches on complex databases and data warehouses, failing to provide accessible and understandable responses, especially on mobile devices.

Innovation Solution

A business intelligence search system that translates textual queries into query languages like SQL, generates visualizations, and learns user preferences through interaction data to provide intuitive access to data stored in databases and data warehouses, using a client-server architecture with components like search query rewriters, index searchers, and text-to-query generators.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If complex query structures and reports in database vernacular are used to access information, then the system can retrieve accurate data from databases and data warehouses, but typical users cannot access this information due to lack of sophistication and training

Engineering Contradiction:
ImproveUser accessibility to database queriesVSAvoidQuery structure complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent introduces a natural language processing intermediary layer between users and database query structures. Users input questions in plain English through a chat interface, and the system automatically translates these into appropriate SQL queries and visualizations, eliminating the need for users to learn complex query languages while maintaining accurate data retrieval

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical system of learning and using complex query syntax with an automated natural language processing system. Instead of users manually constructing queries according to database vernacular rules, the system uses AI to interpret user intent and generate appropriate queries automatically

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If traditional business intelligence systems are designed for sophisticated users with training, then they can effectively query complex data stores, but they fail to provide accessible interfaces for typical users who want simple textual searching

Engineering Contradiction:
ImproveSystem adaptability to different user skill levelsVSAvoidInterface simplicity
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent makes the business intelligence system universal by implementing a multi-functional interface that serves both sophisticated users and typical users. The same chat interface can handle simple natural language queries from novice users and complex analytical requests from experienced users, automatically adjusting the level of abstraction and explanation provided

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent implements dynamic adaptation where the system adjusts its response format, level of detail, and explanation based on user interaction patterns and preferences. The interface dynamically modifies itself to match user expertise levels while maintaining access to complex data warehouse capabilities

Inventive Principle:
Principle #15Dynamics

3Loss of information

If business intelligence systems provide detailed query results in traditional formats, then they contain complete information, but they do not produce responses in formats that users can easily understand

Engineering Contradiction:
ImproveInformation completenessVSAvoidResponse understandability
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The patent transforms query results from traditional tabular formats into visual dimensions such as charts, graphs, and interactive visualizations. This dimensional transformation maintains all the underlying data information while presenting it in visually intuitive formats that are easier for users to understand and interpret

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces a natural language explanation intermediary that translates complex query results into plain English summaries. The system provides both the complete detailed data for accuracy and automated narrative explanations for understandability, bridging the gap between information completeness and user comprehension

Inventive Principle:
Principle #24Intermediary (Mediator)

4Adaptability or versatility

If users need to access data wherever they are located using mobile devices, then users gain flexibility and mobility, but traditional business intelligence systems do not support mobile access with textual query input

Engineering Contradiction:
ImproveMobile device compatibilityVSAvoidTextual search capability on mobile
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent implements a universal chat interface that functions consistently across desktop and mobile devices. The same natural language processing capabilities and visualized response formats are available on mobile devices, enabling users to perform complex data warehouse queries using simple text input anywhere they have internet connectivity

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS10664488B2Semantic searches in a business intelligence system
Publication Date: 2020.05.26 ORACLE INT CORP
  • US10664488B2 patent drawing
  • US10664488B2 patent drawing
  • US10664488B2 patent drawing

AI summary

A computer-implemented method of executing a user query includes presenting a user interface to allow a user to enter a query, receiving a user-entered textual request through the interface, launching a search service to rewrite the textual request into a search query, sending the search query to a presentation server, receiving an answer to the query, and returning the answer to the user as a graphical representation. A computer-implemented method includes receiving a crawl request from a user, launching a crawl manager to monitor the crawl request and track statistics related to the crawl, starting a crawl task based upon the crawl request, indexing a business intelligence presentation server to create a data index, and storing the data index.