Data Visualization Query Parsing for Ambiguous Natural Language
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data visualization systems struggle with handling linguistically colloquial, underspecified, or ambiguous natural language utterances, leading to inefficiencies in data analysis due to the complexity of realizing these utterances against underlying data and analytical functions, and the high cognitive burden on users.
Innovation Solution
A data visualization application employs a context-free grammar and semantic constraints to form intermediate expressions, inferring omitted information using syntactic and semantic rules, translating these into formal queries to generate accurate data visualizations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If natural language interfaces are used for data exploration, then ease of operation is improved, but handling colloquial and ambiguous utterances increases device complexity
Solution Approach 1:
The patent introduces an intermediate language as a mediator between natural language input and the visual analytics system. This intermediate language serves as a formal representation that bridges the gap between colloquial user utterances and the precise analytical functions required by the system, thereby maintaining ease of operation while managing complexity through a structured intermediary layer
Solution Approach 2:
The system segments the natural language processing task into distinct stages: initial parsing to form intermediate expressions, inference rules to resolve ambiguities, and translation to analytical functions. This segmentation allows each component to handle specific aspects of the complexity independently, making the overall system more manageable while preserving user-friendly natural language interaction
2Measurement precision
If manual grammar authoring is performed for each database, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent implements a universal grammar framework that can be applied across multiple databases and data sources without requiring separate manual authoring for each. The inference rules and intermediate language design enable the system to handle diverse data types and analytical functions through a single, reusable grammatical structure, thereby maintaining precision while eliminating the time-consuming repetitive authoring process
Solution Approach 2:
The system performs preliminary action by pre-defining inference rules and grammatical structures that can automatically resolve ambiguities and interpret underspecified utterances. This preliminary preparation of the linguistic framework allows the system to quickly process natural language queries without requiring time-consuming manual grammar customization for each new database
Data Source
AI summary
A computing device parses a user-specified natural language command to form a first expression. The computing device determines that the first expression is ambiguous or underspecified. The computing device, in accordance with the determination, infers first information using one or more inferencing rules, where at least one of the inferencing rules is based on an attribute of data fields and/or data values in a data source. The computing device forms a second expression based on the first expression using and the first information. The computing device retrieves one or more data sets from the data source using according to the second expression. The computing device generates and displays a data visualization of the retrieved one or more data sets.