Natural Language Interface for Data Visualization Intent Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing natural language interfaces in data visualization systems fail to accurately infer user intent, leading to limited and ineffective visualizations, as they often rely on explicit data attributes and chart types, restricting the relevance and interactivity of generated responses.
Innovation Solution
A natural language interface within a data visualization application that uses both context and intent to support analytical workflows by determining explicit and implicit user intents from natural language commands, generating visual specifications, and modifying data visualizations dynamically based on user input, allowing for interactive and adaptive visualization responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If natural language interfaces rely on explicit data attributes and chart types, then the system complexity is reduced and ease of operation is improved, but the accuracy of user intent inference deteriorates and the relevance of generated visualizations is limited
Solution Approach 1:
The patent introduces an intermediary intent inference mechanism that bridges natural language input and data visualization output. The system uses context from the current visualization state and analytical conversation history to infer implicit user intent, acting as a mediator between simple keyword matching and complex user needs understanding.
Solution Approach 2:
The system performs preliminary actions by pre-defining a set of possible user intents and preparing context extraction mechanisms before processing user queries. This allows the system to quickly match natural language commands against predefined intent patterns while leveraging pre-computed context from the current visualization state.
2Productivity
If natural language interfaces infer limited aspects of user intent, then the processing speed is improved and productivity is enhanced, but the adaptability and versatility of the system deteriorate
Solution Approach 1:
The patent implements a dynamic intent inference system that adapts its analysis depth based on the conversation context. The system dynamically adjusts which aspects of intent to infer based on the current visualization state, previous commands, and the apparent complexity of the user's information needs, balancing processing speed with adaptability.
Solution Approach 2:
The system achieves versatility through a universal intent inference framework that handles multiple types of user queries (filtering, aggregation, comparison, trend analysis) using a unified approach. The same context extraction and intent classification mechanisms serve multiple analytical functions, enhancing adaptability without proportionally increasing complexity.
3Ease of operation
If the system generates basic interactive visualizations in response to queries, then the ease of operation is improved, but the reliability and quality of analytical insights deteriorate
Solution Approach 1:
The patent incorporates feedback mechanisms where the system continuously monitors the current visualization state and analytical conversation history to refine intent inference. The feedback from previous user interactions and the current visualization context improves the accuracy of subsequent visualization generation, ensuring higher quality analytical insights while maintaining ease of use.
4Measurement precision
If the system requires expert modeling to create effective queries, then the measurement precision of query formulation is improved, but the ease of operation and accessibility to non-experts deteriorates
Solution Approach 1:
The system enables self-service by automatically inferring user intent from natural language commands without requiring expert knowledge of data modeling or query formulation. The context-aware intent inference mechanism allows non-experts to obtain effective visualizations by simply describing their information needs in natural language, while the system handles the complex query formulation automatically.
Data Source
AI summary
A computing device displays an initial data visualization according to an initial visual specification that specifies a data source, visual variables, and data fields from the data source. The computing device receives a natural language command that includes a request for information from the data source. The computing device extracts one or more first keywords and determines, based on the first keywords and one or more of (i) the data source, (ii) the visual variables, and/or (iii) the data fields of the initial visual specification, that the request does not directly specify a characteristic in the initial visual specification. The computing device generates a modified visual specification. The computing device generates one or more queries based on the modified visual specification. The computing device executes the one or more queries to retrieve data for a modified data visualization. The computing device generates and displays the modified data visualization.