Bidirectional Data Storytelling With Linked Text and Visualizations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storytelling tools lack bidirectional generation capabilities, forcing authors to switch contexts between data analysis, visualization creation, and narrative composition, leading to disjointed and inefficient workflows, and automated tools often rely on significance scores that limit human creativity and accuracy.
Innovation Solution
DataWeaver enables bidirectional composition of data narratives through visualization-to-text and text-to-visualization workflows, using Large Language Models (LLMs) to generate narratives anchored to user interactions and data facts, with features like deictic referencing and tone control, facilitating seamless navigation and accurate narrative generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated tools use significance scores to determine interesting data, then efficiency is improved, but human creativity and accuracy are limited
Solution Approach 1:
The patent introduces an intermediary layer between automated significance scoring and final narrative generation. This layer includes human-in-the-loop validation mechanisms where users can review, adjust, or reject AI-generated narratives before publication, thereby maintaining efficiency while improving accuracy and creativity
2Measurement precision
If authors manually transcribe numerical data into textual data facts, then accuracy is improved, but time consumption increases
Solution Approach 1:
The system performs preliminary automated transcription of numerical data to textual data facts using LLMs before the authoring process begins. This pre-processing step creates draft narratives that authors can then review and refine, significantly reducing the time required while maintaining acceptable accuracy levels
3Device complexity
If tools support only data-to-story workflow, then automation is simplified, but text-first authors are discouraged and workflow flexibility is reduced
Solution Approach 1:
The patent implements a universal authoring tool that supports multiple workflows: data-to-story (traditional), text-to-data (for writers first), and bidirectional iteration. The system can function as different types of tools depending on user needs, making it adaptable to various authoring styles while maintaining a unified interface
4Ease of operation
If authors switch contexts between data analysis, visualization creation, and narrative composition, then task completion is achieved, but workflow efficiency decreases
Solution Approach 1:
The patent merges data analysis, visualization creation, and narrative composition into a single integrated authoring environment. All three functions operate within one tool with shared data models and coordinated workflows, eliminating context switching while ensuring all tasks are completed through unified project management
Data Source
AI summary
A computing device displays, on a user interface, a first visualization node and a text node. The first visualization node is upstream of the text node and includes a chart. In response to receiving a first user interaction with a portion of text content that is displayed in the text node, the device retrieving data of the first visualization node and transmits to a language model a first request that includes the data and the portion of text content. The device receives from the language model suggestions and data visualizations for analyzing the one or more datasets. The device displays the plurality of suggestions and their corresponding data visualizations on the user interface. The device receives user selection of a first suggestion that includes a first data visualization and renders a second visualization node on the user interface. The device displays the first data visualization in the second visualization node.