Visual Data Story Generation with Human-in-the-Loop Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automated data analysis systems fail to accurately identify data trends, misinterpret relationships in complex datasets, and inefficiently generate narratives, leading to excessive computational resources and manual review by analysts, who must integrate multiple tools to create coherent reports.
Innovation Solution
The system intelligently analyzes input data to generate visual data stories with graphical visualizations and natural language summaries, using statistical analysis to determine insights and create a visual-data-story graph for selecting relevant narratives, thereby automating the process and reducing computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional automated data analysis systems generate narratives from large amounts of raw data, then data analysis can be automated, but the systems fail to accurately identify data trends and misidentify relationships among complex datasets
Solution Approach 1:
The patent introduces an intermediary human analyst who reviews and validates the automatically generated insights and narratives. This human-in-the-loop approach allows the system to maintain automation while improving accuracy through human expertise in interpreting complex data relationships and validating trend identifications.
2Quantity of substance
If conventional systems generate an inordinate number of extracted insights, then more potential data narratives can be created, but the process becomes time consuming and tedious for analysts to search through
Solution Approach 1:
The patent extracts only the most relevant and salient insights from the large set of generated insights by having human analysts selectively identify and prioritize those that contribute most meaningfully to the narrative. This extraction approach reduces the volume of insights analysts must review while maintaining narrative quality.
Solution Approach 2:
The system generates an excessive number of insights initially (beyond what is strictly necessary), then uses human judgment to filter and select the appropriate subset. This partial action approach ensures that enough insights are generated to capture all potential narratives, while human curation removes the excess to create a manageable review set.
3Quantity of substance
If conventional systems inefficiently utilize computing resources by generating excessive insights, then more comprehensive analysis can be performed, but computing resources are wasted on uninteresting or irrelevant insights
Solution Approach 1:
The system performs preliminary automated generation of all potential insights and narratives before human review, using computing resources efficiently to create a comprehensive set of candidates. This preliminary action allows the system to leverage automated processing for tasks where it excels while reserving human expertise for higher-value validation and selection tasks.
4Adaptability or versatility
If conventional systems cannot integrate multiple applications and tools, then each tool can be used independently, but data analysts must manually transfer insights to separate report-creation tools to construct presentable reports
Solution Approach 1:
The patent merges the data analysis, insight generation, and report creation functions into an integrated system where insights are automatically transferred and formatted for presentation. This combination eliminates the need for manual data transfer between separate tools while maintaining the versatility of using different analysis approaches and presentation formats within a unified workflow.
5Productivity
If conventional systems determine cursory insights by enumerating combinations of data fields, then statistical facts can be calculated, but in-depth insights that identify unique and meaningful analyses are not provided
Solution Approach 1:
The system performs preliminary automated calculation of statistical facts and basic insights quickly, then uses human analysts to build upon these foundations and develop deeper, more meaningful analyses. This preliminary action allows the system to efficiently generate the data foundation while human expertise adds the interpretive depth that automated systems cannot achieve alone.
Data Source
AI summary
This disclosure describes one or more embodiments of systems, non-transitory computer-readable media, and methods that intelligently and automatically analyze input data and generate visual data stories depicting graphical visualizations from data insights determined from the input data. For example, the disclosed systems automatically extract data insights utilizing an in-depth statistical analysis of dataset groups from data-attribute categories within the input data. Based on the data insights, the disclosed systems can automatically generate exportable visual data stories to visualize the data insights, provide textual or audio-based natural language summaries of the data insights, and animate such data insights in videos. In some embodiments, the disclosed systems generate a visual-data-story graph comprising nodes representing visual data stories and edges representing similarities between the visual data stories. Based on the visual-data-story graph, the disclosed systems can select a relevant visual data story to display on a graphical user interface.