Descriptive Insight Generation for Accessible BI Dashboards
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional business intelligence systems require users to visually analyze data, which can be tedious and may lead to overlooked important changes, and are not accessible to visually impaired users.
Innovation Solution
A system that converts digital visual graphs to structured datasets, generates automated descriptive insights using machine learning, and converts insights into accessible formats like text or speech for visually impaired users, while conserving computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If users manually analyze visual graphs to generate insights, then the insights can be customized and detailed, but the process is tedious and time-consuming
Solution Approach 1:
The system enables self-service automated insight generation by allowing users to configure insight templates and parameters once, after which the system automatically generates insights without requiring manual analysis for each new dataset. The system serves itself by autonomously processing data through configured templates and algorithms.
Solution Approach 2:
Users perform preliminary actions by configuring insight templates, selecting data sources, and setting parameters in advance. These preliminary configurations enable automated insight generation without requiring manual analysis when new data arrives, significantly reducing the time needed for subsequent insight generation while maintaining accuracy.
2Difficulty of detecting and measuring
If visual graphs are used to present data, then the data can be visually analyzed for patterns and trends, but the system is not accessible to visually impaired users
Solution Approach 1:
The system achieves universality by generating insights in multiple formats suitable for different user needs. The same data analysis process produces both visual graph outputs for sighted users and alternative text-based or audio-based insights for visually impaired users, making the system accessible to all users regardless of visual ability.
Solution Approach 2:
The automated insight generation system acts as an intermediary that translates visual data patterns into multiple accessible formats. Instead of requiring users to directly interpret visual graphs, the system mediates by analyzing the data and presenting insights in various formats including text descriptions, audio outputs, and simplified visual representations.
3Loss of information
If comprehensive data analysis is performed to identify all patterns and changes, then complete insights are generated, but computing resources are consumed
Solution Approach 1:
The system applies partial action by focusing analysis on specific aspects of the data that are most relevant to the configured insight templates. Rather than performing exhaustive analysis of all possible patterns, the system selectively analyzes data elements that match the template criteria, reducing computing resource consumption while maintaining insight completeness for the intended purpose.
Solution Approach 2:
The system applies local quality by tailoring the depth and type of analysis to specific data regions and insight types. Different parts of the data receive different levels of analysis intensity based on their relevance to the configured insights, optimizing computing resource allocation while ensuring complete coverage of important patterns and changes.
Data Source
AI summary
A system is described that generates descriptive insights in a manner that does not require observations to be made by a visually-impaired user and that can present insights in a form perceptible by such a user. A structured dataset or a digital visual graph may include business intelligence or other types of data. In the case of a graph, the graph is converted to the structured dataset. Parameter names in the dataset are encoded using parameter metadata. Relationships among the data of the dataset are identified based on the encoded parameter names and content of the parameters. The relationships are evaluated based on domain knowledge to generate insights. The insights are applied to automatically-selected text templates to generate descriptive insights. The descriptive insights may be presented to a user in a user interface (e.g., in a BI dashboard) or converted to a form perceptible by a visually-impaired user (e.g., speech).