Interaction-Based Visualization for Real-Time User Recommendations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data visualization technologies require significant user interaction and time for decoding data, with repetitive processes detracting from the user experience due to overlapping or grouped visual entities, and existing machine learning models ignore valuable user interaction data.
Innovation Solution
A method that captures user interactions with data visualizations, builds image stacks, generates embeddings, finds clusters of similar properties, and provides real-time recommendations to enhance user experience by personalizing and optimizing data visualization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data visualizations display all available data at once, then data completeness is improved, but user confusion and difficulty in understanding increase due to overlapping or grouped visual entities
Solution Approach 1:
The patent segments the complete data set into multiple batches or groups that are displayed sequentially rather than all at once. The system divides the data visualization into manageable portions, allowing users to explore different segments without being overwhelmed by the entire data set simultaneously, thus maintaining data completeness while improving understandability.
Solution Approach 2:
The system performs preliminary actions by pre-processing and organizing data into meaningful segments before presentation. It anticipates user needs by preparing data in advance in a structured manner, enabling progressive disclosure of information that reduces cognitive load while preserving access to the complete data set.
2Ease of operation
If data visualizations show truncated data to simplify display, then ease of understanding is improved, but data completeness deteriorates as some data is hidden or omitted
Solution Approach 1:
The patent implements feedback mechanisms that allow users to request additional data or alternative views when they encounter truncated information. The system responds to user feedback by providing access to the complete data set through interactive controls, ensuring that no data is permanently hidden and users can retrieve omitted information when needed.
Solution Approach 2:
The data visualization system is made dynamic, allowing users to adjust the level of detail and data completeness according to their needs. The system can transition between simplified views and complete data displays, adapting to user interactions and preferences, thus balancing ease of understanding with data completeness.
3Measurement precision
If users manually inspect and interact with charts to decode data, then data accuracy is improved, but time consumption increases significantly due to repetitive decoding processes
Solution Approach 1:
The patent enables the data visualization system to perform self-service functions by automatically analyzing and interpreting data patterns. The system uses machine learning models to pre-process and highlight key insights, reducing the need for manual inspection while maintaining data accuracy. The automated systems serve themselves by generating meaningful interpretations without requiring repeated user decoding efforts.
Solution Approach 2:
The patent replaces manual mechanical inspection processes with automated computational analysis. Machine learning models and algorithms substitute for human users manually decoding charts, performing the analysis automatically while preserving data accuracy. This substitution eliminates repetitive manual processes and significantly reduces time consumption.
4Quantity of substance
If data visualizations use complex visual encodings to represent more information, then information density is improved, but graphical perception difficulty increases making decoding more challenging
Solution Approach 1:
The patent applies local quality by using different visual encoding strategies for different parts of the data visualization. High-information-density encodings are applied where users need detailed information, while simpler encodings are used in areas requiring quick comprehension. This localized approach optimizes the balance between information density and perceptual ease for different regions of the visualization.
Solution Approach 2:
The system adds another dimension to data presentation by incorporating interactive elements and multiple view modes. Instead of relying solely on complex visual encodings in a single view, the system allows users to explore data across different dimensions and perspectives, distributing information density across multiple interactive layers rather than compressing it into a single complex visualization.
Data Source
AI summary
Techniques for augmenting data visualizations based on user interactions to enhance user experience are provided. In one aspect, a method for providing real-time recommendations to a user includes: capturing user interactions with a data visualization, wherein the user interactions include images captured as the user interacts with the data visualization; building stacks of the user interactions, wherein the stacks of the user interactions are built from sequences of the user interactions captured over time; generating embeddings for the stacks of the user interactions; finding clusters of embeddings having similar properties; and making the real-time recommendations to the user based on the clusters of embeddings having the similar properties.