Visualization Recommendation System Using Meta Templates

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing data visualization systems often fail to provide high-quality insights when users have complex data analysis goals that do not fit within pre-defined categories, leading to inefficient data exploration and analysis.

Innovation Solution

A multi-stage approach that selects Meta templates based on user-defined data visualization intent and data fields, combining chart parts to form chart templates using combination rules, allowing for flexible and powerful visualization recommendations without requiring extensive resource-intensive training or lengthy testing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If classification-based systems are used to recommend visualizations, then the system is simple and easy to implement, but it cannot provide high-quality insights for complex data analysis goals that do not fit pre-defined categories

Engineering Contradiction:

Improveability to handle complex data analysis goalsVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the visualization recommendation process into multiple independent stages: (1) extracting visualization intents from user goals, (2) selecting appropriate Meta templates based on extracted intents, (3) combining chart parts according to combination rules, and (4) generating final chart templates. This segmentation allows the system to handle complex analysis goals by breaking them down into manageable components while maintaining system simplicity through modular design.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces dynamic adaptability by enabling the system to flexibly combine chart parts based on extracted visualization intents rather than relying on fixed pre-defined categories. The combination rules allow dynamic assembly of visualization components to match diverse user goals, making the system adaptable to complex data analysis scenarios without requiring extensive reconfiguration.

Inventive Principle:

Principle #15Dynamics

2Reliability

If extensive training and testing are performed to improve visualization recommendation quality, then the system produces high-quality results, but it requires increased resource requirements and lengthy training periods

Engineering Contradiction:

Improvevisualization recommendation qualityVSAvoidtraining time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by pre-defining Meta templates and combination rules that encode visualization best practices and patterns. These pre-established structures serve as a foundation that guides the recommendation process without requiring extensive runtime training. The system leverages this preliminary preparation to quickly generate high-quality recommendations by matching user goals against the pre-configured templates and rules.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system employs self-service mechanisms where the combination rules automatically guide the assembly of chart parts based on extracted visualization intents, without requiring external training data or manual intervention. The Meta templates and combination rules enable the system to self-regulate and produce reliable recommendations through inherent logical constraints and patterns built into the framework.

Inventive Principle:

Principle #25Self-service

3Adaptability or versatility

If standard classification-based systems are used, then the system is resource-efficient, but it cannot combine visualization pieces in new ways to address limitations

Engineering Contradiction:

Improveability to combine visualization piecesVSAvoidresource requirements

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent merges multiple chart parts according to combination rules to form comprehensive chart templates that address complex visualization needs. By combining individual chart components (such as combining trend analysis parts with comparison parts), the system creates new visualization configurations that go beyond standard categories while maintaining resource efficiency through rule-based assembly rather than generating entirely new visualizations from scratch.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The Meta templates and combination rules serve universal purposes by providing a framework that can handle diverse visualization requirements through a common structure. The same Meta templates and combination rules can be applied across different data types and analysis goals, making the system versatile without requiring separate resource-intensive models for each scenario.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10353909B2System and method for visualizing data

Publication Date: 2019.07.16 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Techniques are described for selecting, based on a data visualization intent specification and a defined set of data fields associated with a set of data, two or more Meta templates that meet the data visualization intent specification and that support the set of data fields, for determining chart parts that can be used within the selected Meta templates to form chart templates and for determining, based on a set of combination rules and the specification, the chart templates that meet the data visualization intent specification.