Natural Language Data Visualization Image Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current data visualization tools lack sufficient shape libraries to handle large datasets, forcing users to manually find or create images, which hinders the visual analysis process and leads to a non-optimal user experience.

Innovation Solution

The implementation automatically identifies and selects semantically relevant images from existing shape libraries on the user's computer, local networks, or the Internet, using natural language to map categorical terms to images, thereby streamlining the visual analysis process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a limited shape library is provided in visualization tools, then the device complexity is reduced, but the adaptability to handle large cardinalities of data is insufficient

Engineering Contradiction:

Improveability to handle large cardinalities of dataVSAvoidshape library complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system uses natural language processing to enable a universal image search capability that can handle any categorical term across diverse domains. Instead of requiring domain-specific shape libraries, the system translates user queries into natural language searches across multiple image sources (local files, cloud storage, web), making the visualization tool adaptable to any data set without increasing structural complexity

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces a natural language interface as an intermediary between the user and the image selection process. This mediator translates categorical terms into search queries, automatically retrieves relevant images from multiple sources, and presents matched results to the user, thereby handling large cardinalities without requiring the user to directly manage complex shape libraries

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If users manually create or search for images for each data point, then the visual distinctiveness of data points is improved, but the time required for visualization creation increases significantly

Engineering Contradiction:

Improvevisualization creation speedVSAvoidtime spent on manual image selection

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-configuring multiple image sources (local folders, cloud storage accounts, web search capabilities) and establishing natural language processing rules before the user needs to create visualizations. When a user creates a visualization, the system has already prepared the infrastructure to quickly retrieve and match images, eliminating the need for manual image selection while maintaining visual distinctiveness

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system enables self-service by automatically performing image retrieval, matching, and selection based on natural language queries. The visualization tool autonomously searches through configured image sources, matches images to categorical terms using natural language processing, and integrates selected images into the visualization without requiring user intervention, thereby dramatically reducing the time spent on image selection

Inventive Principle:

Principle #25Self-service

3Ease of operation

If no automated image selection process is provided, then the system simplicity is maintained, but the user experience and visual analysis flow are hindered

Engineering Contradiction:

Improvevisual analysis flowVSAvoidautomated image selection system

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical manual process of image selection with an automated natural language processing system. Instead of users manually browsing and selecting images, the system uses computational linguistics to interpret user intent, automatically search image sources, and retrieve relevant images, thereby smoothing the visual analysis flow while managing complexity through software automation

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11210333B2Using natural language to generate data visualizations

Publication Date: 2021.12.28 TABLEAU SOFTWARE INC

AI summary

A method of visualizing data is performed at a computing device. A user selects a field in a data structure. The field has a set of field values and an associated field name. Each field value corresponds to a record in the data structure. The field values are words in a natural language. For each field value, the process builds a term set of base terms including the field value and the field name. The process retrieves a set of images from an image corpus according to the term sets. The process then selects an image from each retrieved set of images and displays data from the data structure in a data visualization. Each record in the data structure is displayed using the corresponding selected image. Each selected image is displayed, instead of the field value, at a location in the data visualization according to data in the respective record.