Neural Network Reference Image Generation for URL Memorability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current URL shortening techniques consume significant computing and networking resources, generate confusing URLs with visually similar characters, and are prone to phishing attacks due to their lack of memorability.
Innovation Solution
A neural network model is used to generate a reference image based on a combination of images that satisfies a memorability score threshold, which is then used to help users remember complex data such as URLs, telephone numbers, or textual information, thereby reducing the need for resource-intensive URL shortening and mitigating phishing risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of moving object
If URL shortening techniques are used to reduce the length of complex data, then the length of the data is reduced, but computing and networking resources are consumed significantly and the shortened URLs are confusing with visually similar characters
Solution Approach 1:
The patent creates visual copies (images) of complex data instead of creating shortened text representations. The neural network generates images that visually represent the complex data (URLs, phone numbers, etc.), allowing users to recognize and remember the data through visual patterns rather than memorizing shortened character sequences. This eliminates the need for resource-intensive URL shortening services while avoiding confusion from visually similar characters.
Solution Approach 2:
The patent replaces the mechanical text-based URL shortening system with a neural network-based image generation system. Instead of using algorithms to generate shortened text strings that consume computing resources and create confusion, the system uses machine learning models to generate visual representations that are more memorable and less resource-intensive to process and store.
2Device complexity
If URL shortening techniques are used to simplify complex data, then the complexity is reduced, but the shortened URLs are prone to phishing attacks due to lack of memorability
Solution Approach 1:
The patent utilizes visual characteristics including color, shape, and pattern variations in generated images to create distinctive visual representations of complex data. The neural network generates images with unique visual features that make each representation easily distinguishable and memorable, preventing users from being tricked by phishing attacks that rely on creating visually similar but malicious shortened URLs.
3Length of moving object
If traditional URL shortening methods are used, then the URL length is reduced, but user memorability and safety are compromised
Solution Approach 1:
The patent transitions from one-dimensional text-based URL shortening to two-dimensional image-based representations. Instead of compressing URLs into shorter text strings, the system generates visual images that encode the complex data in spatial and visual dimensions, making the data more memorable and easier for users to recognize and recall without needing to memorize character sequences.
Data Source
AI summary
A device may receive complex data from a user device and may provide multiple images to the user device based on receiving the complex data. The device may receive, from the user device, a selection of two or more images from the multiple images, and may determine whether a combination of the two or more images is stored in a data structure. The device may determine a mapping of information identifying the two or more images with the complex data, based on the combination of the two or more images not being stored in the data structure, and may store the information identifying the two or more images, the complex data, and the mapping in the data structure. The device may process the two or more images to generate a reference image that satisfies a memorability score threshold and may provide the reference image to another user device.