Website Branding Reliability Detection After Internet Scraping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distinguishing reliable websites from unauthorized ones that copy branding is challenging, and existing methods that train machine learning models on large labeled datasets are computationally intensive and prone to identifying irrelevant features, wasting resources.
Innovation Solution
Training a model on a set of guidelines associated with a first entity, such as style guides and coding standards, to detect unauthorized websites, reducing the need for extensive labeled datasets and conserving processing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning models are trained on large labeled datasets to detect unauthorized websites, then detection accuracy is improved, but computational intensity and processing resources increase significantly
Solution Approach 1:
The patent extracts only the essential branding elements (logos, colors, fonts) from the full website content for analysis. By taking out only these critical visual identifiers rather than analyzing entire websites, the system achieves accurate detection while significantly reducing computational resources required for processing.
Solution Approach 2:
The system creates simplified representations (copies) of branding elements such as extracted logo images, color palettes, and font characteristics. These copies serve as the data basis for machine learning models, enabling accurate unauthorized website detection without processing the complete original website content, thus reducing computational intensity.
2Measurement precision
If machine learning models are trained on large labeled datasets to detect unauthorized websites, then detection accuracy is improved, but processing time and resource consumption increase
Solution Approach 1:
The patent extracts only the essential branding elements (logos, colors, fonts) from the full website content for analysis. By taking out only these critical visual identifiers rather than analyzing entire websites, the system achieves accurate detection while significantly reducing processing time required for analysis.
Solution Approach 2:
The system segments website analysis into separate modules: logo detection, color analysis, and font identification. Each module processes specific branding elements independently, allowing parallel processing and reducing overall processing time while maintaining comprehensive detection accuracy through the combined results of segmented analyses.
3Reliability
If traditional machine learning models analyze all website features, then comprehensive detection is achieved, but irrelevant features waste processing resources
Solution Approach 1:
The patent applies different analysis methods to different local aspects of website branding. Logo detection uses image recognition algorithms, color analysis uses spectral detection, and font identification uses typographic feature extraction. Each local feature is processed with the most appropriate specialized method, improving detection reliability while avoiding the waste of processing resources on irrelevant features through targeted, localized analysis.
Solution Approach 2:
The system extracts and focuses analysis only on branding-related features (logos, colors, fonts) that are relevant for detecting unauthorized websites. By taking out these specific features from the broader set of all possible website attributes, the system eliminates processing waste on irrelevant data while maintaining high detection reliability through concentrated analysis of meaningful indicators.
Data Source
AI summary
In some implementations, a reliability modeler may receive a plurality of webpages associated with a first entity from an Internet scraping device. The reliability modeler may detect, within the plurality of webpages, at least one of a logo, a font, or a color. The reliability modeler may apply a model, trained on a set of guidelines associated with the first entity, to the logo, the font, or the color. Accordingly, the reliability modeler may determine, based on output from the model, that the plurality of webpages are unlikely to be authorized by the first entity. The reliability modeler may transmit, to a user device, an alert based on determining that the plurality of webpages are unlikely to be associated with the first entity.