URL Pixel Similarity for Preemptive Lookalike Domain Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anti-phishing solutions are inadequate in preemptively mitigating the risks posed by lookalike domains, as they rely on reactive measures and do not generate accurate similarity scores to identify potential phishing threats effectively.
Innovation Solution
The use of a genetic algorithm to generate lookalike domains based on graphical similarity pixel comparison, assigning penalty values to deception methods, and iteratively refining these domains through multiple generations to enhance accuracy and efficiency in identifying potential phishing threats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional anti-phishing solutions are used, then the system is simple to operate, but the ability to preemptively identify lookalike domains is insufficient
Solution Approach 1:
The system performs preliminary actions by proactively generating lookalike domains before actual phishing attacks occur. The genetic algorithm creates multiple generations of potential phishing domains in advance, enabling the system to block these domains preemptively rather than reactively, thus improving reliability while managing complexity through automated generation processes
Solution Approach 2:
The patent replaces manual or simple rule-based anti-phishing mechanisms with an automated genetic algorithm system. This substitution introduces complexity but enables the system to handle the combinatorial explosion of possible lookalike domains through evolutionary computation, achieving better detection capability without requiring manual configuration of each potential threat
2Measurement precision
If graphical similarity pixel comparison is used, then the measurement precision of domain similarity is improved, but the computational complexity increases
Solution Approach 1:
The system creates visual copies or representations of domain names as images, then compares these image representations using pixel analysis. This copying approach allows the genetic algorithm to evaluate graphical similarity of potentially millions of generated domains without performing complex character-by-character analysis on each one, thus improving measurement precision while managing computational complexity through efficient image-based comparison
Solution Approach 2:
The patent changes the parameter space from character-level domain analysis to pixel-level image analysis. By converting domains to visual representations and comparing them based on pixel similarity metrics, the system achieves more accurate graphical similarity measurement while leveraging efficient image processing algorithms to handle the computational load
3Adaptability or versatility
If multiple deception methods are applied, then the coverage of phishing detection is improved, but the quantity of generated lookalike domains increases
Solution Approach 1:
The system segments the domain generation process into multiple generations, where each generation applies specific deception methods to create variations. By dividing the overall generation task into manageable generations and using penalty-based filtering to eliminate low-quality candidates at each stage, the system achieves comprehensive coverage through multiple deception techniques while controlling the total quantity of domains that need to be evaluated
4Manufacturing precision
If penalty values are assigned to each deception method, then the manufacturing precision of lookalike domains is improved, but the loss of time in generation process increases
Solution Approach 1:
The system implements feedback mechanisms by assigning penalty values to deception methods and using these penalties to guide the selection of parent domains for the next generation. This feedback loop allows the genetic algorithm to learn from previous generations and converge faster on high-quality lookalike domains, improving manufacturing precision while reducing the time required through intelligent search guidance rather than exhaustive evaluation
Data Source
AI summary
Systems and methods for generating and utilizing lookalike Uniform Resource Locators (URLs) based on a graphical comparison include receiving an original target domain and a lookalike domain, converting the original target domain and lookalike domain into pixelated images, calculating a similarity based on the images of the original target domain and the lookalike domain, and calculating a percentage difference of the images of the original target domain and the lookalike domain.