Domain Name Filtering With Dynamic Thresholds for Trademark Screening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The sheer number of domains on the Internet makes it difficult and resource-intensive to identify potential trademark infringers, as existing methods like n-gram distance analysis are inefficient and require significant computational power.
Innovation Solution
A domain filter system that computes a similarity score and dynamic threshold between seed and candidate domains, reducing the number of domains requiring further processing by using a linear complexity algorithm, thus improving efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If n-gram distance analysis is used to identify potential trademark infringers, then the identification can be performed, but the computational power required is significant and processing efficiency is low
Solution Approach 1:
The patent segments the domain name analysis process into multiple filtering stages: initial filtering based on domain structure patterns, intermediate filtering using n-gram distance for candidate domains, and final verification. This segmentation allows the system to process hundreds of millions of domains by applying computationally expensive n-gram analysis only to a small subset of candidate domains identified through lighter initial filters.
Solution Approach 2:
The patent performs preliminary filtering actions before applying the computationally intensive n-gram distance analysis. It pre-identifies candidate domains using simpler criteria such as domain structure patterns, length constraints, and basic string matching. This preliminary action reduces the dataset from hundreds of millions of domains to a manageable subset, enabling efficient trademark infringement detection.
2Reliability
If all hundreds of millions of domains are processed to find trademark infringers, then comprehensive detection is achieved, but the time and computational resources required are excessive
Solution Approach 1:
The patent applies different levels of analysis quality to different domains based on their characteristics. High-quality n-gram distance analysis is applied locally to candidate domains that match specific patterns, while the majority of domains receive lighter initial filtering. This local quality approach ensures reliable detection for potential infringers while minimizing overall processing time.
Solution Approach 2:
The patent performs partial action by focusing computational resources on a subset of domains that are most likely to be infringers. Rather than exhaustively analyzing all hundreds of millions of domains, it applies comprehensive analysis only to candidate domains that pass initial filtering criteria, achieving reliable detection with significantly reduced time and resources.
3Productivity
If a simple filtering method is used to reduce domains for processing, then processing speed increases, but the accuracy of identifying potential infringers decreases
Solution Approach 1:
The patent segments the filtering process into multiple stages with increasing precision. The first stage uses simple, fast filters to reduce the domain set from hundreds of millions to a manageable subset. The second stage applies n-gram distance analysis to this reduced set with high precision. This segmentation allows the system to achieve both high filtering speed and accurate infringer identification.
Solution Approach 2:
The patent introduces an intermediary filtering stage between the initial domain set and the final n-gram analysis. This intermediary stage uses pattern matching and structural analysis to identify candidate domains that warrant further investigation. It acts as a mediator that preserves potentially infringing domains while filtering out obvious non-infringers, maintaining accuracy while enabling efficient processing.
Data Source
AI summary
A domain processing system is enhanced with a first-pass domain filter configured for loading character strings representing a pair of domains consisting of a seed domain and a candidate domain in a computer memory, computing a similarity score and a dynamic threshold for the pair of domains, determining whether the similarity score exceeds the dynamic threshold, and iterating the loading, the computing, and the determining for each of a plurality of candidate domains paired with the seed domain. A similarity score between the seed domain and the candidate domain and a corresponding dynamic threshold for the pair are computed. If the similarity score exceeds the corresponding dynamic threshold, the candidate domain is provided to a downstream computing facility. Otherwise, it is dropped. In this way, the first-pass domain filter can significantly reduce the number of domains that otherwise would need to be processed by the downstream computing facility.