Network Address Trust Evaluation via Derivative Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for preventing phishing attacks, such as blacklists, additional website information, and email filtering, are either only effective against known threats, overwhelm users with information, or are limited to email-borne attacks, failing to provide comprehensive protection against unknown or non-email phishing attacks.
Innovation Solution
A method that evaluates network addresses by generating derivatives through Optical Character Recognition, structural modifications, and error compensation, assigning trust levels based on validity, rank, host location, creation date, and character set, and comparing these to detect untrustworthy addresses automatically, providing a secure and flexible solution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If blacklists created from user reports are used to prevent phishing attacks, then known phishing attacks can be blocked, but the method is only effective against known threats and cannot detect unknown phishing attacks
Solution Approach 1:
The system performs preliminary analysis of network addresses by generating derivatives and assigning trust levels before actual phishing attempts occur. This proactive approach enables the detection of unknown phishing attacks by evaluating trustworthiness in advance, rather than relying on post-hoc blacklists.
Solution Approach 2:
The patent introduces an intermediary evaluation system that acts as a mediator between the user and network addresses. This intermediary automatically assesses trust levels by comparing addresses against their derivatives, providing a bridge that enhances both protection against known threats and detection of unknown threats without requiring direct user interaction.
2Reliability
If additional information about websites (hosting location, registration date, rank) is displayed to prevent phishing, then users can make informed decisions, but the method may overwhelm users with information and cause false positives with less popular websites
Solution Approach 1:
The system extracts only the essential trust evaluation results from complex website information, presenting a simplified trust level indicator to users. This extraction approach maintains user awareness of trustworthiness while avoiding information overload by filtering out unnecessary details such as hosting location and registration date.
Solution Approach 2:
The patent applies different levels of information presentation based on local context - providing detailed trust evaluation metrics for high-risk addresses while offering simplified indicators for low-risk addresses. This localized approach prevents false positives by adapting the information display to the specific risk level of each website.
3Reliability
If e-mail filtering using Bayesian filters is used to prevent phishing, then email-borne phishing can be filtered or reformatted, but the method is only effective against phishing disseminated by email and not other delivery methods
Solution Approach 1:
The patent creates a universal trust evaluation system that functions across multiple delivery methods including email, direct URL access, and other communication channels. By generating derivatives and comparing trust levels, the system provides multi-functional protection that adapts to various phishing delivery methods rather than being limited to email filtering.
Data Source
AI summary
The invention relates to a method for evaluating or accessing a network address, comprising the steps of: receiving a network address (50); generating derivatives (60) of the received network address (50); assigning a trust level to the generated derivatives (60) and the received network address (50); comparing the trust levels of the derivatives (60) with the trust level of the received network address (50); and issuing a response based on the comparison.