Harmful Site Detection via Link Circulation Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing harmful site blocking techniques rely heavily on manual maintenance of databases, which is inefficient due to the constant emergence of new harmful sites and frequent changes in website content and addresses, making real-time analysis cumbersome and time-consuming.

Innovation Solution

A device and method that analyzes connections between harmful sites, normalizes URLs, and uses a database to store both harmful and non-harmful site URLs, allowing for efficient determination of harmful sites by creating a directed graph and calculating a harmful site rank based on statistical analysis, while removing non-harmful site URLs from consideration to reduce computation and enhance efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual maintenance of harmful site database is performed, then the database can be updated, but it is time-consuming and inefficient due to constant emergence of new harmful sites

Engineering Contradiction:
Improveaccuracy of harmful site identificationVSAvoidtime for database maintenance
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system automatically identifies and updates harmful sites through web crawling and link circulation analysis without requiring manual intervention. The harmful site collection device autonomously navigates web pages, extracts links, analyzes circulation patterns, and updates the database automatically, making the system self-maintaining and eliminating time-consuming manual updates while maintaining high accuracy through automated analysis algorithms

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system performs preliminary web crawling and link extraction to build a database of potential harmful sites before actual harmful site identification is needed. By pre-collecting and organizing link information, the system prepares data structures in advance that enable rapid automated analysis when new harmful sites emerge, reducing the time required for both database maintenance and real-time identification

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If real-time analysis of website contents is performed, then harmful sites can be identified accurately, but the process is cumbersome and time-consuming

Engineering Contradiction:
Improveaccuracy of harmful site detectionVSAvoidspeed of harmful site identification
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system performs preliminary web crawling and link extraction to pre-process and organize web page information before actual harmful site analysis. By preparing data structures, extracting links in advance, and organizing circulation patterns beforehand, the system reduces the computational burden during real-time analysis, enabling both high accuracy through thorough analysis and high speed through optimized data structures

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system divides the harmful site identification process into separate modules: web crawling, link extraction, circulation analysis, and harmful site determination. Each module processes specific aspects independently, allowing parallel execution and optimized computation. This segmentation enables the system to maintain high accuracy through comprehensive analysis while improving productivity through efficient modular processing and reduced computational overhead

Inventive Principle:
Principle #1Segmentation

3Reliability

If all extracted URLs are analyzed for harmful sites, then no harmful sites are missed, but computational load increases significantly

Engineering Contradiction:
Improvecompleteness of harmful site detectionVSAvoidcomputational resources consumed
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system extracts and removes non-harmful site URLs from the analysis pool by comparing extracted URLs against a pre-stored non-harmful site database. By filtering out known safe sites before进行深入分析, the system significantly reduces the number of URLs requiring computational analysis while ensuring no harmful sites are missed, as the filtering only removes previously verified non-harmful sites

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs partial analysis by focusing computational resources only on URLs that pass the initial non-harmful site filter and exhibit suspicious circulation patterns. Rather than analyzing every extracted URL equally, the system applies analysis selectively to URLs that require it based on their circulation characteristics, reducing overall computational load while maintaining complete detection of harmful sites through the circulation analysis mechanism

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2937801B1Harmful site collection device and method
Publication Date: 2021.03.31 FOUND OF SOONGSIL UNIV IND COOP
  • EP2937801B1 patent drawingFigure 1
  • EP2937801B1 patent drawingFigure 2
  • EP2937801B1 patent drawingFigure 3

AI summary

Provided are a harmful site collection device and method for determining a harmful site by analyzing a connection between harmful sites. The harmful site collection device extracts a URL linked to a web page of a harmful site; checks a link circulation on the basis of link information on a web page of the URL linked to the harmful site to determine whether the web page of the URL linked to the harmful site is a harmful site; and, when a URL of a prestored non-harmful site is extracted while the link circulation is checked, stops checking the link circulation that includes the URL of the non-harmful site. Accordingly, the harmful site collection device can more easily determine a harmful site merely with information on a URL linked to a web page and can reduce the amount of computation using information on a URL of a prestored non-harmful site.