Malware Detection via Hierarchical URL Cache Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing malware detection systems face challenges in balancing accuracy and efficiency, often resulting in high resource consumption and false positives, as they struggle to keep pace with rapidly evolving malware websites.

Innovation Solution

A method and system that utilize a multilevel cache to analyze network requests by parsing URLs, evaluating fully qualified domain names and parent hostnames for suspiciousness, and flagging machines with a likelihood of infection based on cache presence and scoring mechanisms, reducing false positives and resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If malware detection systems check network requests against websites known to distribute malware, then detection accuracy is improved, but the system cannot keep pace with rapidly generated new malware websites

Engineering Contradiction:
Improvedetection accuracyVSAvoidability to detect new malware websites
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system segments the URL analysis into multiple hierarchical levels: first checking the complete URL against a cache of known malicious URLs, then extracting and checking the fully qualified domain name (FQDN) against a separate cache, and finally evaluating the parent hostname if previous checks fail. This segmentation allows the system to handle both known malicious sites and newly generated malware websites efficiently.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by maintaining pre-populated caches of suspicious URLs and domain names before actual malware detection is needed. These caches are built in advance through monitoring and analysis, enabling rapid comparison against incoming network requests without requiring real-time computation for each check.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If malware detection methods consume system resources such as memory, communication bandwidth, and processing bandwidth to increase detection accuracy, then detection capability is improved, but resource consumption increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidsystem resource consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system applies partial action by not performing complete analysis on every URL. Instead, it uses a hierarchical approach where most URLs are quickly resolved through cache lookups of complete URLs or FQDNs. Only when these fail does the system perform the more resource-intensive parent hostname evaluation, applying excessive action only where necessary to maintain detection accuracy while minimizing overall resource consumption.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The system uses caching to store copies of previously analyzed suspicious URLs and domain names. These cached copies allow rapid comparison against new network requests without repeating the full analysis process, significantly reducing processing bandwidth and memory usage for detecting patterns in malware website generation.

Inventive Principle:
Principle #26Copying

3Reliability

If malware detection systems use comprehensive analysis methods, then detection accuracy is improved, but the rate of false positives increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidfalse positive rate
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The system segments the detection process into hierarchical levels with increasing stringency. The first level checks complete URLs against a cache, the second level checks extracted FQDNs, and the third level evaluates parent hostnames. This segmentation allows the system to achieve high detection accuracy through multiple checks while reducing false positives by requiring evidence at multiple levels before flagging a URL as malicious.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3123668B1A system to identify machines infected by malware applying linguistic analysis to network requests from endpoints
Publication Date: 2021.01.13 CA TECH INC
  • EP3123668B1 patent drawingFigure 1~2
  • EP3123668B1 patent drawingFigure 3A~3C
  • EP3123668B1 patent drawingFigure 4A

AI summary

A method to identify machines infected by malware is provided. The method includes determining whether a universal resource locator in a network request is present in a first cache and determining whether a fully qualified domain name from the uniform resource locator is present in a second cache. The method includes evaluating a parent hostname as to suspiciousness. The method includes indicating the computing device has a likelihood of infection, responsive to one of: the universal resource locator being present in the first cache with a first indication of suspiciousness, the fully qualified domain name being present in the second cache with a second indication of suspiciousness, or the evaluating the parent hostname having a third indication of suspiciousness, wherein at least one method operation is performed by the processor. A system and computer readable media are provided.