Ransom Note Classification in Received Files for Early Ransomware Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing ransomware detection systems are slow to detect malicious code, allowing it to cause damage before proper classification, and lack effective methods to identify ransom notes quickly.

Innovation Solution

A framework that utilizes sentiment analysis on tokenized and lemmatized text data to classify files as ransomware by generating a score based on vectorized features, suspending the process if the score exceeds a threshold, and alerting the user.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If static classification and runtime analysis are used to detect ransomware, then detection accuracy is improved, but detection speed deteriorates allowing malicious code to cause damage before classification

Engineering Contradiction:
Improvedetection accuracyVSAvoiddetection speed
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The patent applies preliminary action by extracting and analyzing text content from files before they are executed. The system creates a text representation of file contents, extracts textual features, and classifies them as ransomware-related before the malicious code can cause harm. This pre-execution analysis enables early detection and neutralization of ransomware threats.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent applies extraction by separating text content analysis from traditional executable code analysis. The system extracts text representations from files, isolates textual features, and analyzes them independently using NLP techniques. This extraction approach enables rapid classification of ransomware based on characteristic text patterns without requiring full runtime execution.

Inventive Principle:
Principle #2Taking out (Extraction)

2Reliability

If traditional ransomware detection systems wait for multiple file change events to occur, then they can confirm malicious behavior, but by then malicious code has already caused damage

Engineering Contradiction:
Improvedetection reliabilityVSAvoidtime to neutralize threat
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary classification of files based on their text content before multiple file change events occur. By analyzing textual features and generating ransomware probability scores early in the file lifecycle, the system can neutralize threats before they propagate across the network or cause extensive damage.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces text content analysis as an intermediary between file reception and full execution. This intermediary layer provides early warning signals through text-based classification, allowing the system to intervene before the malicious code completes its harmful actions. The text analysis acts as a mediator that enables early detection without requiring confirmation through multiple file change events.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If sentiment analysis with tokenization and lemmatization is applied to classify files, then ransomware identification accuracy is improved, but processing complexity increases

Engineering Contradiction:
Improveransomware identification accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical code analysis methods with natural language processing techniques. Instead of analyzing binary executable code through complex disassembly and debugging processes, the system substitutes text-based NLP methods including tokenization, lemmatization, and sentiment analysis. This substitution simplifies the analysis process while maintaining high accuracy in identifying ransomware characteristic text patterns.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent applies parameter changes by transforming raw text data into standardized NLP features through tokenization and lemmatization. These parameter transformations convert unstructured text into structured numerical representations that can be efficiently processed by classification algorithms. The parameter changes enable the system to handle linguistic variations while maintaining consistent classification accuracy.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3673400B1Classifying ransom notes in received files for ransomware process detection and prevention
Publication Date: 2025.09.24 ENDGAME INC
  • EP3673400B1 patent drawingFigure 1
  • EP3673400B1 patent drawingFigure 2
  • EP3673400B1 patent drawingFigure 3

AI summary

The present invention analyzes the text of a received file to determine if the file likely is a forensic artifact of a ransomware attack on a computer system. If the computer system concludes that the file is likely an artifact of a ransomware attack, the system terminates or ignores all related processes, thereby minimizing the harm caused to the computer system.