Sketch Matrix Dimensionality Reduction for Malware Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing malware classification techniques face challenges in efficiently processing high-dimensional feature spaces of computer programs, which are computationally burdensome and impractical for large datasets, making it difficult to classify malicious vs. benign programs effectively.
Innovation Solution
The method involves determining a sketch matrix that approximates a reference dataset using a random projection matrix, reducing dimensionality by identifying approximate principal components, and generating a reduced dimension representation, allowing for more efficient classification of target computer programs based on similarity measures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional malware classification techniques process high-dimensional feature spaces of computer programs, then classification accuracy can be maintained, but computational overhead becomes excessively high and processing efficiency deteriorates
Solution Approach 1:
The patent applies dimensionality reduction techniques to transform the high-dimensional feature space into a lower-dimensional representation. By projecting features onto a reduced subspace while preserving essential variance, the system maintains classification accuracy while dramatically reducing computational overhead and improving processing efficiency.
Solution Approach 2:
The patent extracts and removes redundant or less informative features from the high-dimensional feature space. By identifying and eliminating features that contribute minimally to classification accuracy, the system reduces dimensionality while preserving the most discriminative features, thereby improving processing efficiency without sacrificing classification performance.
2Productivity
If the quantity of features in the reference dataset is reduced through dimensionality reduction, then computational overhead decreases and processing speed improves, but information loss may occur affecting classification reliability
Solution Approach 1:
The patent transforms the feature representation by changing parameters through dimensionality reduction techniques. By systematically selecting which features to retain and how to combine them in the reduced space, the method preserves the most informative aspects of the original data while reducing dimensionality, thus maintaining classification reliability while improving processing speed.
Solution Approach 2:
The patent replaces the mechanical processing of all high-dimensional features with a transformed representation in reduced space. By substituting the full feature set with a carefully constructed reduced representation that captures essential patterns, the system achieves faster processing while maintaining the reliability needed for accurate malware classification.
Data Source
AI summary
In one aspect, a computer-implemented method is disclosed. The computer-implemented method may include determining a sketch matrix that approximates a matrix representative of a reference dataset. The reference dataset may include at least one computer program having a predetermined classification. A reduced dimension representation of the reference dataset may be generated based at least on the sketch matrix. The reduced dimension representation may have a fewer quantity of features than the reference dataset. A target computer program may be classified based on the reduced dimension representation. The target computer program may be classified to determine whether the target computer program is malicious. Related systems and articles of manufacture, including computer program products, are also disclosed.