# Malware Classification Using Static Disassembly and Machine Learning

arXiv: https://arxiv.org/abs/2201.07649 · HTML: https://mlsanity.com/paper/2201.07649/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 16 references that are in the map.
- First posted on arXiv 10 Dec 2021.

## Builds on (most cited first, 3 of 3)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Malware Detection by Eating a Whole EXE](https://mlsanity.com/paper/1710.09435/) (2017)
- [Microsoft Malware Classification Challenge](https://mlsanity.com/paper/1802.10135/) (2018)

## Built on by (most cited first, 1 of 1)

- [Obfuscated Malware Detection: Investigating Real-world Scenarios through Memory Analysis](https://mlsanity.com/paper/2404.02372/) (2024)

## Abstract (arXiv)

Network and system security are incredibly critical issues now. Due to the rapid proliferation of malware, traditional analysis methods struggle with enormous samples. In this paper, we propose four easy-to-extract and small-scale features, including sizes and permissions of Windows PE sections, content complexity, and import libraries, to classify malware families, and use automatic machine learning to search for the best model and hyper-parameters for each feature and their combinations. Compared with detailed behavior-related features like API sequences, proposed features provide macroscopic information about malware. The analysis is based on static disassembly scripts and hexadecimal machine code. Unlike dynamic behavior analysis, static analysis is resource-efficient and offers complete code coverage, but is vulnerable to code obfuscation and encryption. The results demonstrate that features which work well in dynamic analysis are not necessarily effective when applied to static analysis. For instance, API 4-grams only achieve 57.96% accuracy and involve a relatively high dimensional feature set (5000 dimensions). In contrast, the novel proposed features together with a classical machine learning algorithm (Random Forest) presents very good accuracy at 99.40% and the feature vector is of much smaller dimension (40 dimensions). We demonstrate the effectiveness of this approach through integration in IDA Pro, which also facilitates the collection of new training samples and subsequent model retraining.