A research team has designed a decentralized detection framework that merges locality-sensitive hashing with blockchain technology to identify polymorphic malware from memory-forensic artifacts. Documented by researchers Hatem Mosa and Mohammad Azzeh in the journal Cluster Computing, the investigation demonstrates how distributed ledgers can be combined with similarity-preserving algorithms to recognize evasive malicious software. Experimental evaluation of the integrated framework demonstrated that implementing locality-sensitive hashing directly within a decentralized blockchain network achieved 97 percent precision and 97 percent recall, while simultaneously eliminating unknown classification misses.
The study was motivated by the continuous technical evolution of polymorphic and metamorphic malware, whose evasion techniques have caused considerable operational and economic losses across diverse computing domains. To analyze these evasive behaviors under controlled conditions, the researchers utilized the CIC-MalMem-2022 malware dataset, evaluating a total corpus of 58,596 records by extracting memory-forensic artifacts directly from malware memory dumps for comparative evaluation.
Using these forensic memory artifacts, the authors implemented and compared locality-sensitive hashing techniques to establish how reliably similarity algorithms could separate malicious variants. In direct comparative evaluation, the MinHash algorithm exhibited performance superior to that of the SimHash algorithm across evaluated parameters. When tuned to an optimized threshold value of 0.5, which yielded the most favorable analytical results, MinHash achieved 99 percent precision and 97 percent recall. In this standalone configuration, MinHash recorded a collision rate of 98.30 percent, accompanied by an initial locality-sensitive hashing miss rate of 1.70 percent representing unknown classifications.
Following the initial algorithmic comparison, the investigators executed the subsequent phase of their design by integrating locality-sensitive hashing directly into a blockchain network to establish a fully decentralized detection framework. Within this distributed blockchain environment, the unified framework demonstrated robust diagnostic capability, sustaining 97 percent precision and 97 percent recall across the sample records. Furthermore, the collaborative decentralized structure successfully lowered the locality-sensitive hashing miss rate from the baseline of 1.70 percent to exactly 0 percent, confirming that no malware samples within the network remained as unknown classifications.
Subject of Research: Decentralized polymorphic malware detection using locality-sensitive hashing and blockchain
Article Title: A decentralized framework for polymorphic malware detection using locality-sensitive hashing and blockchain
Article References: Mosa, H., & Azzeh, M. (2027). A decentralized framework for polymorphic malware detection using locality-sensitive hashing and blockchain. Cluster Computing, 30(1), Article 54. https://doi.org/10.1007/s10586-026-06783-6
Image Credits: AI Generated
DOI: 10.1007/s10586-026-06783-6
Keywords: polymorphic malware, locality-sensitive hashing, blockchain, MinHash, SimHash, CIC-MalMem-2022, memory forensics, malware detection, decentralized framework, memory dumps, obfuscation, evasion techniques, precision and recall, collision rate, cybersecurity, metamorphic malware
Cite Scienmag News
Denise Maddox. (November 20, 2026). Decentralized Blockchain Framework Detects Polymorphic Malware With Zero Miss Rate. Scienmag. https://scienmag.com/decentralized-blockchain-framework-detects-polymorphic-malware-with-zero-miss-rate/
Denise Maddox. "Decentralized Blockchain Framework Detects Polymorphic Malware With Zero Miss Rate." Scienmag, 20 November 2026, https://scienmag.com/decentralized-blockchain-framework-detects-polymorphic-malware-with-zero-miss-rate/. Accessed 20 November 2026.
Denise Maddox. "Decentralized Blockchain Framework Detects Polymorphic Malware With Zero Miss Rate." Scienmag. November 20, 2026. https://scienmag.com/decentralized-blockchain-framework-detects-polymorphic-malware-with-zero-miss-rate/








