Peiser, Stefan CarlStefan CarlPeiserFriborg, LudwigLudwigFriborgScandariato, RiccardoRiccardoScandariato2021-09-022021-09-022020-09International Conference on Information Security and Privacy Protection (SEC 2020)http://hdl.handle.net/11420/10253In this paper, we explore the idea of using locality sensitive hashes as input features to a feed-forward neural network with the goal of detecting JavaScript malware through static analysis. An experiment is conducted using a dataset containing 1.5M evenly distributed benign and malicious samples provided by the anti-malware company Cyren. Four different locality sensitive hashing algorithms are tested and evaluated: Nilsimsa, ssdeep, TLSH, and SDHASH. The results show a high prediction accuracy, as well as low false positive and negative rates. These results show that LSH based neural networks are a competitive option against other state-of-the-art JavaScript malware classification solutions.en1868-4238IFIP advances in information and communication technology2020143154JavaScriptLSHMalwareNeural networkMLE@TUHHJavaScript Malware Detection Using Locality Sensitive HashingConference Paper10.1007/978-3-030-58201-2_10Conference Paper