Scholarlink

Effective and scalable detection techniques are crucial given the sharp rise in malware that targets Internet of Things (IoT) devices. An opcode-based malware detection system utilizing Long Short-Term Memory (LSTM) networks is presented in this study. It uses Word2Vec Continuous Bag of Words (CBOW) model to create meaningful opcode embeddings by interpreting opcode sequences as equivalent to natural language data. A two-stage LSTM network is then used to process these embeddings in order to differentiate between malware and benign files. A high classification accuracy is achieved through extensive testing with various word window sizes to maximize the embedding quality. Our framework also addresses the increasing security issues in resource-constrained situations because it is lightweight, making it appropriate for deployment on IoT devices.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Opcode-Based Malware Detection Using LSTM Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Katyayni Upadhyay,
  • Neelam Dayal

摘要

Effective and scalable detection techniques are crucial given the sharp rise in malware that targets Internet of Things (IoT) devices. An opcode-based malware detection system utilizing Long Short-Term Memory (LSTM) networks is presented in this study. It uses Word2Vec Continuous Bag of Words (CBOW) model to create meaningful opcode embeddings by interpreting opcode sequences as equivalent to natural language data. A two-stage LSTM network is then used to process these embeddings in order to differentiate between malware and benign files. A high classification accuracy is achieved through extensive testing with various word window sizes to maximize the embedding quality. Our framework also addresses the increasing security issues in resource-constrained situations because it is lightweight, making it appropriate for deployment on IoT devices.