深度学习在恶意软件分析中的应用研究  

Research on application of deep learning in malware analysis

作　　者：吴晓梅 WU Xiaomei(Vocational Secondary School in Gaotai County,Zhangye,Gansu 734300,China)

机构地区：[1]甘肃省高台县职业中等专业学校,甘肃张掖734300

出　　处：《计算机应用文摘》2023年第21期40-42,共3页Chinese Journal of Computer Application

摘　　要：文章研究了基于长短时记忆的恶意软件检测系统。其中,通过引入LSTM模型,针对Windows系统开发了一种恶意软件检测系统,并使用公开数据集进行了测试。该系统结构包括LSTM层、全连接层和Softmax层,可用于学习序列模式和实现恶意软件分类。实验结果表明,该系统在准确率、精确率和F1值等评估指标方面表现出良好的性能,具有实际应用的潜力。This article studies a malware detection system based on long-term and short-term memory.By introducing the LSTM model,a malware detection system was developed for Windows systems and tested using public datasets.The system structure includes an LSTM layer,a fully connected layer,and a Softmax layer for learning sequence patterns and implementing malware classification.The experimental results show that the system has achieved good performance in evaluation indicators such as accuracy,accuracy,and F value,and has the potential for practical application.

关 键 词：深度学习 软件分析 WINDOWS系统 LSTM 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]