摘要
大数据环境中网络安全问题日益严重,传统漏洞检测技术面临极大挑战。文章提出面向大数据环境的网络信息安全漏洞检测技术,设计多阶段数据采集与处理框架,结合深度学习与图神经网络提取关键特征,利用无监督与有监督学习方法检测数据中的异常行为,采用滑动窗口算法触发反馈机制实现自动修复和响应。实验表明,该技术在准确率、误报率和检测时间上相较于传统技术优势显著。
The network security problem in the big data environment is becoming increasingly serious,and the traditional vulnerability detection technology is facing great challenges.A network information security vulnerability detection technology for the big data environment is proposed.A multi-stage data acquisition and processing framework is designed.Deep learning and graph neural network are combined to extract key features.Unsupervised and supervised learning methods are used to detect abnormal behavior in data.The sliding window algorithm is used to trigger the feedback mechanism to achieve automatic repair and response.Experiments show that this technology has significant advantages over traditional technologies in terms of accuracy,false positive rate and detection time.
作者
张晓鲁
ZHANG Xiaolu(Digital Operation and Maintenance Center of the Fourth Oil Production Plant,Daqing Heilongjiang 163511,China)
出处
《信息与电脑》
2025年第12期72-74,共3页
Information & Computer
关键词
大数据
漏洞检测
异常检测
深度学习
实时监控
big data
vulnerability detection
anomaly detection
deep learning
real-time monitoring
