微型电脑应用2023，Vol.39Issue(12)：62-65,4.
一种基于BLCNA模型的恶意URL检测技术
A Malicious URL Detection Technology Based on BLCNA Model
摘要
Abstract
Aiming at the low accuracy and time-consuming of traditional methods for identifying malicious URLs,this paper proposes a combined neural network model(BLCNA)based on attention mechanism to detect malicious URLs.It extracts the semantic information and visual information of the URL for encoding processing.Combines the bidirectional long short-term memory network(BiLSTM)and the capsule network(CapsNet)to build a neural network joint model to capture semantic and visual features simultaneously.The attention mechanism is used to increase the key feature weights.The classification of the URL is completed based on the valid features.The experimental results show that the proposed method is superior to other methods in detecting malicious URLs,and the accuracy rate can reach 99.79％.关键词
电力网络安全/URL/注意力机制/特征提取/神经网络Key words
power network security/URL/attention mechanism/feature extraction/neural network分类
信息技术与安全科学引用本文复制引用
沈伍强,张金波,许明杰,杨春松..一种基于BLCNA模型的恶意URL检测技术[J].微型电脑应用,2023,39(12):62-65,4.基金项目
南方电网公司科技项目资助(037800KK52190012) （037800KK52190012）
