Phishing website detection based on two-way feature extraction and adaptive fusion

Guo Kai; Wang Heng; Ma Ziqiang

doi:10.11959/j.issn.2096-109x.2025076

您当前的位置：

首页 >

文章列表页 >

Phishing website detection based on two-way feature extraction and adaptive fusion

Papers | 更新时间：2026-05-26

- Phishing website detection based on two-way feature extraction and adaptive fusion
- Chinese Journal of Network and Information Security Vol. 12, Issue 1, Pages: 88-103(2026)
- 作者机构：
  
  宁夏大学信息工程学院，宁夏银川 750021
- 作者简介：
- 基金信息：
  
  The National Natural Science Foundation of China(62463025);General Project of Key Research and Development Program of Ningxia Hui Autonomous Region(2022BDE03008)
- DOI：10.11959/j.issn.2096-109x.2025076
  CLC： TP393.08
- Received：05 March 2025，
  
  Revised：2025-12-21，
  
  Published：25 February 2026
- 稿件说明：
移动端阅览
郭凯, 王恒, 马自强. 基于双路特征提取和自适应融合的钓鱼网站检测[J]. 网络与信息安全学报, 2026, 12(1): 88-103.

Guo Kai, Wang Heng, Ma Ziqiang. Phishing website detection based on two-way feature extraction and adaptive fusion[J]. Chinese Journal of Network and Information Security, 2026, 12(1): 88-103.
郭凯, 王恒, 马自强. 基于双路特征提取和自适应融合的钓鱼网站检测[J]. 网络与信息安全学报, 2026, 12(1): 88-103. DOI： 10.11959/j.issn.2096-109x.2025076.

Guo Kai, Wang Heng, Ma Ziqiang. Phishing website detection based on two-way feature extraction and adaptive fusion[J]. Chinese Journal of Network and Information Security, 2026, 12(1): 88-103. DOI： 10.11959/j.issn.2096-109x.2025076.

摘要

随着网络钓鱼攻击手段的不断演进，传统检测方法在特征提取、模型泛化能力等方面面临严峻挑战。针对这些问题，提出了一种基于双路特征提取与自适应融合的钓鱼网站检测模型（DPFEAT）。首先，构建了时序与空间双路特征提取机制，通过引入特征交互层增强整体特征的表征能力；其次，设计了自适应融合模块，用于实现多维特征的动态融合，从而有效提升模型对复杂攻击特征的捕获能力。在训练策略上，采用预热－余弦衰减和循环学习率调度机制，显著增强了训练过程的稳定性。同时，结合Focal Loss与Dice Loss构建组合损失函数，并引入动态权重平衡机制，有效缓解了类别不平衡与边界样本识别困难的问题。此外，还设计了可解释性分析框架，用于评估各模块对模型性能的贡献。实验结果表明，DPFEAT在准确率与受试者工作特征曲线下面积（AUC-ROC）值上分别达到95.41%和98.83%，显著优于现有方法。

Abstract

With the continuous evolution of phishing attack methods

traditional detection approaches face severe challenges in terms of feature extraction and model generalization capability. To address these issues

a phishing detection model named DPFEAT (phishing detection model with dual-path feature extraction and adaptive fusion)

which is based on dual-path feature extraction and adaptive fusion

was proposed. Firstly

a dual-path feature extraction mechanism for temporal and spatial features was constructed

and a feature interaction layer was introduced to enhance the overall feature representation capability. Secondly

an adaptive fusion module was designed to achieve dynamic fusion of multi-dimensional features

thereby effectively improving the model's ability to capture complex attack characteristics. Regarding the training strategy

a warmup-cosine decay and cyclical learning rate scheduling mechanism was adopted

which significantly improved the stability of the training process. Simultaneously

a combined loss function was constructed by integrating Focal Loss and Dice Loss

and a dynamic weight balancing mechanism was introduced

which effectively alleviated the issues of class imbalance and the difficulty in identifying boundary samples. Furthermore

an interpretability analysis framework was designed to evaluate the contribution of each module to the model's performance. Experimental results demonstrated that the proposed DPFEAT model achieved an accuracy of 95.41% and an receiver operating characteristicarea under the curve (AUC-ROC) score of 98.83%

which is significantly superior to existing methods.

关键词

Keywords

references

朱世起 , 努尔布力 . 钓鱼网站检测研究现状与发展趋势的计量分析 [J ] . 计算机工程与应用 , 2020 , 56 ( 15 ): 92 - 100 .

Zhu S Q , Nurbol . Bibliometric analysis of current studies and developing trends on phishing sites detection [J ] . Computer Engineering and Applications , 2020 , 56 ( 15 ): 92 - 100 .

蔡召 , 荆涛 , 任爽 . 以太坊钓鱼诈骗检测技术综述 [J ] . 网络与信息安全学报 , 2023 , 9 ( 2 ): 21 - 32 .

Cai Z , Jing T , Ren S . Survey on Ethereum phishing detection technology [J ] . Chinese Journal of Network and Information Security , 2023 , 9 ( 2 ): 21 - 32 .

王雨琪 , 刘博文 , 林果园 . 基于URL语言特征的钓鱼网站检测算法 [J ] . 计算机工程与应用 , 2019 , 55 ( 24 ): 84 - 90 .

Wang Y Q , Liu B W , Lin G Y . Phishing detection algorithm based on language features of URL [J ] . Computer Engineering and Applications , 2019 , 55 ( 24 ): 84 - 90 .

顾晓清 , 王洪元 , 倪彤光 , 等 . 基于贝叶斯和支持向量机的钓鱼网站检测方法 [J ] . 计算机工程与应用 , 2015 , 51 ( 4 ): 87 - 90, 95 .

Gu X Q , Wang H Y , Ni T G , et al . Phishing detection approach based on Bayes and support vector machine [J ] . Computer Engineering and Applications , 2015 , 51 ( 4 ): 87 - 90, 95 .

Zhou J X , Cui H B , Li X N , et al . A novel phishing website detection model based on LightGBM and domain name features [J ] . Symmetry , 2023 , 15 ( 1 ): 180 .

Huang Y . Phishing detection using convolutional neural networks and feature engineering [J ] . IEEE Transactions on Information Forensics and Security , 2018 , 13 ( 10 ): 2547 - 2558 .

Wang L , Liu Y . Phishing detection using LSTM and feature engineering [J ] . IEEE Transactions on Information Forensics and Security , 2020 , 15 ( 9 ): 2041 - 2050

文伟平 , 朱一帆 , 吕子晗 , 等 . 针对品牌的网络钓鱼扩线与检测方案 [J ] . 信息网络安全 , 2023 , 23 ( 12 ): 1 - 9 .

Wen W P , Zhu Y F , Lyu Z H , et al . Brand-specific phishing expansion and detection solutions [J ] . Netinfo Security , 2023 , 23 ( 12 ): 1 - 9 .

弋晓洋 , 张健 . 基于图像的网络钓鱼邮件检测方法研究 [J ] . 信息网络安全 , 2021 , 21 ( 9 ): 52 - 58 .

Yi X Y , Zhang J . Image-based phishing email detection method and implementation [J ] . Netinfo Security , 2021 , 21 ( 9 ): 52 - 58 .

张桥 , 卜佑军 , 陈博 , 等 . 一种基于MPAN的钓鱼URL检测方法 [J ] . 信息工程大学学报 , 2021 , 22 ( 4 ): 443 - 449 .

Zhang Q , Bu Y J , Chen B , et al . Phishing URL detection method based on MPAN [J ] . Journal of Information Engineering University , 2021 , 22 ( 4 ): 443 - 449 .

Ma C Y , Wu A M , Ma W Z , et al . Malicious URL recognition based on multi-feature fusion and machine learning [C ] // Proceedings of the 2022 41st Chinese Control Conference (CCC) . Piscataway : IEEE Press , 2022 : 3014 - 3019 .

张思睿 , 延志伟 , 董科军 , 等 . 基于注意力机制与特征融合的钓鱼网站检测算法 [J ] . 网络与信息安全学报 , 2024 , 10 ( 4 ): 123 - 131 .

Zhang S R , Yan Z W , Dong K J , et al . Phishing detection algorithm based on attention and feature fusion [J ] . Chinese Journal of Network and Information Security , 2024 , 10 ( 4 ): 123 - 131 .

Bahaghighat M , Ghasemi M , Ozen F . A high-accuracy phishing website detection method based on machine learning [J ] . Journal of Information Security and Applications , 2023 , 77 : 103553 .

Tan C C L , Chiew K L , Yong K S C , et al . Hybrid phishing detection using joint visual and textual identity [J ] . Expert Systems with Applications , 2023 , 220 : 119723 .

Johnson O V , Chen X Y , Khaw K W , et al . Ps-CALR: periodic-shift cosine annealing learning rate for deep neural networks [J ] . IEEE Access , 2023 , 11 : 139171 - 139186 .

Lin T Y , Goyal P , Girshick R , et al . Focal loss for dense object detection [C ] // Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV) . Piscataway : IEEE Press , 2017 : 2999 - 3007 .

Liu Q , Liu H T , Zhao Y , et al . Dual-branch network with dual-sampling modulated dice loss for hard exudate segmentation in color fundus images [J ] . IEEE Journal of Biomedical and Health Informatics , 2022 , 26 ( 3 ): 1091 - 1102 .

Adebowale M A , Lwin K T , Hossain M A . Intelligent phishing detection scheme using deep learning algorithms [J ] . Journal of Enterprise Information Management , 2023 , 36 ( 3 ): 747 - 766 .

Aljofey A , Jiang Q S , Qu Q , et al . An effective phishing detection model based on character level convolutional neural network from URL [J ] . Electronics , 2020 , 9 ( 9 ): 1514 .

Zhu Y M , Wang M , Yin X F , et al . Deep learning in diverse intelligent sensor based systems [J ] . Sensors , 2023 , 23 ( 1 ): 62 .

Benavides-Astudillo E , Fuertes W , Sanchez-Gordon S , et al . Comparative study of deep learning algorithms in the detection of phishing attacks based on HTML and text obtained from web pages [C ] // Applied Technologies . Cham : Springer Nature Switzerland , 2023 : Proceedings of the 4th International Conference on 386 - 398 .

Ponni P , Prabha D . Randomized active learning to identify phishing URL [M ] // Advanced Communication and Intelligent Systems . Cham : Springer Nature Switzerland , 2023 : 533 - 539 .

Mijwil M M , Salem I. E , Ismaeel M M . The significance of machine learning and deep learning techniques in cybersecurity: a comprehensive review [J ] . Iraqi Journal for Computer Science and Mathematics , 2023 : 87 - 101 .

Ghosh M , Srivastava S , Upadhyaya A , et al . CATALOG: exploiting joint temporal dependencies for enhanced phishing detection on ethereum [C ] // Proceedings of the ACM on Web Conference 2025 . New York : ACM , 2025 : 969 - 977 .

Lin Y , Liu R F , Divakaran D M , et al . Phishpedia: a hybrid deep learning based approach to visually identify phishing webpages [C ] // Proceedings of the 30th USENIX Security Symposium . Berkeley : USENIX Association , 2021 : 3793 - 3810 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Anomaly detection in DNS: an empirical study

FL-IoV: comprehensive intrusion detection method for internet of vehicles based on federated learning

Multi-task adversarial attribution method based on hierarchical structure

Research progress of color image steganalysis

Deep learning-based method for mobile social networks with strong sparsity for link prediction

Related Author

HOU Ronghao

HE Ming

KONG Kaichuan

CHEN Yong

ZUO Peng

JIANG Yuchang

JIANG Libing

HUANG Kai

Related Institution

China Internet Network Information Center

School of Cyber Science and Engineering, Jinan University

Beijing Unionpay Card Technology Co., Ltd., Shenzhen Branch

School of Computer Science, Nanjing University of Posts and Telecommunications

School of Automation, Nanjing University of Science and Technology

⁰