Abstract
Convolutional neural network based on attention mechanism and a bidirectional independent recurrent neural network tandem joint algorithm (CATIR) are proposed. In natural language processing related technologies, word vector features are extracted based on URLs, and the extracted URL information features and host information features are merged. The proposed CATIR algorithm uses CNN (Convolutional Neural Network) to obtain the deep local features in the data, uses the Attention mechanism to adjust the weights, and uses IndRNN (Independent Recurrent Neural Network) to obtain the global features in the data. The experimental results shows that the CATIR algorithm has significantly improved the accuracy of malicious URL detection based on traditional algorithms to 96.9%.
Highlights
With the development of the Internet, the Internet is becoming more and more closely connected with users' lives, and more and more personal information of users exists on the Internet, so the security of information on the Internet is especially important
(2) We use CNN to get deep local features in the data and use Attention mechanism to adjust the weights, and concatenate CNN and Attention mechanism to get more effective data features; we use IndRNN to get global features in the data and it is processed in series with CNN and attention mechanism to get more comprehensive data information, which is used for the analysis and detection of network malicious URL
In order to get the hidden features of all kinds of features more fully to improve the detection effect of malicious URL detection, we need to learn the information features of host and URL to get block features with the help of algorithm, secondly, face features are obtained by learning word vector features and block features with the help of algorithm[22][23][24], and face features are input to the CATIR tandem joint algorithm in this paper for training and used for malicious URL analysis and detection, and the overall schematic diagram is as follows Figure 1
Summary
With the development of the Internet, the Internet is becoming more and more closely connected with users' lives, and more and more personal information of users exists on the Internet, so the security of information on the Internet is especially important. Jianguo Jiang et al.[5] proposed a character-level deep neural network-based online detection scheme that uses a natural language processing approach to map URL and DNS strings into vector form, and a convolutional neural network framework designed to automatically extract malicious features and train a classification model.Selva Ganapathy et al.[6] used a stacked constrained Boltzmann machines for binary classification by feature selection in deep neural networks, using IBK-kNN, binary correlation and Powerset with SVM labels for multi-category classification, tested on a sample of 27,700 URLs with good results. (2) We use CNN to get deep local features in the data and use Attention mechanism to adjust the weights, and concatenate CNN and Attention mechanism to get more effective data features; we use IndRNN to get global features in the data and it is processed in series with CNN and attention mechanism to get more comprehensive data information, which is used for the analysis and detection of network malicious URL
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
