Figure - available from: Security and Communication Networks
This content is subject to copyright. Terms and conditions apply.
Source publication
The traditional malicious uniform resource locator (URL) detection method excessively relies on the matching rules formulated by the network security personnel, which is hard to fully express the text information of the URL. Thus, an improved multilayer recurrent convolutional neural network model based on the YOLO algorithm is proposed to detect m...
Similar publications
Monitoring and predicting the security state of network environment is the key to solve network security. On the basis of studying the principle of network security situation awareness, a network security situation prediction model based on artificial immunity and Kalman algorithm is designed according to the artificial immunity evaluation model. F...
Citations
... The YOLO network has been widely utilized across various domains, encompassing tasks such as detecting malicious URLs [1], identifying small ships in optical images [2], analyzing smoking behavior in images [3], and recognizing vehicle targets [4]. These applications underscore the versatility and effectiveness of the YOLO algorithm in object detection tasks. ...
Object detection is a pivotal aspect of computer vision, essential for diverse recognition tasks. This study centers on exploring deep learning methodologies for object detection, specifically targeting the identification of URLs in images captured by mobile phones. We conduct a comparative analysis of three models from the YOLO family – YOLOv3, YOLOv4, and YOLOv5 – recognized for their efficacy in object detection. Our research addresses the unique challenge of detecting URLs in images, particularly considering the limited availability of URL-labeled dataset. Through rigorous experimentation and evaluation, we demonstrate the generalization capabilities of YOLOv3, YOLOv4, and YOLOv5, as measured by average precision scores. Furthermore, we highlight the resilience of the YOLOv4 model against various image-related challenges. Our findings contribute significantly to the advancement of computer vision, specifically in the domain of object detection for real-world applications. By evaluating the performance of cutting-edge deep learning models, we provide valuable insights into their effectiveness for URL detection, thereby enriching our understanding of their practical utility. This research serves as a foundation for future investigations aimed at leveraging deep learning techniques to enhance object detection accuracy across diverse contexts.
... (1) It is not completely truncated at negative values, thus ensuring better information inflow. (2) Its positive value is unbounded, with gradient tending to 1 in the left and right limits, thus avoiding gradient saturation [25]. (3) The gradient descent of Mish is superior, ensuring the smoothing of each point to the maximum extent possible [26]. ...
Recognizing cashmere and wool fibers has been a challenging problem in the textile industry due to their similar morphological structure, chemical composition, and physicochemical properties. Traditional manual methods for identifying these fibers are inefficient, labor-intensive, and inaccurate. To address these issues, we present a novel method for recognizing cashmere and wool fibers using an improved version of ShuffleNetV2 and Transfer Learning, which we implement as a new cashmere and wool classification network (CWCNet).The approach leverages depthwise separable dilated convolution to extract more feature information for fiber classification. We also introduce a new activation function that enhances the nonlinear representation of the model and allows it to more fully extract negative feature information. Experimental results demonstrate that CWCNet achieves an accuracy rate of up to 98.438% on our self-built dataset, which is a 2.084% improvement over the original ShuffleNetV2 model. Furthermore, our proposed method outperforms classical models such as EfficientNetB0, MobileNetV2, Wide-ResNet50, and ShuffleNetV2 in terms of recognition accuracy while remaining lightweight.The method is capable of extracting more information on fiber characteristics and has the potential to replace manual labor as technological advancements continue to be made. This will greatly benefit engineering applications in the textile industry by providing more efficient and accurate fiber classification.
... Chen et al. processed a dataset consisting of 200,000 URLs, including 100,000 normal URLs and 100,000 Malicious URLs, using the YOLO algorithm on an artificial neural network [4]. The features extracted were used to evaluate malicious URLs by a bidirectional LSTM recurrent neural network algorithm, and it was claimed that a success rate of 90% was achieved [5]. ...
Today, the increased use of the internet has become important in our lives and new communication technologies, social networks, e-commerce, online banking, and among other applications have a significant impact on the promotion and growth of business. In our study, we aimed to work with a large dataset and to achieve the best results in detecting malicious URL addresses using an artificial intelligence model. A 7-layer RNN model was used in the study, and two similar national and international datasets were combined and merged to create a big new dataset consisting of 579,112 URL addresses. Then, this new data set is divided into training and test sets. first data set was trained at the model and then the second data set was processed test. When this data set was processed in our model, we achieved a success rate of over 91%. This rate is a very good result of detecting malicious url addresses. Your contribution with this work is significant in developing more effective methods for detecting harmful sites as internet usage increases, parallel use of artificial intelligence models makes detection of such sites more effective and potentially assist users in protecting from various types of cyber-attacks is targeted.
... Its advantage is that it is fast, easy, and has a meager false-positive (FP) rate. However, the main problem with this method is that it has a high false-negative (FN) rate and fails to detect newly generated URLs [1], [5], [6]. Nevertheless, it has been widely utilized in several major browsers, including Mozilla Firefox, Safari, and Chrome, among others, due to its simplicity and efficiency [5]. ...
Network information security risks multiply and become more dangerous. Hackers today generally target end-to-end technology and take advantage of human weaknesses. Furthermore, hackers take advantage of technology weaknesses by applying various methods to attack. Nowadays, one of the greatest dangers to the modern digital world is malicious URLs, and stopping them is one of the biggest challenges in the field of cyber security. Detecting harmful URLs using machine learning and deep learning algorithms have been the subject of various academic papers. However, time and accuracy are the two biggest challenges of these tools. This paper proposes a multilayer perceptron (MLP) model that utilizes two significant aspects to make it more practical, lightweight, and fast: Using only lexical features and a decision tree (DT) algorithm to select the best relevant subset of features. The effectiveness of the experimental outcomes is evaluated in terms of time, accuracy, and error reduction. The results show that a MLP model using 35 features could achieve an accuracy of 94.51% utilizing only URL lexical features. Furthermore, the model is improved in time after applying the DT as feature selection with a slight improvement in accuracy and loss.
One of the challenges the world is facing is cyber-attacks and threats such as data breaches, data leaks, etc., in which sensitive, confidential or protected data has been compromised. According to the information accessed by CERT (Indian Computer Emergency Response Team), in past five years, 47 incidents of data leak and 142 data breaches have occurred in India. Henceforth, cyber-security is emphasized more as economies of country, institutions, and individual lives are highly interconnected and growing more complex, protection of privacy and confidentiality has become crucial. Data analysis and the right implication of handling the data security issues are well addressed by ML algorithms. Many research indicators exhibited the comparison of various ML models to combat the data security issues. In this chapter, a novel approach of applying neural network for data security outperforms the data security issues. Self-Learning by the neural network models finds better pattern recognition for learning and the performance would be independent of any bias factors.
Attackers perform malicious activities by sending URLs to victims via e-mail, SMS, social network messages, and other means. Recently, intruders have been generating malicious URLs algorithmically. They also use shortening or obfuscation services to bypass firewalls and other security barriers. Some machine learning methods have been presented in order to identify malicious URLs from normal ones, all of which are subject to classification errors. On the other hand, it is impractical to have a complete and up-to-date blacklist due to large number of daily generated malicious URLs. Therefore, calculating the URLs security risk would be more helpful than URLs classification. In this way a user can correctly decide whether to use an unfamiliar URL if they know its associated security risk. In this study, the problem of URLs security risk computation is introduced and two effective novel criteria for this problem are proposed. Based on these criteria, a security risk score can be estimated for each incoming URL. In the first criterion, based on previous malicious and non-malicious URL instances, the extracted features of a URL are divided into two categories, those increase the risk and those reduce the security risk. In the second criterion, security risk score of an unknown URL is estimated based on its distances to nearest known malicious and also safe URLs. For both criterion, corresponding formulations and algorithms are also designed and are described. Extensive empirical evaluations on various real datasets show the effectiveness of the proposed criteria in terms of malicious URL detection rate. Moreover, our experiments show that the proposed metrics significantly outperforms previously proposed risk score criteria.
Object detection is a key technique in computer vision, as it is considered a necessary step in any recognition process. It is the procedure for determining the instance of the class to which the object belongs and estimating its location by displaying its bounding box. It was widely accepted that advances in object detection have generally gone through two periods: ”the traditional object detection period”, where detection was performed through classical machine learning techniques, and ”the deep learning based detection period”, where classical machine learning techniques have been completely replaced by methods based on deep neural networks. In this paper, we will focus on object detection based on deep learning. The main objective is to carry out a comparative study of three models of the YOLO family, already proven to be effective for object detection that are YOLOv3, YOLOv4, and YOLOv5 in the context of the detection of URLs in photos taken by a mobile phone. The experimental results, expressed in terms of average precision, showed the generalization ability of the three models, YOLOv3, YOLOv4, and YOLOv5. In addition, the stability of the YOLOV4 model against several difficulties added to the images.
Website link detection is an important means to ensure the security of the external chain. In the past, it was mainly realized through blacklisting and feature engineering-based machine learning, which has the problems of slow detection speed and weak model generalization ability. The development of neural networks has brought a new solution to the security detection of the external chain of the website. To address the performance bottleneck caused by the variable content length of web pages, this article introduces an innovative approach: a website external link security detection algorithm based on multi-modal fusion. It extracts text, dynamic script, and image features separately, and constructs a deep fusion model that combines these multi-modal features. Compared with the previous research results, the proposed method is superior to the traditional single-mode method, and can quickly and accurately identify malicious web pages. The accuracy and F1 value are improved by 2.7% and 0.026.
