Phishing is a kind of online attack that attempts to defraud sensitive information of network users. Current phishing webpage detection methods mainly use manual feature collection, and there are problems that feature extraction is complicated and the possible correlation between features cannot be avoided. To solve the problems, a new phishing webpage detection model is proposed, among which the main components are automatic learning representations from multi-aspects features through representation learning and extracting features by hybrid deep learning network. Firstly, the model treats URL, HTML page content, and DOM (Document Object Model) structure of webpages as character sequences respectively, and uses representation learning technology to automatically learn the representation of the webpages; then, sends multiple representations to a hybrid deep learning network composed of a convolutional neural network and a bidirectional long and short-term memory network through different channels to extract local and global features, and use the attention mechanism to strengthen the influence of important features; finally, the output of multiple channels is fused to realize classification prediction. Through four sets of experiments to verify the detection effect of the model, the results show that the overall classification effect of the model is better than the existing classic phishing webpage detection methods, the accuracy reaches 99.05%, and the false positive rate is only 0.25%. It is proved that the strategies of extracting webpage features from all aspects through representation learning and hybrid deep learning network can effectively improve the detection effect of phishing webpages.

Video anomaly detection is a critical component of intelligent video surveillance systems, extensively deployed and researched in industry and academia. However, existing methods have a strong generalization ability for predicting anomaly samples. They cannot utilize high-level semantic and temporal contextual information in videos, resulting in unstable prediction performance. To alleviate this issue, we propose an encoder–decoder model named SMAMS, based on spatiotemporal masked autoencoder and memory modules. First, we represent and mask some of the video events using spatiotemporal cubes. Then, the unmasked patches are inputted into the spatiotemporal masked autoencoder to extract high-level semantic and spatiotemporal features of the video events. Next, we add multiple memory modules to store unmasked video patches of different feature layers. Finally, skip connections are introduced to compensate for crucial information loss caused by the memory modules. Experimental results show that the proposed method outperforms state-of-the-art methods, achieving AUC scores of 99.9%, 94.8%, and 78.9% on the UCSD Ped2, CUHK Avenue, and Shanghai Tech datasets.

At present, the existing abnormal event detection models based on deep learning mainly focus on data represented by a vectorial form, which pay little attention to the impact of the internal structure characteristics of feature vector. In addition, a single classifier is difficult to ensure the accuracy of classification. In order to address the above issues, we propose an abnormal event detection hybrid modulation method via feature expectation subgraph calibrating classification in video surveillance scenes in this paper. Our main contribution is to calibrate the classification of a single classifier by constructing feature expectation subgraphs. First, we employ convolutional neural network and long short-term memory models to extract the spatiotemporal features of video frame, and then construct the feature expectation subgraph for each key frame of every video, which could be used to capture the internal sequential and topological relational characteristics of structured feature vector. Second, we project expectation subgraphs on the sparse vector to combine with a support vector classifier to calibrate the results of a linear support vector classifier. Finally, the experiments on a common dataset named UCSDped1 and a coal mining video dataset in comparison with some existing works demonstrate that the performance of the proposed method is better than several the state-of-the-art approaches.

Social robot accounts controlled by artificial intelligence or humans are active in social networks, bringing negative impacts to network security and social life. Existing social robot detection methods based on graph neural networks suffer from the problem of many social network nodes and complex relationships, which makes it difficult to accurately describe the difference between the topological relations of nodes, resulting in low detection accuracy of social robots. This paper proposes a social robot detection method with the use of an improved neural network. First, social relationship subgraphs are constructed by leveraging the user’s social network to disentangle intricate social relationships effectively. Then, a linear modulated graph attention residual network model is devised to extract the node and network topology features of the social relation subgraph, thereby generating comprehensive social relation subgraph features, and the feature-wise linear modulation module of the model can better learn the differences between the nodes. Next, user text content and behavioral gene sequences are extracted to construct social behavioral features combined with the social relationship subgraph features. Finally, social robots can be more accurately identified by combining user behavioral and relationship features. By carrying out experimental studies based on the publicly available datasets TwiBot-20 and Cresci-15, the suggested method’s detection accuracies can achieve 86.73% and 97.86%, respectively. Compared with the existing mainstream approaches, the accuracy of the proposed method is 2.2% and 1.35% higher on the two datasets. The results show that the method proposed in this paper can effectively detect social robots and maintain a healthy ecological environment of social networks.