Explore the vast range of titles available.

Detection of anomalies in video surveillance plays a key role in ensuring the safety and security of public spaces. The number of surveillance cameras is growing, making it harder to monitor them manually. So, automated systems are needed. This change increases the demand for automated systems that detect abnormal events or anomalies, such as road accidents, fighting, snatching, car fires, and explosions in real-time. These systems improve detection accuracy, minimize human error, and make security operations more efficient. In this study, we proposed the Composite Recurrent Bi-Attention (CRBA) model for detecting anomalies in surveillance videos. The CRBA model combines DenseNet201 for robust spatial feature extraction with BiLSTM networks that capture temporal dependencies across video frames. A multi-attention mechanism was also incorporated to direct the model’s focus to critical spatiotemporal regions. This improves the system’s ability to distinguish between normal and abnormal behaviors. By integrating these methodologies, the CRBA model improves the detection and classification of anomalies in surveillance videos, effectively addressing both spatial and temporal challenges. Experimental assessments demonstrate that the CRBA model achieves high accuracy on both the University of Central Florida (UCF) and the newly developed Road Anomaly Dataset (RAD). This model enhances detection accuracy while also improving resource efficiency and minimizing response times in critical situations. These advantages make it an invaluable tool for public safety and security operations, where rapid and accurate responses are needed for maintaining safety.

With the development of artificial intelligence, more and more people hope that computers can understand human language through natural language technology, learn to think like human beings, and finally replace human beings to complete the highly difficult tasks with cognitive ability. As the key technology of natural language understanding, sentence representation reasoning technology mainly focuses on the sentence representation method and the reasoning model. Although the performance has been improved, there are still some problems such as incomplete sentence semantic expression, lack of depth of reasoning model, and lack of interpretability of the reasoning process. In this paper, a multi-layer semantic representation network is designed for sentence representation. The multi-attention mechanism obtains the semantic information of different levels of a sentence. The word order information of the sentence is also integrated by adding the relative position mask between words to reduce the uncertainty caused by word order. Finally, the method is verified on the task of text implication recognition and emotion classification. The experimental results show that the multi-layer semantic representation network can promote sentence representation’s accuracy and comprehensiveness.

To tackle the challenges of weak sensing capacity for multi-scale objects, high missed detection rates for occluded targets, and difficulties for model deployment in detection tasks of intelligent roadside perception systems, the PDT-YOLO algorithm based on YOLOv7-tiny is proposed. Firstly, we introduce the intra-scale feature interaction module (AIFI) and reconstruct the feature pyramid structure to enhance the detection accuracy of multi-scale targets. Secondly, a lightweight convolution module (GSConv) is introduced to construct a multi-scale efficient layer aggregation network module (ETG), enhancing the network feature extraction ability while maintaining weight. Thirdly, multi-attention mechanisms are integrated to optimize the feature expression ability of occluded targets in complex scenarios, Finally, Wise-IoU with a dynamic non-monotonic focusing mechanism improves the accuracy and generalization ability of model sensing. Compared with YOLOv7-tiny, PDT-YOLO on the DAIR-V2X-C dataset improves mAP50 and mAP50:95 by 4.6% and 12.8%, with a parameter count of 6.1 million; on the IVODC dataset by 15.7% and 11.1%. We deployed the PDT-YOLO in an actual traffic environment based on a robot operating system (ROS), with a detection frame rate of 90 FPS, which can meet the needs of roadside object detection and edge deployment in complex traffic scenes.

Semantic segmentation of high-resolution remote sensing images is of great application value in fields like natural disaster monitoring. Current multimodal semantic segmentation methods have improved the model’s ability to recognize different ground objects and complex scenes by integrating multi-source remote sensing data. However, these methods still face challenges such as blurred boundary segmentation and insufficient perception of multi-scale ground objects when achieving high-precision classification. To address these issues, this paper proposes MMA-Net, a semantic segmentation network enhanced by two key modules: cross-layer multimodal fusion module and multi-scale multi-attention module. These modules effectively improve the model’s ability to capture detailed features and model multi-scale ground objects, thereby enhancing boundary segmentation accuracy, detail feature preservation, and consistency in multi-scale object segmentation. Specifically, the cross-layer multimodal fusion module adopts a staged fusion strategy to integrate detailed information and multimodal features, realizing detail preservation and modal synergy enhancement. The multi-scale multi-attention module combines cross-attention and self-attention to leverage long-range dependencies and inter-modal complementary relationships, strengthening the model’s feature representation for multi-scale ground objects. Experimental results show that MMA-Net outperforms state-of-the-art methods on the Potsdam and Vaihingen datasets. Its mIoU reaches 88.74% and 84.92% on the two datasets, respectively. Ablation experiments further verify that each proposed module contributes to the final performance.

The popularity and deepening of social networks have increased the risk of personal information leakage for users. To enhance the security of social networks, this study constructed an access control model based on the preferences of social network users. This model utilizes graph neural networks to generate access control strategies based on user preferences, and introduces a multi-layer attention mechanism to optimize the graph neural network. To better capture user preference information, the study sets the learning rate to 0.0001. The experimental results demonstrated that in the Twitter dataset, the accuracy of the proposed model reached 95.7% and the F1 score reached 96.2%, which were significantly higher than those of other models. These results indicated that the model could more accurately classify access control in social networks and reduce false positives. The area under the receiver operation characteristic curve of the proposed model was 0.982, which was higher than other models. The decision time was 13.77 seconds, significantly lower than other models. This indicated that the model could more effectively distinguish different types of user access requests and provide more reliable guarantees for secure access to social networks. The user's preferred social network access control model based on graph neural networks has superior performance, effectively ensuring the information security of social network users and laying the foundation for further development of access control technology

The growing incidence of cyberattacks necessitates a robust and effective Intrusion Detection Systems (IDS) for enhanced network security. While conventional IDSs can be unsuitable for detecting different and emerging attacks, there is a demand for better techniques to improve detection reliability. This study introduces a new method, the Deep Adaptive Multi-Layer Attention Network (DAMLAN), to boost the result of intrusion detection on network data. Due to its multi-scale attention mechanisms and graph features, DAMLAN aims to address both known and unknown intrusions. The real-world NSL-KDD dataset, a popular choice among IDS researchers, is used to assess the proposed model. There are 67,343 normal samples and 58,630 intrusion attacks in the training set, 12,833 normal samples, and 9711 intrusion attacks in the test set. Thus, the proposed DAMLAN method is more effective than the standard models due to the consideration of patterns by the attention layers. The experimental performance of the proposed model demonstrates that it achieves 99.26% training accuracy and 90.68% testing accuracy, with precision reaching 98.54% on the training set and 96.64% on the testing set. The recall and F1 scores again support the model with training set values of 99.90% and 99.21% and testing set values of 86.65% and 91.37%. These results provide a strong basis for the claims made regarding the model’s potential to identify intrusion attacks and affirm its relatively strong overall performance, irrespective of type. Future work would employ more attempts to extend the scalability and applicability of DAMLAN for real-time use in intrusion detection systems.

Due to the different modalities of data such as images and text, the difficulty of sentiment analysis increases. To achieve cross-modal sentiment analysis, the study firstly designs a cross-modal sentiment analysis method based on bi-directional long and short-term memory networks and bi-linear convolutional neural networks. At the same time, concepts such as image attributes are introduced in the experiment to detect irony in graphic and textual data. Finally, a hybrid strategy cross-modal sentiment analysis method is established in the experiment. After comparison, the proposed method has the highest subject working characteristic curve and PR, which are 5% and 3% higher than the comparative methods, respectively. The model has the lowest error take, with a minimum value of only 0.71%. The average F1 value and average accuracy reached 92.61% and 88.97%, respectively. When the validation set size is 400, the recognition time of the proposed method is 2.1 seconds. When iterating 50, the recognition time of this method is 0.9 seconds. In practical applications, the proposed method has accurately analyzed six types of graphic and textual content with different emotional tendencies. This method has the best detection results for both single graphic and cross-modal modes.

The knowledge tracing method based on deep learning is used to assess learners’ cognitive states, laying the foundation for personalized education. However, deep learning methods are inefficient when processing long-term series data and are prone to overfitting. To improve the accuracy of cognitive state prediction, we design a Multi-layer Attention Self-supervised Knowledge Tracing Method (MASKT) using self-supervised learning and the Transformer method. In the pre-training stage, MASKT uses a random forest method to filter out positive and negative correlation feature embeddings; then, it reuses noise-processed restoration tasks to extract more learnable features and enhance the learning ability of the model. The Transformer in MASKT not only solves the problem of long-term dependencies between input and output using an attention mechanism, but also has parallel computing capabilities that can effectively improve the learning efficiency of the prediction model. Finally, a multidimensional attention mechanism is integrated into cross-attention to further optimize prediction performance. The experimental results show that, compared with various knowledge tracing models on multiple datasets, MASKT’s prediction performance remains 2 percentage points higher. Compared with the multidimensional attention mechanism of graph neural networks, MASKT’s time efficiency is shortened by nearly 30%. Due to the improvement in prediction accuracy and performance, this method has broad application prospects in the field of cognitive diagnosis in intelligent education.

Relation extraction aims to identify semantic relations between entities in text. In recent years, this task has been extended to the joint extraction of entities and relations, which requires the simultaneous identification of entities and their relations from sentences. However, existing methods, limited by the existing tagging scheme, fail to identify more complex entities, which in turn limits the performance of the joint extraction task. This article presents a joint extraction model for entities and relations called MLRA-LSTM-CRF that uses multi-label tagging and relational alignment to transform this task into a multi-label tag recognition problem. The proposed model first tags the entities and their relations according to the multi-label tagging scheme and then uses a joint entity and relation extraction module with a multi-layer attention mechanism to extract the triplets in the sentence. Finally, the relational alignment module is used to align the predicted relation classification results. Experimental results on the New York Times and Wiki-KBP datasets indicate that MLRA-LSTM-CRF is significantly better than that of several state-of-the-art models and baseline.

Target images are complex and diverse under the influence of scale, occlusion and appearance factors in real scenes, and they affect the performance of target detection algorithms. They also make the existing target detection algorithms suffer from the following problems. On the one hand, the neurons in the target detection algorithm architecture cannot learn the complex interaction and semantic features inside the target image. On the other hand, the feature expression of different target images is insufficient and the channel reduction leads to the loss of position information and other problems. herefore, a multi-layer attention mechanism of considering both node and semantic level attention in the model architecture is proposed. In this method, the fusion of neighbors and semantic information is weighted, and node representations is learned under a hierarchical aggregation manner.Just because of this, it can improve the effectiveness and interpretability of the model, and solve the problem of complex interaction and rich semantic feature acquisition within images. Furthermore, we propose an adaptive feature fusion network which can adaptively filter the useless information of other layers and retain the feature information that is beneficial to target recognition. A feature enhancement module, is introduced to enhance the identifiability of the top-level target features of the feature network, and which can alleviate the problem of loss of target position. Finally, the extensive tests using PASCAL VOC and MSCOCO datasets, the experimental result shows that our method not only has the best recognition performance, but also has better stability and robustness.