In recent years, low recall rates and high dependencies on data labelling have become the biggest obstacle to developing deep anomaly detection (DAD) techniques. Inspired by the success of generative adversarial networks (GANs) in detecting anomalies in computer vision and imaging, we propose an anomaly detection model called FlowGANAnomaly for detecting anomalous traffic in network intrusion detection systems (NIDS). Unlike traditional GAN-based approaches, which are composed of a flow encoder, a convolutional encoder-decoder-encoder, a flow decoder and a convolutional encoder, the architecture of this model consists of a generator (G) and a discriminator (D). FlowGANAnomaly maps the different types of traffic feature data from separate datasets to a uniform feature space, thus can capture the normality of network traffic data more accurately in an adversarial manner to mitigate the problem of the high dependence on data labeling. Moreover, instead of simply detecting the anomalies by the output of D, we proposed a new anomaly scoring method that integrates the deviation between the output of two Gs' convolutional encoders with the output of D as weighted scores to improve the low recall rate of anomaly detection. We conducted several experiments comparing existing machine learning algorithms and existing deep learning methods (AutoEncoder and VAE) on four public datasets (NSL-KDD, CIC-IDS2017, CIC-DDoS2019, and UNSW-NB15). The evaluation results show that FlowGANAnomaly can significantly improve the performance of anomaly-based NIDS.

With the rapid advancement of the Internet of Things, network security has garnered increasing attention from researchers. Applying deep learning (DL) has significantly enhanced the performance of Network Intrusion Detection Systems (NIDSs). However, due to its complexity and “black box” problem, deploying DL-based NIDS models in practical scenarios poses several challenges, including model interpretability and being lightweight. Feature selection (FS) in DL models plays a crucial role in minimizing model parameters and decreasing computational overheads while enhancing NIDS performance. Hence, selecting effective features remains a pivotal concern for NIDSs. In light of this, this paper proposes an interpretable feature selection method for encrypted traffic intrusion detection based on SHAP and causality principles. This approach utilizes the results of model interpretation for feature selection to reduce feature count while ensuring model reliability. We evaluate and validate our proposed method on two public network traffic datasets, CICIDS2017 and NSL-KDD, employing both a CNN and a random forest (RF). Experimental results demonstrate superior performance achieved by our proposed method.

The proliferation of consumer electronics and 5G advancements have driven explosive growth in network traffic. Classifying traffic on mobile edge devices is critical for consumer electronics services. While deep learning-based Traffic Classification (TC) methods show promise, the dynamic and heterogeneous nature of edge networks causes network traffic Data Distribution Drift (DDD) and further Continual Learning (CL), degrading TC performance. Although various supervised drift detection methods have been explored, most inevitably highly depend on data labeling. Additionally, unsupervised methods incur high resource consumption and false positive rate. To address this issues, we propose an Unsupervised Real-Time Drift Detection and Adaptation Method for Continual Traffic Classification (UDDA-TC). Firstly, we propose a sliding window-based adaptive weighted Cumulative Sum (CUSUM) detector for stable drift monitoring and efficient DDD detection. Building on this, we introduce a Weighted Logits Variation (WLV)-based unsupervised real-time drift detection method enhances detection sensitivity without labels. Finally, to integrate CL into drift adaptation on edge devices, we propose a training set reconstruction strategy combines adaptive multi-window sampling with representative historical data retention to minimize labeling efforts, storage costs, and privacy risks for edge deployment. Experiments on public and private network traffic datasets showed that UDDA-TC outperforms other methods across metrics.

Social media platforms such as Weibo and TikTok have become more influential than traditional media. Sentiment in social media comments reflects users’ attitudes and impacts society, making sentiment analysis (SA) crucial. AI driven models, especially deep-learning models, have achieved excellent results in SA tasks. However, most existing models are not interpretable enough. First, deep learning models have numerous parameters, and their transparency is insufficient. People cannot easily understand how the models extract features from input data and make sentiment judgments. Second, most models lack intuitive explanations. They cannot clearly indicate which words or phrases are key for emotion prediction. Moreover, extracting sentiment factors from comments is challenging because a comment often contains multiple sentiment characteristics. To address these issues, we propose a dual-branch combination network (DCN) for SA of social media short comments, achieving both word-level and sentence-level interpretability. The network includes a key word feature extraction network (KWFEN) and a key word order feature extraction network (KWOFEN). KWFEN uses popular emotional words and SHAP for word-level interpretability. KWOFEN employs position embedding and an attention layer to visualize attention weights for sentence-level interpretability. We validated our method on the public dataset weibo2018 and TSATC. The results show that our method effectively extracts positive and negative sentiment factors, establishing a clear mapping between model inputs and outputs, demonstrating good interpretability performance.

With the development of Web 3.0 and Metaverse technologies, the ability of autonomous vehicles has been dramatically improved. These technologies have decentralized features that break the traditional data-sharing mode, grant users control over their data, and achieve benefits through data sharing, promoting the widespread circulation of data. To ensure data exchange security, flexibility, and reliability, this paper proposes FASDSA: A Flexible, Adaptive, and Secure Data Sharing Architecture for CAVs with Web 3.0 and Metaverse. This architecture has three advantages: First, it adopts a decentralized, federated learning and CAV role division method, which allows different computational power CAVs to participate in data sharing according to their roles, achieving flexible data privacy protection. Second, it has the ability of tampered model detection based on interpretable analysis, which can effectively ensure that the model is not tampered with. Third, it has a reward mechanism based on work contribution and trust assessment, which uses blockchain technology to ensure the continuous security operation of this architecture. To verify the performance of FASDSA, we used the UNSW-NB15 dataset to conduct three experiments. The experimental results indicate that compared to traditional methods, FASDSA possesses greater flexibility and security while maintaining similar or even superior model performance.

Multimodal emotion recognition (MER) has been widely adopted in affective computing and human–computer interaction. However, real-world multimodal streams frequently suffer from missing or incomplete modalities due to sensor failures, privacy constraints, and heterogeneous acquisition costs, which often causes severe performance degradation for models trained under complete inputs. To address this issue, we propose hierarchical VAE–diffusion for emotion reconstruction (HVDER), a latent-space generative framework for incomplete MER. HVDER first employs modality-specific variational autoencoders (VAEs) to project language, visual, and audio features into a unified low-dimensional latent space with KL-regularized structure. Then, a two-stage coarse-to-fine conditional diffusion module completes missing modality latents by recovering global emotion semantics followed by refining local discriminative details. Finally, the completed and observed modality representations are fused for emotion prediction, optimized end-to-end with a multiobjective loss integrating reconstruction, diffusion matching, cross-modal alignment, and classification supervision. Extensive experiments on CMU-MOSEI and CMU-MOSI validate the effectiveness and robustness of HVDER. Under fixed modality-missing settings, HVDER achieves average ACC2/F1 of 76.8/75.4 on CMU-MOSEI and 73.4/72.9 on CMU-MOSI, outperforming representative baselines across all modality combinations. Under random missing with a high missing rate of MR = 0.7, HVDER maintains ACC2/F1 of 75.4/72.5 on CMU-MOSEI and 68.0/67.2 on CMU-MOSI, indicating a stronger performance lower bound in severely incomplete scenarios. Ablation studies further confirm that latent-space modeling and the coarse-to-fine diffusion design jointly contribute to the main performance gains.