Most network embedding (NE) methods are either based on the proximity for community-guided tasks or on the structural similarity for role-oriented tasks. While being prevalent and effective, there still exists some potential issues that need further attention: 1) community and role are always regarded as orthogonal problems. They have rarely been combined to model the latent structures within the complex network. However, the generation of the network is usually jointly driven by these two mechanisms; and 2) few works study the interaction between roles or communities, which leads to the generation process of the network cannot being effectively modeled. To solve these problems, we propose a unified network embedding framework via mutual fusion of community and role (UMFCR). We combine the Gaussian mixture model (GMM) with a variational graph auto-encoder to generate node embeddings and discover the membership distribution of each node. An elaborate fusion pattern is then designed to produce the generation process for each link from the perspective of both community and role. The promising experimental results on real-world data demonstrate the necessity of fusing these two mechanisms and the superior performance of the model on different network tasks.

Biometrics is one of the most prominent sources of knowledge for human identification in the rapidly advancing field of cybersecurity. Among various types of biometric information, electrocardiogram (ECG) and photoplethysmography (PPG) signals gained considerable attention due to its inherent resistance to spoofing. However, existing researches are predominantly conducted in controlled environments, where ECG or PPG signals are susceptible to interference from motion and noise in real-world scenarios, resulting in suboptimal model robustness. Additionally, the few multimodal ECG identification studies mainly based on different conversion methods of ECG signal, resulting in information redundancy between modalities. To address these, we propose a lightweight identity recognition model based on cross-domain multimodal feature fusion of ECG and PPG signals. It incorporates an enhanced MultiRes block that forwards upsampled data for multi-resolution analysis of signal features, and employs spatial pyramid pooling for multi-scale feature extraction. Using this network structure, we separately extract and fuse features from ECG and PPG signals, then perform identity recognition based on the fused features. Extensive experiments were conducted to evaluate the proposed algorithm. The proposed fusion model reaches a high accuracy of 97.26%, improving by 17.06% and 40.48% compared to single-modal ECG and PPG identification models, respectively. By visualizing the fused features alongside single-modal ECG and PPG features, we intuitively explain the effectiveness of multimodal feature fusion for feature extraction, offering a new approach for multimodal biometric recognition.

To combat the growing IoT malware threat, many studies propose ML-based classification solutions, but the lack of comprehensive evaluations limits insights for developing new solutions and selecting models. Given this necessity, this work evaluates IoT malware classification models under three key challenges: temporal, architectural, and spatial inconsistencies between development and deployment datasets, which can be regarded as variables characterizing the dataset, and the challenges arise from the variable values inconsistency between the two stages. To improve the conclusions’ comprehensiveness, effectiveness, and generalizability, the evaluation is organized hierarchically across three levels based on model generation, sample variation, and inconsistency assumptions. Given the complexity of the model development pipeline, our evaluation treats each model individually and aims to conclude impacts across all models. The analysis reveals that temporal and architectural inconsistencies significantly degrade model performance, with architectural inconsistency having a greater impact, despite cross-architecture designs. Temporal inconsistency effects vary with temporal value differences, while spatial inconsistency has minimal impact, even with substantial spatial variation. Furthermore, we use one-way ANOVA to identify features contributing to family distinguishability, temporal stability, and architectural generalizability that benefit future solution design. Meanwhile, we studied a specific example to study the model performance degradation under architectural inconsistency. Finally, we summarize the lessons learned and outline potential research directions to address these challenges.

Recently, electrocardiogram (ECG) signals have garnered significant attention in the field of identity authentication due to its biological uniqueness. For identity authentication, the ECG signals that collected by wearing smart devices need to be determined whether the signal belongs to an enrolled one. Constrained by the computational efficiency of smart devices in practical scenarios, it is essential to reduce the complexity of the method to lower the computational load. To maintain the accuracy of identity authentication, most research efforts rely on both R-wave extraction and segmentation for subsequent authentication. Moreover, many methods constantly require the model to be retrained during the user enrollment stage, leading to performance degradation and waste of training resources. Hence, we propose a simple yet effective ECG identity authentication method that applies blind segmentation and is free from retraining, which greatly simplifies the authentication process. To mitigate the Equal Error Rate (EER) during the verification phase, a combination of AAM-softmax and triplet losses is employed, along with the incorporation of the hard negative mining within batch samples. Extensive experiments demonstrate that our method outperforms competitors by a large margin, e.g., achieving 0.40% EER on the large-scale Autonomic dataset.

The evolution of IoT malware has ignited interest in the creation of malware family classification models. Nonetheless, these models encounter security concerns stemming from issues related to their interpretability and vulnerabilities exposed within the training pipeline. Recent research highlighted the limitations of learning-based malware classifiers, which are susceptible to backdoor attacks due to relying on human-engineered features to simplify the mapping from features to binary perturbations. In contrast, our study aligns with the current trajectory of the malware classification field, where we emphasize the detection of backdoor attacks targeted at models employing features extracted from within the model itself. To thoroughly assess model vulner-abilities, we have devised a dynamic trigger generation method based on sample features, which we refer to as “BENIGN”. This approach is used to contaminate and launch attacks on the model while also implementing a tailored training process to achieve specific attack objectives. Through experiments, we analyze the impact of variables involved in its training procedures on the attack stability and success rates. Last, we evaluate mitigation methods and emphasize the challenges and adaptability needed to defend against these attack strategies.

Cardiotocography (CTG) is an important medical diagnostic tool when it comes to monitoring fetal wellbeing. It records Fetal Heart Rate (FHR) and uterine contraction activity, and can be used to detect whether the fetus is receiving oxygen adequately or in distress. Unfortunately, the interpretation of CTG recordings is highly subjective which can lead to unnecessary medical intervention that represents a risk for both the mother and the fetus. In this regard, intelligent CTG (ICTG) classification is a challenging research that can assist obstetricians in making clinical decisions, thereby improving the efficiency and accuracy of pregnancy management. But, many of these models focus on one specific modality that lack generalization to unseen or test data samples. In this study, a multi-modal fusion learning approach is proposed for pathological diagnosis of fetal distress. It combines signal and image modalities for multi-modal inputs and develops a Multi-modal Encoder Network (MENet) model based on DNN for capturing the underlying distribution of multi-modal data samples. Experimental results demonstrate that under the constraints of same classifier structure, MENet performs well in terms of classification accuracy and stability, far superior to several existing ICTG algorithms.

Fetal Heart Rate (FHR), an important recording in Cardiotocography (CTG)-based fetal health status monitoring, is the only information that clinical obstetricians can directly obtain and use. A challenge, however, is that missing samples are very common in FHR due to various causes such as fetal movements and sensor malfunctions. The aim is the development of an inpainting tool which is suitable for different missing lengths$q$and various total missing percentages$Q$, as well as for use in online mode. This study focused on two major impediments to existing inpainting methods: the longer the missing length, the more difficult it is to recover with mathematical methods; the reliance on tens of thousands of training samples, and the computational burden caused by full batch-based dictionary learning algorithms. We present a regularized minimization approach to signal recovery, which combines a${{\rm{L}}_{{0}{\rm{.6}}}}{\rm{ - norm}}$minimized sparse dictionary learning algorithm (MSDL) and a model optimization strategy for using a mini-batch version for signal recovery. Using 100 FHR recordings with 2 protocols designed to simulate missing clinical data scenarios, the combined method performed favorably in terms of 5 data analysis metrics and 3 clinical indicators. Comparing 4 inpainting methods, we were able to prove the superiority of the proposed algorithm for both large$q$and large$Q$. The experimental results showed the lowest values (2.64 (MAE), 4.68 (RMSE)) when${\rm{Q}} = {\rm{5\% }}$with short interval lengths. The developed architecture provides a reference value for the practical application of recovering missing samples online.

Physiological signal based biometric analysis has recently attracted attention as a means of meeting increasing privacy and security requirements. The real-time nature of an electrocardiogram (ECG) and the hidden nature of the information make it highly resistant to attacks. This paper focuses on three major bottlenecks of existing deep learning driven approaches: the lengthy time requirements for optimizing the hyperparameters, the slow and computationally intense identification process, and the unstable and complicated nature of ECG acquisition. We present a novel deep neural network framework for learning human identification feature representations directly from ECG time series. The proposed framework integrates deep bidirectional long short-term memory (BLSTM) and adaptive particle swarm optimization (APSO). The overall approach not only avoids the inefficient and experience-dependent search for hyperparameters, but also fully exploits the spatial information of ordinal local features and the memory characteristics of a recognition algorithm. The effectiveness of the proposed approach is thoroughly evaluated in two ECG datasets, using two protocols, simulating the influence of electrode placement and acquisition sessions in identification. Comparing four recurrent neural network structures and four classical machine learning and deep learning algorithms, we prove the superiority of the proposed algorithm in minimizing overfitting and self-learning of time series. The experimental results demonstrated an average identification rate of 97.71%, 99.41%, and 98.89% in training, validation, and test sets, respectively. Thus, this study proves that the application of APSO and LSTM techniques to biometric human identification can achieve a lower algorithm engineering effort and higher capacity for generalization.

Mobile botnet has recently evolved due to the rapid growth of smartphone technologies. Unlike legacy botnets, mobile devices are characterized by limited power capacity, calculation capabilities, and wide communication methods. As such, the logical topology structure and communication mode have to be redesigned for mobile botnets to narrow energy gap and lower the reduction speed of nodes. In this paper, we try to design a novel Gini‐based energy balancing scheme (Gleer) for the atomic network, which is a fundamental component of the heterogeneous multilayer mobile botnet. Firstly, for each operation cycle, we utilize the dynamic energy threshold to categorize atomic network into two groups. Then, the Gini coefficient is introduced to estimate botnet energy gap and to regulate the probability for each node to be picked as a region C&C server. Experimental results indicate that our proposed method can effectively prolong the botnet lifetime and prevent the reduction of network size. Meanwhile, the stealthiness of botnet with Gleer scheme is analyzed from users’ perspective, and results show that the proposed scheme works well in the reduction of user’ detection awareness.