In hard-label black-box adversarial attacks, where only the top-1 predicted label is accessible, the prohibitive query complexity poses a major obstacle to practical deployment. In this paper, we focus on optimizing a representative class of attacks that search for the optimal ray direction yielding the minimum ℓ₂-norm perturbation required to move a benign image into the adversarial region. Inspired by Nesterov's Accelerated Gradient (NAG), we propose a momentum-based algorithm, ARS-OPT, which proactively estimates the gradient with respect to a future ray direction inferred from accumulated momentum. We provide a theoretical analysis of its convergence behavior, showing that ARS-OPT enables more accurate directional updates and achieves faster, more stable optimization. To further accelerate convergence, we incorporate surrogate-model priors into ARS-OPT's gradient estimation, resulting in PARS-OPT with enhanced performance. The superiority of our approach is supported by theoretical guarantees under standard assumptions. Extensive experiments on ImageNet and CIFAR-10 demonstrate that our method surpasses 13 state-of-the-art approaches in query efficiency.

Traditional open set recognition (OSR) methods, due to their supervised learning nature, primarily focus on extracting features of known event classes, failing to effectively capture the distinguishing features of unknown event classes, which limits recognition performance. This paper proposes a self-supervised contrastive learning-based open-set recognition (SCOR) method for DAS perimeter security systems. This method leverages self-supervised contrastive learning to distinguish between known and unknown class samples, significantly improving the recognition rate for unknown classes by integrating a consistency training strategy and a dual-level threshold detection approach. Moreover, the incorporation of center loss enhances the feature clustering of known class samples, thereby boosting the classification accuracy. The algorithm’s hyperparameters are optimized efficiently using Bayesian optimization hyperband (BOHB). These designs yield a more compact and separable visual semantic space, allowing the detected unknown samples to be further refined into more accurate latent subclasses through clustering. Experimental results based on multiple DAS datasets show that the proposed method achieves an accuracy of 94.1-98.2 % for known events and 88.7%-92.3% for unknown events, with an average improvement of 2.7%-7.8% in recognition accuracy compared to the best-performing alternative methods. Furthermore, the algorithm successfully enables efficient online detection through a self-built DAS perimeter security system and upper-level software in a laboratory environment.

In recent years, various international security events have occurred frequently and interacted between real society and cyberspace. Traditional traffic monitoring mainly focuses on the local anomalous status of events due to a large amount of data. BGP-based event monitoring makes it possible to perform differential analysis of international events. For many existing traffic anomaly detection methods, we have observed that the window-based noise reduction strategy effectively improves the success rate of time series anomaly detection. Motivated by this, we propose an unsupervised anomaly detection model, MAD-MulW, which introduces a multi-window serial framework. The W-GAT module adaptively updates sample weights within the window to reduce noise, while the W-LAE module captures temporal trends through predictive reconstruction, enhancing inter-class separation. Our model has been experimentally validated on multiple BGP anomalous events with an average F1 score of over 90%, which demonstrates the significant improvement effect of the stage windows and adaptive strategy on the efficiency and stability of the timing model. The source code is available at://github.com/2024ChenYP/MAD-MulW.

Automatic modulation recognition is essential for large-scale wireless communications, but traditional methods often ignore unknown signals in open-set conditions, leading to their incorrect classification as known types and thereby compromising system reliability and communication security. To handle this challenge, a novel automatic modulation open set recognition (AMOSR) model based on deep metric learning with embedding enhancement is proposed in this article. First, for each known example, deep neural model is employed to separately extract the original in-phase and quadrature (IQ) signal and its instantaneous features, which are then fused to obtain embedding. Second, random erasing is employed to the original IQ signal and instantaneous features separately to obtain an augmented example, which has similar structure but different semantics with known example, and the embedding of this example is obtained by using first step. Then, the embedding space, in which tuplet loss and margin loss are combined with the embeddings of known and augmented examples, is trained to improve the overall performance of the model. Finally, after training, AMOSR is implemented using the class centers of known classes. Experiments on three automatic modulation datasets show that our model has better average performance than several mainstream methods in the field of computer vision.

It is known that deep neural networks may exhibit dangerous behaviors under various security threats (e.g., backdoor attacks, adversarial attacks and safety property violation) and there exists an ongoing arms race between attackers and defenders. In this work, we propose a complementary perspective to utilize recent progress on ''neural network repair'' to mitigate these security threats and repair various kinds of neural network defects (arising from different security threats) within a unified framework, offering a potential silver bullet solution to real-world scenarios. To substantially push the boundary of existing repair techniques (suffering from limitations such as lack of guarantees, limited scalability, considerable overhead, etc) in addressing more practical contexts, we propose ProRepair, a novel provable neural network repair framework driven by formal preimage synthesis and property refinement. The key intuitions are: (i) synthesizing a precise proxy box to characterize the feature space preimage, which can derive a bounded distance term sufficient to guide the subsequent repair step towards the correct outputs, and (ii) performing property refinement to enable surgical corrections and scale to more complex tasks. We evaluate ProRepair across four security threats repair tasks on six benchmarks and the results demonstrate it outperforms existing methods in effectiveness, efficiency and scalability. For point-wise repair, ProRepair corrects models while preserving performance and achieving significantly improved generalization, with a speed-up of 5× to 2000× over existing provable approaches. In region-wise repair, ProRepair successfully repairs all 36 safety property violation instances (compared to 8 by the best existing method), and can handle 18× higher dimensional spaces.

The impressive performance of Large Language Model (LLM) has prompted researchers to develop Multi-modal LLM (MLLM), which has shown great potential for various multi-modal tasks. However, current MLLM often struggles to effectively address fine-grained multi-modal challenges. We argue that this limitation is closely linked to the models’ visual grounding capabilities. The restricted spatial awareness and perceptual acuity of visual encoders frequently lead to interference from irrelevant background information in images, causing the models to overlook subtle but crucial details. As a result, achieving fine-grained regional visual comprehension becomes difficult. In this paper, we break down multi-modal understanding into two stages, from Coarse to Fine (CoF). In the first stage, we prompt the MLLM to locate the approximate area of the answer. In the second stage, we further enhance the model’s focus on relevant areas within the image through visual prompt engineering, adjusting attention weights of pertinent regions. This, in turn, improves both visual grounding and overall performance in downstream tasks. Our experiments show that this approach significantly boosts the performance of baseline models, demonstrating notable generalization and effectiveness. Our CoF approach is available online at https://github.com/Gavin001201/CoF.

3D human pose estimation (HPE) is crucial due to its extensive applications. While current 3D HPE methods focus on human skeleton topology for accuracy, they often overlook joint angle information, which is vital in 2D-to-3D pose lifting. This paper introduces the Joint Angle Network (JANE) model to leverage joint angle features from 2D poses. We propose an angle graph convolution module to extract joint angle features and a multi-stage parallel fusion module to integrate these features with input image data. Experiments on Human3.6M show that our method improves accuracy by 8.7% and 2.6% over non-spatiotemporal baselines using 2D ground truth and 2D pose detectors, respectively.

One of the most practical and challenging types of black-box adversarial attacks is the hard-label attack, where only the top-1 predicted label is available. One effective approach is to search for the optimal ray direction from the benign image that minimizes the $\ell_p$ norm distance to the adversarial region. The unique advantage of this approach is that it transforms the hard-label attack into a continuous optimization problem. The objective function value is the ray's radius, which can be obtained via binary search at a high query cost. Existing methods use a "sign trick" in gradient estimation to reduce the number of queries. In this paper, we theoretically analyze the quality of this gradient estimation and propose a novel prior-guided approach to improve ray search efficiency both theoretically and empirically. Specifically, we utilize the transfer-based priors from surrogate models, and our gradient estimators appropriately integrate them by approximating the projection of the true gradient onto the subspace spanned by these priors and random directions, in a query-efficient manner. We theoretically derive the expected cosine similarities between the obtained gradient estimators and the true gradient, and demonstrate the improvement achieved by incorporating priors. Extensive experiments on the ImageNet and CIFAR-10 datasets show that our approach significantly outperforms 11 state-of-the-art methods in terms of query efficiency.

Deep neural networks (DNNs) are suffering from ethical issues such as individual discrimination. In response, extensive NN repair techniques have been developed to adjust models and mitigate such undesired behaviors. However, existing fairness repair methods are typically data-centric, which often lack provable guarantees and generalization to unseen samples. To overcome these limitations, we propose PROF, a novel fairness repair framework with provable guarantees. The key intuition of PROF is to leverage interval bound propagation (a widely used NN verification technique) to soundly capture model outputs over the whole set ${\mathcal{S}}\left(x\right)$ around a biased sample x. The derived bounds are utilized to guide fairness repair which encourages the model to produce consistent outputs on ${\mathcal{S}}\left(x\right)$. Specifically, we integrate fairness constraints and model modifications into a unified constraint-solving formulation, which can be transformed to a Mixed-Integer Linear Programming (MILP) problem solvable by off-the-shelf solvers. The solution to the MILP problem effectively induces a repaired model with guaranteed fairness over the whole set ${\mathcal{S}}\left(x\right)$. We evaluate PROF on four widely used benchmark datasets and demonstrate that it achieves provable fairness repair, with generalization of up to 95.93% on full datasets and 93.16% on the entire input space. Notably, PROF can be easily configured to support multiple sensitive attributes and more practical fairness definitions, while providing provable repair guarantees and delivering around 90% fairness improvement. Our code is available in this $\color{red}{\text{repository}}$.

The rapid advancement of generative AI has raised concerns about the authenticity of digital images, as highly realistic fake images can now be generated at low cost, potentially increasing societal risks. In response, several datasets have been established to train detection models aimed at distinguishing AI-generated images from real ones. However, existing datasets suffer from limited generalization, low image quality, overly simple prompts, and insufficient image diversity. To address these limitations, we propose a high-quality, large-scale dataset comprising over 730,000 images across multiple categories, including both real and AI-generated images. The generated images are synthesized via state-of-the-art methods, including text-to-image generation (guided by over 10,000 carefully designed prompts), image inpainting, image refinement, and face swapping. Each generated image is annotated with its generation method and category. Inpainting images further include binary masks to indicate inpainted regions, providing rich metadata for analysis. Compared to existing datasets, detection models trained on our dataset demonstrate superior generalization capabilities. Our dataset not only serves as a strong benchmark for evaluating detection methods but also contributes to advancing the robustness of AI-generated image detection techniques. Building upon this, we propose a lightweight detection method based on image noise entropy, which transforms the original image into an entropy tensor of Non-Local Means (NLM) noise before classification. Extensive experiments demonstrate that models trained on our dataset achieve strong generalization, and our method delivers competitive performance, establishing a solid baseline for future research. The dataset and source code are publicly available at https://real-hd.github.io.

Privacy and security concerns are becoming increasingly critical for recommender systems, as model extraction attack provides an effective way to probe system robustness by replicating the model’s recommendation logic — potentially exposing sensitive user preferences and proprietary algorithmic knowledge. Despite the promising performance of existing model extraction methods, they still face two key challenges: unrealistic assumptions on the requirement of accessible member or surrogate data and generalization problem where surrogate model architecture constraints lead to overfitting on generated data. To tackle these challenges, in this paper, we first thoroughly analyze how the architecture of surrogate models influences extraction attack performance, highlighting the superior effectiveness of the graph convolution architecture. Based on this, we propose a novel Data-free Black-box Graph convolution-based Recommender Model Extraction method, dubbed DBGRME. Specifically, DBGRME contains: (1) an interaction generator to alleviate the need for member data requirements in a data-free scenario; and (2) a generalization-aware graph convolution-based surrogate model to capture diverse and complex recommender interaction patterns for mitigating the overfitting issue. Experimental results on various datasets and victim models demonstrate the superiority of our attack in data-free scenarios (e.g., surpassing PTQ data-require methods with 17.4% improvement on LightGCN). Code is available: \url{https://github.com/Vencent-Won/DBGRME.git}.

Understanding the enigmatic black-box representations within Deep Neural Networks (DNNs) is an essential problem in the community of deep learning . An initial step towards tackling this conundrum lies in quantifying the degree of similarity between these representations. Various approaches have been proposed in prior research, however, as the field of representation similarity continues to develop, existing metrics are not compatible with each other and struggling to meet the evolving demands. To address this, we propose a comprehensive similarity measurement framework inspired by the natural graph structure formed by samples and their corresponding features within the neural network . Our novel Graph-Based Similarity (GBS) framework gauges the similarity of DNN representations by constructing a weighted, undirected graph based on the output of hidden layers. In this graph, each node represents an input sample, and the edges are weighted in accordance with the similarity between pairs of nodes. Consequently, the measure of representational similarity can be derived through graph similarity metrics, such as layer similarity. We observe that input samples belonging to the same category exhibit dense interconnections within the deep layers of the DNN. To quantify this phenomenon, we employ a motif-based approach to gauge the extent of these interconnections. This serves as a metric to evaluate whether the representation derived from one model can be accurately classified by another. Experimental results show that GBS gets state-of-the-art performance in the sanity check. We also extensively evaluate GBS on downstream tasks to demonstrate its effectiveness, including measuring the transferability of pretrained models and model pruning.

Unmanned Aerial Vehicle (UAV) recognition using Deep Learning (DL) is critical for ensuring the safety of low-altitude airspace. However, the limited availability of labeled UAV signal data poses significant challenges to achieving high recognition accuracy and robustness. To address this, we propose a novel method, Self-Supervised learning with Self-Adaptive Pseudo-Labeling (SS-SAPL), designed to enhance UAV recognition performance. The method operates in two stages: a self-supervised pre-training stage and a semi-supervised fine-tuning stage. In the pre-training stage, contrastive learning with weak and strong data augmentations is employed to extract generic feature representations from all UAV signal samples. In the fine-tuning stage, Pseudo-Labeling (PL) is combined with a Self-Adaptive Threshold (SAT) and Self-Adaptive Fairness (SAF) mechanism to improve the accuracy of PSeudo-Labels (PSLs) and leverage both labeled and unlabeled data for refining feature representations. Simulation results demonstrate the effectiveness of our method. For UAV signals at 2.4 GHz with only 30 labeled samples, our approach achieves a recognition accuracy of 82.38%, outperforming state-of-the-art methods by at least 6.63%. In mixed-frequency scenarios (2.4 GHz and 5.8 GHz) with only 10 labeled samples, our method exceeds 92.13% accuracy, surpassing competitors by at least 4.63%. These results highlight the robustness and practical value of the proposed method in challenging environments.

Radio Frequency Fingerprinting Identification (RFFI) leverages signal processing to extract unique characteristics from wireless signals for device identification. In recent years, deep learning (DL) has significantly advanced signal identification, catalyzing progress in RFFI research. This paper proposes an enhanced RFFI method to manage variable-length signal inputs, typically problematic for neural networks such as convolutional neural networks (CNNs) and multilayer perceptrons (MLPs), by treating these signals as images to solve data formatting problems. The robust representation of the variable-length signal ultimately achieves over 90% accuracy, meeting the expected results. Furthermore, conventional DL-based RFFI methods require a comprehensive analysis of the entire RF signal, consuming significant computational resources and vulnerable to environmental variations. We address these issues by proposing an incremental learning (IL)-based RFFI method that allows dynamic model updates and improves recognition and generalization performance. Our method’s efficacy, tested on the power amplifiers (PA) dataset, enables real-time data stream processing.

Specific emitter identification (SEI) is a technique that identifies individual emitters based on the inherent characteristics reflected in the radio frequency signals due to the individual differences of the emitters. Deep learning (DL) has become the primary research method for identifying and authenticating wireless devices in SEI. However, in the real world, electromagnetic signals continuously change with the channel environment and time, causing models trained on datasets collected from known specific domains to exhibit significant performance degradation when applied to unknown channel environments. This limitation makes general DL methods unsuitable, and domain generalization (DG) becomes a key method to address this issue. To overcome the limitations of SEI identification performance across different scenarios, we propose a robust SEI method by mixing augmentation, named SigMix. Specifically, we innovatively introduce the Mixup method into the SEI task, mixing data from different source domains and then performing pairwise linear interpolation before using it for training the neural network. The SigMix method helps the model learn more comprehensive features by generating new samples in the training data, thereby improving the model’s generalization ability. To validate the effectiveness of the SigMix method, while also considering the impact of different receivers on identification performance, we evaluate a dataset spanning both time and receivers. The experimental results indicate that the average identification accuracy of the proposed SigMix method in unknown domains reaches 84.40%, significantly outperforming existing DG methods, demonstrating the robustness and generalization of our proposed SigMix method in SEI tasks. Our code is available for download at://github.com/frownean/SigMix.

Road traffic state estimation is an essential component of intelligent transportation systems (ITSs). However, some road sections lack fixed detectors, making it difficult to obtain complete traffic state data for the entire city. Thus, through the fusion of data collected from both the fixed and mobile detectors, we propose a novel model framework, a reasoning model based on the multisource causal interpretation graph (MS-CIG), to infer traffic state data for the road sections without detectors. First, a cross-layer random walk strategy is proposed to achieve fusion and embedding of road network topology graphs (constructed by the fixed detector data) and road network logic graphs (constructed by the mobile detector data). Second, the spatial similarity between road sections can be calculated by combining road sections static feature data, Point of Interest (PoI) distribution and embedding features; thereby, a weighted spatiotemporal graph of the road network is constructed. Finally, an interpretive module is combined to provide interpretability analysis for graph-based semi-supervised inference, thereby making the inferred data generated by graph-based semi-supervised more reasonable and accurate. Through the above processing, we are able to obtain the complete traffic state data for the entire city. Experimental evaluations on a real traffic data set demonstrate the superiority of the proposed method.

Specific emitter identification (SEI) is a passive physical layer authentication technology that mines subtle hardware differences between emitters to identify devices. However, traditional deep learning-based SEI is trained for scenarios with massive signal samples and performs poorly in sample-limited scenarios. To solve this problem, we proposed a robust few-shot SEI (FS-SEI) method using dual-level data augmentation, consisting of phase shift position prediction and manifold cutMix (P3MC). We perform data augmentation in both the sample space and the feature space to accelerate the complex valued time series lightweight adaptive network (CV-TSLANet) to learn robust features and use machine learning to identify ADS-B emitters. Our experimental results show that the performance of our proposed FS-SEI method reaches 90% when the number of samples per category is 30. We have open-sourced the proposed FS-SEI method at https://github.com/IcedWatermelonJuice/P3MC.

As the Internet rapidly expands, the increasing complexity and diversity of network activities pose significant challenges to effective network governance and security regulation. Network traffic, which serves as a crucial data carrier of network activities, has become indispensable in this process. Network traffic detection aims to monitor, analyze, and evaluate the data flows transmitted across the network to ensure network security and optimize performance. However, existing network traffic detection methods generally suffer from several limitations: 1) a narrow focus on characterizing traffic features from a single perspective; 2) insufficient exploration of discriminative features for different traffic; 3) poor generalization to different traffic scenarios. To address these issues, we propose a multi-view correlation-aware framework namedFlowIDfor network traffic detection.FlowIDcaptures multi-view traffic features via temporal and interaction awareness, while a hypergraph encoder further explores higher-order relationships between flows. To overcome the challenges of data imbalance and label scarcity, we design a dual-contrastive proxy task, enhancing the framework’s ability to differentiate between various traffic flows through flow-to-flow and group-to-group contrast. Extensive experiments on five real-world datasets demonstrate thatFlowIDsignificantly outperforms existing methods in accuracy, robustness, and generalization across diverse network scenarios, particularly in detecting malicious traffic.

Abstract Link prediction, which aims to infer missing edges or predict future edges based on currently observed graph connections, has emerged as a powerful technique for diverse applications such as recommendation, relation completion, etc. While there is rich literature on link prediction based on node representation learning, direct link embedding is relatively less studied and less understood. One common practice in previous work characterizes a link by manipulate the embeddings of its incident node pairs, which is not capable of capturing effective link features. Moreover, common link prediction methods such as random walks and graph auto-encoder usually rely on full-graph training, suffering from poor scalability and high resource consumption on large-scale graphs. In this paper, we propose Inductive Subgraph Embedding for Link Prediciton (SE4LP) — an end-to-end scalable representation learning framework for link prediction, which utilizes the strong correlation between central links and their neighborhood subgraphs to characterize links. We sample the “link-centric induced subgraphs” as input, with a subgraph-level contrastive discrimination as pretext task, to learn the intrinsic and structural link features via subgraph classification. Extensive experiments on five datasets demonstrate that SE4LP has significant superiority in link prediction in terms of performance and scalability, when compared with state-of-the-art methods. Moreover, further analysis demonstrate that introducing self-supervision in link prediction can significantly reduce the dependence on training data and improve the generalization and scalability of model.

Graph neural networks (GNNs) have achieved remarkable advances in graph-oriented tasks. However, real-world graphs invariably contain a certain proportion of heterophilous nodes, challenging the homophily assumption of traditional GNNs and hindering their performance. Most existing studies continue to design generic models with shared weights between heterophilous and homophilous nodes. Despite the incorporation of high-order messages or multi-channel architectures, these efforts often fall short. A minority of studies attempt to train different node groups separately but suffer from inappropriate separation metrics and low efficiency. In this paper, we first propose a new metric, termed Neighborhood Confusion (NC), to facilitate a more reliable separation of nodes. We observe that node groups with different levels of NC values exhibit certain differences in intra-group accuracy and visualized embeddings. These pave the way for Neighborhood Confusion-guided Graph Convolutional Network (NCGCN), in which nodes are grouped by their NC values and accept intra-group weight sharing and message passing. Extensive experiments on both homophilous and heterophilous benchmarks demonstrate that our framework can effectively separate nodes and yield significant performance improvement compared to the latest methods.

Recent years have seen a rapid growth of machine learning in cheminformatics problems. In order to tackle the problem of insufficient training data in reality, more and more researchers pay attention to data augmentation technology. However, few researchers pay attention to the problem of construction rules and domain information of data, which will directly impact the quality of augmented data and the augmentation performance. While in graph-based molecular research, the molecular connectivity index, as a critical topological index, can directly or indirectly reflect the topology-based physicochemical properties and biological activities. In this paper, we propose a novel data augmentation technique that modifies the topology of the molecular graph to generate augmented data with the same molecular connectivity index as the original data. The molecular connectivity index combined with data augmentation technology helps to retain more topology-based molecular properties information and generate more reliable data. Furthermore, we adopt five benchmark datasets to test our proposed models, and the results indicate that the augmented data generated based on important molecular topology features can effectively improve the prediction accuracy of molecular properties, which also provides a new perspective on data augmentation in cheminformatics studies.

Automatic Modulation Open Set Recognition (AMOSR) has practical significance in detecting unknown classes. However, a challenge arises when unknown samples closely resemble known samples, posing a formidable task for accurate detection. A novel AMOSR framework based on multi-view discriminators’ joint judgment is proposed to handle this challenge. Firstly, utilizing signal domain knowledge, multi-dimensional features are extracted through varied signal time-frequency transforms and encoders, baesd on which multiple discriminators are created. Secondly, Constrained Clustering Prototype Loss and Geodesic Contrastive Loss are introduced to pretrain these discriminators, providing more space for unknown signals. Then, collaborative learning is employed to further fine-tune the aforementioned discriminators, enhancing information sharing between modalities. Furthermore, a set of indicators is constructed, and multi-criteria fusion is performed using the TOPSIS algorithm to evaluate the discrimination capabilities of different classifiers in both closed-set and open-set scenarios. Furthermore, a decision tree is constructed to segregate test signals into known and unknown classes, in which discriminators with higher confidence levels are given precedence. Finally, TOPSIS hierarchical ensemble pruning algorithm that considers diversity and open-set recognition capabilities is adopted to reduce model complexity while maintaining original performance. Extensive experiments conducted on modulation datasets demonstrate the superiority of this framework over state-of-the-art AMOSR results.

Domain adaptation (DA) techniques are becoming increasingly proficient in cross-domain fault diagnosis tasks. However, DA-based methods are not always applicable due to the target domain data is not always accessible. Although there have been some interesting domain generalization methods for fault diagnosis under unseen conditions, most of them can only be used to mine the fault features on source domain distributions, and the improvement of model generalization performance is limited. To solve this problem, the multiplicative noise Gaussian perturbation strategy and the additive noise linear fusion strategy are proposed to capture fault information beyond source domain distributions. The former is used to randomly perturb feature statistics of multisource domains to simulate the uncertainty of domain shift, while the latter is used to perform the additive noise linear operation on feature statistics of multiple source domains to ensure the authenticity of the generated feature styles. Further, the feature statistics generated by both strategies are mixed with random convex weights to obtain new feature styles, achieving the best compromise between reliability and diversity. The network can learn more fault information from features with diversified styles. Extensive experimental results on both public and real datasets verify the effectiveness of our approach.

Due to the decentralized and public nature of the blockchain ecosystem, malicious activities on the Ethereum platform impose immeasurable losses on users. At the same time, the transparency of cryptocurrency transactions provides a unique opportunity to analyze illegal activities, such as phishing scams, from a network perspective. Most existing phishing scam detection methods focus primarily on analyzing account interaction networks, which limits their ability to uncover transaction behavior patterns embedded within transaction interactions. To address this, we construct theTransactionSubGraphNetwork (TSGN) by using transaction subgraphs as basic elements and further propose a novel framework for Ethereum phishing account detection. Specifically, we rebuild the graph structures via three well-designed mapping mechanisms, yielding TSGN and its two variants, i.e., Directed-TSGN and Temporal-TSGN, to obtain direction-aware and time-aware transfer flow features. By further incorporating the mapping strategy into transaction multidigraphs, we develop the Multiple-TSGN, which could preserve more transaction flow features while concurrently reducing the time consumption of modeling large-scale networks. TSGN models based on transaction subgraph interactions can capture complex higher-order dependencies, which lay beyond the reach of models that exclusively capture pairwise account interactions. As a general framework, our model can incorporate various feature extraction methods to improve the performance of phishing detection. Extensive experimental results on Ethereum datasets show that our method achieves superior performance in phishing detection, yielding 3.27%$\sim$6.71% relative improvement over previous state-of-the-art.

The rampant fraudulent activities on Ethereum hinder the healthy development of the blockchain ecosystem, necessitating the reinforcement of regulations. However, multiple imbalances involving account interaction frequencies and interaction types in the Ethereum transaction environment pose significant challenges to data mining-based fraud detection research. To address this, we first propose the concept of meta-interactions to refine interaction behaviors in Ethereum, and based on this, we present a dual self-supervision enhanced Ethereum fraud detection framework, named Meta-IFD. This framework initially introduces a generative self-supervision mechanism to augment the interaction features of accounts, followed by a contrastive self-supervision mechanism to differentiate various behavior patterns, and ultimately characterizes the behavioral representations of accounts and mines potential fraud risks through multi-view interaction feature learning. Extensive experiments on real Ethereum datasets demonstrate the effectiveness and superiority of our framework in detecting common Ethereum fraud behaviors such as Ponzi schemes and phishing scams. Additionally, the generative module can effectively alleviate the interaction distribution imbalance in Ethereum data, while the contrastive module significantly enhances the framework’s ability to distinguish different behavior patterns. The source code will be available inhttps://github.com/GISec-Team/Meta-IFD.

Specific emitter identification (SEI) is crucial in the Internet of Everything (IoE). Over the past decade, deep learning (DL) and broad learning (BL)-enabled SEI technologies have emerged. Both DL- and BL-based SEI methods rely on extensive radio frequency (RF) signal samples and corresponding labels, but labeling unknown signals is a considerable overhead and costly task. Consequently, many researchers have begun exploring semi-supervised learning techniques to address the semi-supervised SEI (SS-SEI) problem with limited labeled RF signals. However, existing SS-SEI solutions often prioritize identification performance, leading to high computational overheads and lacking iterability and scalability. To overcome these challenges, this paper proposes a novel SS-SEI solution, termed deep cloud and broad edge (DCBE). This approach integrates a DL-based SEI method at the cloud server with an updatable BL-based SEI method at the edge node. Initially, several DL-based SEI models are trained using labeled historical data at the cloud server. Meanwhile, an updatable BL-based SEI method is deployed locally on the edge node to identify unlabelled signals. When the DCBE solution is operational, edge nodes capture real-time unlabelled RF signals. The pre-trained DL-based SEI method and the locally BL-based SEI method jointly identify these RF signals. The identification results, along with the new real-time RF signals, are then used to update the weights of the BL-based SEI method at the edge nodes. The DCBE SS-SEI solution is validated using an open-source, large-scale, real-world automatic dependent surveillance-broadcast (ADS-B) dataset. Experimental results demonstrate that the proposed DCBE solution offers significant advantages in terms of SS-SEI performance, reduced computational overhead without GPU dependency, and system robustness in complex environments.

In the case of label space alignment, the existing domain adaptation (DA)-based fault diagnosis approaches have achieved high accuracy. In real industrial scenarios, however, the label space of the target domain is usually a subset of the label space of the source domain, called partial DA (PDA). The main challenge of PDA lies in how to separate common samples from private samples. In existing works, different weights are usually assigned to different samples based on the prediction score of the classifier, but the negative transfer caused by the data distribution alignment of private and common samples is ignored. To address this problem, class-consistency matching is proposed in this article, which uses label consensus score to identify classes in target clusters to discover common and private samples. In addition, parameter-free cosine attention wavelet blocks (PCAWBs) are designed to learn the complementary spatial-domain and frequency-domain features to enrich the domain-invariant features extracted by the shared encoder. Experiments on the real motor system demonstrate that the proposed method significantly outperforms state-of-the-art PDA fault diagnosis approaches.

As Industry 4.0 continues to evolve, the integration of cyber-physical systems (CPS) into modern engineering systems marks a significant paradigm shift that not only enhances operational efficiency but also opens up new possibilities for innovation and improved quality of life across various sectors. In intricate communication environments, the precise identification of device identities within CPS holds significant importance for augmenting reliability and robustness. For this purpose, we introduce a robust method for specific emitter identification that utilizes a deep residual shrinkage network, aimed at enhancing the model's ability to accurately recognize emitters, even when operating under conditions of low signal-to-noise ratios. This is an end-to-end recognition method that reduces the dependence on expert knowledge. Through the utilization of specially crafted subnetworks, adaptive thresholding is employed to enable each in-phase and quadrature (IQ) signal to possess its unique set of thresholds. The proposed approach reduces noise influence on the model by incorporating a soft threshold within the nonlinear transformation layer of the deep architecture. Experimental results using real-world data demonstrate that this method surpasses the performance of commonly utilized state-of-the-art specific emitter identification models.

Recently, the field of machine learning has undergone a transition from model-centric to data-centric. The advancements in diverse learning tasks have been propelled by the accumulation of more extensive datasets, subsequently facilitating the training of larger models on these datasets. However, these datasets remain relatively under-explored. To this end, we introduce a pioneering approach known as RK-core, to empower gaining a deeper understanding of the intricate hierarchical structure within datasets. Across several benchmark datasets, we find that samples with low coreness values appear less representative of their respective categories, and conversely, those with high coreness values exhibit greater representativeness. Correspondingly, samples with high coreness values make a more substantial contribution to the performance in comparison to those with low coreness values. Building upon this, we further employ RK-core to analyze the hierarchical structure of samples with different coreset selection methods. Remarkably, we find that a high-quality coreset should exhibit hierarchical diversity instead of solely opting for representative samples. The code is available at https://github.com/yaolu-zjut/Kcore.

In recent years, with the rise of deep learning, it has become a hot research topic to combine time-frequency analysis technology with deep learning to recognize radar signals. For the application of deep learning in radar signal recognition, however, the discovery of adversarial examples poses a tremendous security risk. Based on experiments, it appears that the radar signal recognition model based on the time-frequency image have been shown to be less vulnerable to adversarial attack methods based on time domain. Therefore, we propose a cross-modal attack (CMA). Firstly, we establish a surrogate model architecture locally, including three parts: time-frequency analysis, data quantization, and classifier. Secondly, we train this architecture as a whole and generate adversarial examples utilizing the trained surrogate model architecture parameters and adversarial attack methods. Finally, we carry out the CMA on the radar signal recognition model based on the time-frequency image by adding adversarial perturbations to the original signal. According to experimental results, the CMA can reduce the model recognition accuracy by more than 30%, demonstrating good attack performance, when the perturbation strength is 0.1 and the signal-to-noise ratio is 0 dB.

Along with the prevalent use of deep neural networks (DNNs), concerns have been raised on the security threats from DNNs such as backdoors in the network. While neural network repair methods have shown to be effective for fixing the defects in DNNs, they have been also found to produce biased models, with imbalanced accuracy across different classes, or weakened adversarial robustness, allowing malicious attackers to trick the model by adding small perturbations. To address these challenges, we propose INNER, an INterpretability-based NEural Repair approach. INNER formulates the idea of neuron routing for identifying fault neurons, in which the interpretability technique model probe is used to evaluate each neuron's contribution to the undesired behaviour of the neural network. INNER then optimizes the identified neurons for repairing the neural network. We test INNER on three typical application scenarios, including backdoor attacks, adversarial attacks, and wrong predictions. Our experimental results demonstrate that INNER can effectively repair neural networks, by ensuring accuracy, fairness, and robustness. Moreover, the performance of other repair methods can be also improved by re-using the fault neurons found by INNER, justifying the generality of the proposed approach.

Molecular property prediction faces the challenge of limited labeled data as it necessitates a series of specialized experiments to annotate target molecules. Data augmentation techniques can effectively address the issue of data scarcity. In recent years, Mixup has achieved significant success in traditional domains such as image processing. However, its application in molecular property prediction is relatively limited due to the irregular, non-Euclidean nature of graphs and the fact that minor variations in molecular structures can lead to alterations in their properties. To address these challenges, we propose a novel data augmentation method called Mix-Key tailored for molecular property prediction. Mix-Key aims to capture crucial features of molecular graphs, focusing separately on the molecular scaffolds and functional groups. By generating isomers that are relatively invariant to the scaffolds or functional groups, we effectively preserve the core information of molecules. Additionally, to capture interactive information between the scaffolds and functional groups while ensuring correlation between the original and augmented graphs, we introduce molecular fingerprint similarity and node similarity. Through these steps, Mix-Key determines the mixup ratio between the original graph and two isomers, thus generating more informative augmented molecular graphs. We extensively validate our approach on molecular datasets of different scales with several Graph Neural Network architectures. The results demonstrate that Mix-Key consistently outperforms other data augmentation methods in enhancing molecular property prediction on several datasets.

Graph neural networks (GNNs) have achieved remarkable success in various real-world applications. However, recent studies highlight the vulnerability of GNNs to malicious perturbations. Previous adversaries primarily focus on graph modifications or node injections to existing graphs, yielding promising results but with notable limitations. Graph modification attack (GMA) requires manipulation of the original graph, which is often impractical, while graph injection attack (GIA) necessitates training a surrogate model in the black-box setting, leading to significant performance degradation due to divergence between the surrogate architecture and the actual victim model. Furthermore, most methods concentrate on a single attack goal and lack a generalizable adversary to develop distinct attack strategies for diverse goals, thus limiting precise control over victim model behavior in real-world scenarios. To address these issues, we present a gradient-free generalizable adversary that injects a single malicious node to manipulate the classification result of a target node in the black-box evasion setting. Specifically, we model the single-node injection label specificity attack as a Markov decision process (MDP) and propose gradient-free generalizable single node injection attack, namely G2-SNIA, a reinforcement learning framework employing proximal policy optimization (PPO). By directly querying the victim model, G2-SNIA learns patterns from exploration to achieve diverse attack goals with extremely limited attack budgets. Through comprehensive experiments over three acknowledged benchmark datasets and four prominent GNNs in the most challenging and realistic scenario, we demonstrate the superior performance of our proposed G2-SNIA over the existing state-of-the-art baselines. Moreover, by comparing G2-SNIA with multiple white-box evasion baselines, we confirm its capacity to generate solutions comparable to those of the best adversaries.

Graph neural networks (GNNs) have found successful applications in various graph-related tasks. However, recent studies have shown that many GNNs are vulnerable to adversarial attacks. In a vast majority of existing studies, adversarial attacks on GNNs are launched via direct modification of the original graph such as adding/removing links, which may not be applicable in practice. In this article, we focus on a realistic attack operation via injecting fake nodes. The proposed global attack strategy via node injection (GANI) is designed under the comprehensive consideration of an unnoticeable perturbation setting from both structure and feature domains. Specifically, to make the node injections as imperceptible and effective as possible, we propose a sampling operation to determine the degree of the newly injected nodes, and then generate features and select neighbors for these injected nodes based on the statistical information of features and evolutionary perturbations obtained from a genetic algorithm, respectively. In particular, the proposed feature generation mechanism is suitable for both binary and continuous node features. Extensive experimental results on benchmark datasets against both general and defended GNNs show strong attack performance of GANI. Moreover, the imperceptibility analyses also demonstrate that GANI achieves a relatively unnoticeable injection on benchmark datasets.

Measuring the influence of scientists and their activities on science and society is important and indeed essential for many studies. Despite the substantial efforts devoted to exploring the influence’s measures and patterns of an individual scientific enterprise, it remains unclear how to quantify the mutual impact of multiple scientific activities. This work quantifies the relationship between the scientists’ interactive activities and their influences with different patterns in the AMiner dataset. Specifically, inflation treatment and field normalization are introduced to process the big data of paper citations as the scientist’s influence, and then the evolution of the influence is investigated for scientific activities in the citation and cooperation patterns through the Hawkes process. The results show that elite scientists have higher individual and interaction influences than ordinary scientists in all patterns found in the study, with permutation tests verifying the significance of the new findings. Moreover, the study compares the patterns found in two largest disciplines, i.e.,STEMandHumanities, revealing the higher value of individual influence inSTEMthan inHumanities. Furthermore, it is found that the opposite trend ofSTEMandHumanitiesin the cooperation pattern suggests different cooperation habits of scientists in different disciplines. Overall, this investigation provides a feasible approach to addressing the scientific influence issue and deepening the quantitative understanding of the mutual influence of multiple scientific activities in science and society.

Critical node detection (CND) is commonly used to detect nodes with a high impact on network robustness. It has been widely used in disease propagation, social networks, communications, and other fields. As a nondeterministic polynomial-time (NP)-complete problem, the efficiency of solving CND severely limits the scale of the available network. Fortunately, the evolutionary algorithm (EA) is an effective method to solve this problem. However, although EA improves the global search capability of the algorithm by preserving gene diversity, it also introduces many inferior genes, thus expanding the candidate solution space, reducing the search efficiency, and making it difficult to apply the pruning algorithm directly to its solution space. Hence, indirectly reducing the solution space of EA by deleting inferior genes is a feasible pruning method; however, the interaction of multiple genes affects the quality of CND solutions, making it a challenge to pick out inferior individual genes. Therefore, this work proposes a multipopulation synergistic gene screening algorithm based on the parallelism of EA and combined with Ensemble learning for identifying low-quality genes and removing them as a way of pruning the solution space of the algorithm and improving the search efficiency. The algorithm encodes all nodes in the graph as the gene pool of EA and treats a single population as a weak learner to screen the dominant genes in the gene pool and achieve fast pruning of EA’s solution space by integrating the dominant individuals in multiple populations. In this work, the experiments demonstrate the effectiveness of the proposed method and analyze the effect of different network structures on the algorithm.

Graph neural network (GNN) has achieved remarkable success in various graph learning tasks, such as node classification, link prediction, and graph classification. The key to the success of GNN lies in its effective structure information representation through neighboring aggregation. However, the attacker can easily perturb the aggregation process through injecting fake nodes, which reveals that GNN is vulnerable to the graph injection attack (GIA). Existing GIA methods primarily focus on damaging the classical feature aggregation process while overlooking the neighborhood aggregation process via label propagation. To bridge this gap, we propose the label-propagation-based global injection attack (LPGIA) which conducts the GIA on the node classification task. Specifically, we analyze the aggregation process from the perspective of label propagation and transform the GIA problem into a global injection label specificity attack problem. To solve this problem, LPGIA utilizes a label-propagation-based strategy to optimize the combinations of the nodes connected to the injected node. Then, LPGIA leverages the feature mapping to generate malicious features for injected nodes. In extensive experiments against representative GNNs, LPGIA outperforms the previous best-performing injection attack method in various datasets, demonstrating its superiority and transferability.

With the rapid development of artificial intelligence, a number of machine learning algorithms, such as graph neural networks (GNNs), have been proposed to facilitate network analysis or graph data mining. Although effective, recent studies show that these advanced methods may suffer from adversarial attacks, i.e., they may lose effectiveness when only a small fraction of links are unexpectedly changed. This article investigates three well-known adversarial attack methods, i.e., Nettack, Meta Attack, and GradArgmax. It is found that different attack methods have their specific attack preferences on changing the target network structures. Such attack patterns are further verified by experimental results on some real-world networks, revealing that, generally, the top-4 most important network attributes on detecting adversarial samples suffice to explain the preference of an attack method. Based on these findings, the network attributes are utilized to design machine learning models for adversarial sample detection and attack method recognition with outstanding performance.

Automatic modulation recognition (AMR) of radio signal is an important research topic in the area of non-cooperative communication and cognitive radio. Recently deep learning (DL) techniques enable significant progress in AMR. However, the techniques of adversarial machine learning cause the threats of adversarial attacks in DL-based AMR. In this paper, we aim to make AMR model robust, accurate and lightweight, thus propose a multi-distillation mechanism for robust training of DL-based AMR models, namely Adversarial Multi-Distillation (AMD). In the framework of AMD, by knowledge distillation, two powerful teacher models transfer the learned classification knowledge and defense knowledge, respectively, to the student model to form robust training. Our experiments with public dataset RML2016.10a show that the proposed method can significantly improve the defense of AMR models to against adversarial perturbations and keep relatively high classification accuracy, which enables robust decision making with lightweight models under adversarial attacks.

The backdoor attacks have posed a severe threat to deep neural networks (DNNs). Online training platforms and third-party model training providers are more vulnerable to backdoor attacks due to uncontrollable data sources, untrusted developers or unmonitorable training processes. Researchers have proposed to detect the backdoor in the well-trained models, and then remove them by some mitigation techniques, e.g., retraining and pruning. However, they are still limited from two aspects: (i) real-time - they cannot detect in time at the beginning of training due to their reliance on well-trained models; (ii) mitigation effect - the later discovery of backdoors usually leads to 1) deeper backdoors, 2) less effective mitigation, and 3) greater costs. To address these challenges, we rethink the evolution of the backdoor, and intend to cope with backdoors along with the online training process, that is to detect the backdoors sooner rather than later. We propose BackdoorTracer, a novel framework that detects the backdoor in the training phase. BackdoorTracer constructs the model into an equivalent graph based on the activated neural path during training, thereby detecting the backdoor through multiple graph metrics. BackdoorTracer can incorporate any existing backdoor mitigation approaches that require accessing training to stop the impact of backdoors as soon as possible. It differs from previous works in several key aspects: (i) lightweight - BackdoorTracer is independent of the training process, and thus it has little negative impact on the training efficiency and testing accuracy; (ii) generalizable - it works different modalities of data, models and different backdoor attacks. BackdoorTracer outperforms the state-of-the-art (SOTA) detection approaches in experiments on 5 modes, 10 models and 9 backdoor attack scenarios. Compared with the existing 5 backdoor detection methods, our method can detect backdoors earlier ($\sim ~1.5$epochs) and higher detection rate (~ +10%), effectively improving the effectiveness of backdoor defense (ASR. ~ -78%, ACC. +47%). Finally, we make BackdoorTracer a plug-and-play backdoor detector, which enables real-time backdoor tracing in the training phase.

Deep learning technology has found a promising application in lightweight model design, for which pruning is an effective means of achieving a large reduction in both model parameters and float points operations (FLOPs). The existing neural network pruning methods mostly start from the consideration of the importance of model parameters and design parameter evaluation metrics to perform parameter pruning iteratively. These methods were not studied from the perspective of network model topology, so they might be effective but not efficient, and they require completely different pruning for different datasets. In this article, we study the graph structure of the neural network and propose a regular graph pruning (RGP) method to perform a one-shot neural network pruning. Specifically, we first generate a regular graph and set its node-degree values to meet the preset pruning ratio. Then, we reduce the average shortest path-length (ASPL) of the graph by swapping edges to obtain the optimal edge distribution. Finally, we map the obtained graph to a neural network structure to realize pruning. Our experiments demonstrate that the ASPL of the graph is negatively correlated with the classification accuracy of the neural network and that RGP has a strong precision retention capability with high parameter reduction (more than 90%) and FLOPs reduction (more than 90%) (the code for quick use and reproduction is available at https://github.com/Holidays1999/Neural-Network-Pruning-through-its-RegularGraph-Structure).

Graph neural networks (GNNs) have been proved powerful in graph-oriented tasks. However, many real-world graphs are heterophilous, challenging the homophily assumption of classical GNNs. To solve the universality problem, many studies deepen networks or concatenate intermediate representations, which does not inherently change neighbor aggregation and introduces noise. Recent studies propose new metrics to characterize the homophily, but rarely consider the correlation of the proposed metrics and models. In this paper, we first design a new metric, Neighborhood Homophily (NH), to measure the label complexity or purity in node neighborhoods. Furthermore, we incorporate the metric into the classical graph convolutional network (GCN) architecture and propose Neighborhood Homophily-based Graph Convolutional Network (NHGCN). In this framework, neighbors are grouped by estimated NH values and aggregated from different channels, and the resulting node predictions are then used in turn to estimate and update NH values. The two processes of metric estimation and model inference are alternately optimized to achieve better node classification. NHGCN achieves top overall performance on both homophilous and heterophilous benchmarks, with an improvement of up to 7.4% compared to the current SOTA methods.

Despite the popularization of deep neural networks (DNNs) in many fields, it is still challenging to deploy state-of-the-art models to resource-constrained devices due to high computational overhead. Model pruning provides a feasible solution to the aforementioned challenges. However, the interpretation of existing pruning criteria is always overlooked. To counter this issue, we propose a novel layer pruning method by exploring the Stochastic Re-initialization. Our SR-init method is inspired by the discovery that the accuracy drop due to stochastic re-initialization of layer parameters differs in various layers. On the basis of this observation, we come up with a layer pruning criterion, i.e., those layers that are not sensitive to stochastic re-initialization (low accuracy drop) produce less contribution to the model and could be pruned with acceptable loss. Afterward, we experimentally verify the interpretability of SR-init via feature visualization. The visual explanation demonstrates that SR-init is theoretically feasible, thus we compare it with state-of-the-art methods to further evaluate its practicability. As for ResNet56 on CIFAR-10 and CIFAR-100, SR-init achieves a great reduction in parameters (63.98% and 37.71%) with an ignorable drop in top-1 accuracy (-0.56% and 0.8%). With ResNet50 on ImageNet, we achieve a 15.59% FLOPs reduction by removing 39.29% of the parameters, with only a drop of 0.6% in top-1 accuracy. Our code is available at https://github.com/huitang-zjut/SR-init.

The concept of$k$-core in complex networks plays a key role in many applications, e.g., understanding the global structure or identifying central/critical nodes, of a network. A malicious attacker with a jamming ability can exploit the vulnerability of the$k$-core structure to attack the network and invalidate the network analysis methods, e.g., reducing the$k$-shell values of nodes can deceive graph algorithms, leading to the wrong decisions. In this article, we investigate the robustness of the$k$-core structure under adversarial attacks by deleting edges, for the first time. First, we give the general definition of the targeted$k$-core attack, map it to the set cover problem, which is NP-hard, and further introduce a series of evaluation metrics to measure the performance of attack methods. Then, we propose the$Q$index theoretically as the probability that the terminal node of an edge does not belong to the innermost core, which is further used to guide the design of our heuristic attack methods, namely, COREATTACK and GreedyCOREATTACK. The experiments on a variety of real-world networks demonstrate that our methods behave much better than a series of baselines, in terms of much smaller edge change rate (ECR) and false attack rate (FAR), achieving state-of-the-art attack performance. More impressively, for certain real-world networks, only deleting one edge from the$k$-core may lead to the collapse of the innermost core, even if this core contains dozens of nodes. Such a phenomenon indicates that the$k$-core structure could be extremely vulnerable under adversarial attacks, and its robustness, thus, should be carefully addressed to ensure the security of many graph algorithms. An open-source implementation is available athttps://github.com/Yocenly/COREATTACCK.

In network link prediction, it is possible to hide a target link from being predicted with a small perturbation on network structure. This observation may be exploited in many real world scenarios, for example, to preserve privacy, or to exploit financial security. There have been many recent studies to generate adversarial examples to mislead deep learning models on graph data. However, none of the previous work has considered the dynamic nature of real-world systems. In this work, we present the first study of adversarial attack on dynamic network link prediction (DNLP). The proposed attack method, namely time-aware gradient attack (TGA), utilizes the gradient information generated by deep dynamic network embedding (DDNE) across different snapshots to rewire a few links, so as to make DDNE fail to predict target links. We implement TGA in two ways: one is based on traversal search, namely TGA-Tra; and the other is simplified with greedy search for efficiency, namely TGA-Gre. We conduct comprehensive experiments which show the outstanding performance of TGA in attacking DNLP algorithms.

Community detection, which focuses on clustering vertex interactions, plays a significant role in network analysis. However, it also faces numerous challenges like missing data and adversarial attack. How to further improve the performance and robustness of community detection for real-world networks has raised great concerns. In this paper, we explore robust community detection by enhancing network structure, with two generic algorithms presented: one is named robust community detection via genetic algorithm (RobustECDGA), in which the modularity and the number of clusters are combined in a fitness function to find the optimal structure enhancement scheme; the other is called robust community detection via similarity ensemble (RobustECD-SE), integrating multiple information of community structures captured by various vertex similarities, which scales well on large-scale networks. Comprehensive experiments on real-world networks demonstrate, by comparing with two traditional enhancement strategies, that the new methods help six representative community detection algorithms achieve more significant performance improvement. Moreover, experiments on the corresponding adversarial networks indicate that the new methods could also optimize the network structure to a certain extent, achieving stronger robustness against adversarial attack. The source code of this paper is released on https://github.com/jjzhou012/robustECD release.

Recently, the rapid diffusion of malicious information in online social networks causes great harm to our society. Therefore, it is of great significance to localize diffusion sources as early as possible to stem the spread of malicious information. This paper proposes a novel sensor-based method, called greedy full-order neighbor localization (denoted as GFNL), to solve this problem under a low infection propagation in line with the real world. More specifically, GFNL includes two main components, i.e., the greedy-based sensor deployment strategy (DS) and direction-path-based source estimation strategy (ES). In more detail, to ensure sensors can observe a propagation information as early as possible, a set of sensors is deployed in a network to minimize the geodesic distance (i.e., the distance of the shortest path) between the candidate set and the sensor set based on DS. Then when a fraction of sensors observe a propagation, ES infers the source based on the idea that the distance of the actual propagation path is proportional to the observed time. Compared with some state-of-the-art methods, comprehensive experiments have proved the superiority and robustness of our proposed GFNL.

The last decade has witnessed the rapid growth of open-source software (OSS). Still, all contributors may find it difficult to assimilate into the OSS community even they are enthusiastic to make contributions. We thus suggest that partner recommendation across different roles may benefit both the users and developers, i.e., once we are able to make successful recommendation for those in need, it may dramatically contribute to the productivity of developers and the enthusiasm of users, thus further boosting OSS projects’ development. Motivated by this potential, we model the partner recommendation as link prediction task from email data via network embedding methods. In this article, we introduce time-series snapshot network (TSSN) that is a mixture network to model the interactions among users and developers. Based on the established TSSN, we perform temporal biased walk (TBW) to automatically capture both temporal and structural information of the email network, i.e., the behavioral similarity between individuals in the OSS email network. Experiments on ten Apache data sets demonstrate that the proposed TBW significantly outperforms a number of advanced random walk-based embedding methods, leading to the state-of-the-art recommendation performance.

Prognostic prediction is the task of estimating a patient's risk of disease development based on various predictors. Such prediction is important for healthcare practitioners and patients because it reduces preventable harm and costs. As such, a prognostic prediction model is preferred if: 1) it exhibits encouraging performance and 2) it can generate intelligible rules, which enable experts to understand the logic of the model's decision process. However, current studies usually concentrated on only one of the two features. Toward filling this gap, in the present study, we develop a novel knowledge-aware Bayesian model taking into consideration accuracy and transparency simultaneously. Real-world case studies based on four years' territory-wide electronic health records are conducted to test the model. The results show that the proposed model surpasses state-of-the-art prognostic prediction models in accuracy and c-statistic. In addition, the proposed model can generate explainable rules.

Blockchain technology has the characteristics of decentralization, traceability and tamper-proof, which creates a reliable decentralized trust mechanism, further accelerating the development of blockchain finance. However, the anonymization of blockchain hinders market regulation, resulting in increasing illegal activities such as money laundering, gambling and phishing fraud on blockchain financial platforms. Thus, financial security has become a top priority in the blockchain ecosystem, calling for effective market regulation. In this paper, we consider identifying Ethereum accounts from a graph classification perspective, and propose an end-to-end graph neural network framework namedEthident, to characterize the behavior patterns of accounts and further achieve account de-anonymization. Specifically, we first construct an Account Interaction Graph (AIG) using raw Ethereum data. Then we design a hierarchical graph attention encoder namedHGATEas the backbone of our framework, which can effectively characterize the node-level account features and subgraph-level behavior patterns. For alleviating account label scarcity, we further introduce contrastive self-supervision mechanism as regularization to jointly train our framework. Comprehensive experiments on Ethereum datasets demonstrate that our framework achieves superior performance in account identification, yielding 1.13% ~ 4.93% relative improvement over previous state-of-the-art. Furthermore, detailed analyses illustrate the effectiveness ofEthidentin identifying and understanding the behavior of known participants in Ethereum (e.g. exchanges, miners, etc.), as well as that of the lawbreakers (e.g. phishing scammers, hackers, etc.), which may aid in risk assessment and market regulation.

Deep Neural Networks (DNN) are known to be vulnerable to adversarial samples, the detection of which is crucial for the wide application of these DNN models. Recently, a number of deep testing methods in software engineering were proposed to find the vulnerability of DNN systems, and one of them, i.e., Model Mutation Testing (MMT), was used to successfully detect various adversarial samples generated by different kinds of adversarial attacks. However, the mutated models in MMT are always huge in number (e.g., over 100 models) and lack diversity (e.g., can be easily circumvented by high-confidence adversarial samples), which makes it less efficient in real applications and less effective in detecting high-confidence adversarial samples. In this study, we propose Graph-Guided Testing (GGT) for adversarial sample detection to overcome these aforementioned challenges. GGT generates pruned models with the guide of graph characteristics, each of them has only about 5% parameters of the mutated model in MMT, and graph guided models have higher diversity. The initial experiments on CIFAR10 validate that GGT performs much better than MMT with respect to both effectiveness and efficiency.

Open Source Software success and sustainability is critically important for the digital infrastructure as OSS is used broadly and yet 83+% of such projects fail. To increase chances of success many projects join established software communities, e.g. the Apache Software Foundation (ASF), with clearly established rules and support. Specifically at ASF, projects that strive to join ASF and are at a nascent development stage are digitally housed in the ASF incubator (ASFI), which provides a mature governance environment and expert help toward long-term sustainability. Projects in ASFI eventually conclude their incubation by graduating, if successful on the path to sustainability. Otherwise, they get retired. In ASF, digital traces of developer activities for projects in ASFI are publicly available, together with monthly project status.Here we present a longitudinal dataset of developer coding and communication activities of 269 projects from the Apache Software Foundation Incubator (ASFI). Each project in ASFI is evaluated while in incubation and is eventually "graduated" or "retired", a label indicating the project sustainability promise with respect to their technical development and community diversity. This extrinsically labeled dataset offers heretofore unavailable sustainability data of OSS project development under ASF regulations and governance. We hope its availability will foster more research interest in studying sustainability in OSS projects.

Although OSS development is very popular, ultimately more than 80% of OSS projects fail. Identifying the factors associated with OSS success can help in devising interventions when a project takes a downturn. OSS success has been studied from a variety of angles, more recently in empirical studies of large numbers of diverse projects, using proxies for sustainability, e.g., internal metrics related to productivity and external ones, related to community popularity. The internal socio-technical structure of projects has also been shown important, especially their dynamics. This points to another angle on evaluating software success, from the perspective of self-sustaining and self-governing communities.

Community detection, aiming to group nodes based on their connections, plays an important role in network analysis since communities, treated as meta-nodes, allow us to create a large-scale map of a network to simplify its analysis. However, for privacy reasons, we may want to prevent communities from being discovered in certain cases, leading to the topics on community deception. In this article, we formalize this community detection attack problem in three scales, including global attack (macroscale), target community attack (mesoscale), and target node attack (microscale). We treat this as an optimization problem and further propose a novel evolutionary perturbation attack (EPA) method, where we generate adversarial networks to realize the community detection attack. Numerical experiments validate that our EPA can successfully attack network community algorithms in all three scales, i.e., hide target nodes or communities and further disturb the community structure of the whole network by only changing a small fraction of links. By comparison, our EPA behaves better than a number of baseline attack methods on six synthetic networks and three real-world networks. More interestingly, although our EPA is based on the Louvain algorithm, it is also effective in attacking other community detection algorithms, validating its good transferability.

The adversarial attack methods based on gradient information can adequately find the perturbations, that is, the combinations of rewired links, thereby reducing the effectiveness of the deep learning model-based graph embedding algorithms, but it is also easy to fall into a local optimum. Therefore, this article proposes a momentum gradient attack (MGA) against the graph convolutional network (GCN) model, which can achieve more aggressive attacks with fewer rewiring links. Compared with directly updating the original network using gradient information, integrating the momentum term into the iterative process can stabilize the updating direction, which makes the model jump out of poor local optimum and enhances the method with stronger transferability. Experiments on node classification and community detection methods based on three well-known network embedding algorithms show that MGA has a better attack effect and transferability.

Recently, a graph neural network (GNN) was proposed to analyze various graphs/networks, which has been proven to outperform many other network analysis methods. However, it is also shown that such state-of-the-art methods suffer from adversarial attacks, i.e., carefully crafted adversarial networks with slight perturbation on clean one may invalid these methods on lots of applications, such as network embedding, node classification, link prediction, and community detection. Adversarial training has been testified as an efficient defense strategy against adversarial attacks in computer vision and graph mining. However, almost all the algorithms based on adversarial training focus on global defense through overall adversarial training. In a more practical scene, certain users would be targeted to attack, i.e., specific labeled users. It is still a challenge to defend against target node attack by existing adversarial training methods. Therefore, we propose smoothing adversarial training (SAT) to improve the robustness of GNNs. In particular, we analytically investigate the robustness of graph convolutional network (GCN), one of the classic GNNs, and propose two smooth defensive strategies: smoothing distillation and smoothing cross-entropy loss function. Both of them smooth the gradients of GCN and, consequently, reduce the amplitude of adversarial gradients, benefiting gradient masking from attackers in both global attack and target label node attack. The comprehensive experiments on five real-world networks testify that the proposed SAT method shows state-of-the-art defensibility against different adversarial attacks on node classification and community detection. Especially, the average attack success rate of different attack methods can be decreased by about 40% by SAT at the cost of tolerable embedding performance decline of the original network.

Real-world networks exhibit prominent hierarchical and modular structures, with various subgraphs as building blocks. Most existing studies simply consider distinct subgraphs as motifs and use only their numbers to characterize the underlying network. Although such statistics can be used to describe a network model, or even to design some network algorithms, the role of subgraphs in such applications can be further explored so as to improve the results. In this article, the concept of subgraph network (SGN) is introduced and then applied to network models, with algorithms designed for constructing the 1st-order and 2nd-order SGNs, which can be easily extended to build higher-order ones. Furthermore, these SGNs are used to expand the structural feature space of the underlying network, beneficial for network classification. Numerical experiments demonstrate that the network classification model based on the structural features of the original network together with the 1st-order and 2nd-order SGNs always performs the best as compared to the models based only on one or two of such networks. In other words, the structural features of SGNs can complement that of the original network for better network classification, regardless of the feature extraction method used, such as the handcrafted, network embedding and kernel-based methods.

In social networks, by removing some target-sensitive links, privacy protection might be achieved. However, some hidden links can still be re-observed by link prediction methods on observable networks. In this paper, the conventional link prediction method named Resource Allocation Index (RA) is adopted for privacy attacks. Several defense methods are proposed, including heuristic and evolutionary approaches, to protect targeted links from RA attack. In particular, incremental computation is proposed for accelerating the calculation of fitness in evolutionary approaches. This is the first time to study privacy protection for targeted links against similarity based link prediction attacks. Some links are randomly selected from original network as targeted links for experimentation. The experimental results on nine real-world networks demonstrate the superiority of the evolutionary perturbations, especially EDA, for defending against RA attack. Moreover, experimental results show that the proposed perturbation generated by EDA is transferable and can even defend against other link prediction attacks which are based on high order similarity between pairwise nodes, although it is designed to prevent RA attack.

Predicting the potential relations between nodes in networks, known as link prediction, has long been a challenge in network science. However, most studies just focused on link prediction of static network, while real-world networks always evolve over time with the occurrence and vanishing of nodes and links. Dynamic network link prediction (DNLP) thus has been attracting more and more attention since it can better capture the evolution nature of networks, but still most algorithms fail to achieve satisfied prediction accuracy. Motivated by the excellent performance of long short-term memory (LSTM) in processing time series, in this article, we propose a novel encoder-LSTM-decoder (E-LSTM-D) deep learning model to predict dynamic links end to end. It could handle long-term prediction problems, and suits the networks of different scales with fine-tuned structure. To the best of our knowledge, it is the first time that LSTM, together with an encoder-decoder architecture, is applied to link prediction in dynamic networks. This new model is able to automatically learn structural and temporal features in a unified framework, which can predict the links that never appear in the network before. The extensive experiments show that our E-LSTM-D model significantly outperforms newly proposed DNLP methods and obtain the state-of-the-art results.

Link prediction has long been the focus in the analysis of network-structured data. Though straightforward and efficient, heuristic approaches like Common Neighbors perform link prediction with pre-defined assumptions and only use superficial structural features. While it is widely acknowledged that a vertex could be characterized by a bunch of neighbor vertices, network embedding algorithms and newly emerged graph neural networks still exploit structural features on the whole network, which may inevitably bring in noises and limits the scalability of those methods. In this paper, we propose an end-to-end deep learning framework, namely hyper-substructure enhanced link predictor (HELP), for link prediction. HELP utilizes local topological structures from the neighborhood of the given vertex pairs, avoiding useless features. For further exploiting higher-order structural information, HELP also learns features from hyper-substructure network (HSN).Extensive experiments on six benchmark datasets have shown the state-of-the-art performance of HELP on link prediction.

Graph classification, which aims to identify the category labels of graphs, plays a significant role in drug classification, toxicity detection, protein analysis etc. However, the limitation of scale of benchmark datasets makes it easy for graph classification models to fall into over-fitting and undergeneralization. Towards this, we introduce data augmentation on graphs and present two heuristic algorithms: \emrandom mapping and \emmotif-similarity mapping, to generate more weakly labeled data for small-scale benchmark datasets via heuristic modification of graph structures. Furthermore, we propose a generic model evolution framework, named \emM-Evolve, which combines graph augmentation, data filtration and model retraining to optimize pre-trained graph classifiers. Experiments conducted on six benchmark datasets demonstrate that \emM-Evolve helps existing graph classification models alleviate over-fitting when training on small-scale benchmark datasets and %achieve significant improvement of classification performance. yields an average improvement of 3-12% accuracy on graph classification tasks.

Software defect prediction aims to automatically locate defective code modules to better focus testing resources and human effort. Typically, software defect prediction pipelines are comprised of two parts: the first extracts program features, like abstract syntax trees, by using external tools, and the second applies machine learning-based classification models to those features in order to predict defective modules. Since such approaches depend on specific feature extraction tools, machine learning classifiers have to be custom-tailored to effectively build most accurate models.

Community detection plays an important role in social networks, since it can help to naturally divide the network into smaller parts so as to simplify network analysis. However, on the other hand, it arises the concern that individual information may be overmined, and the concept community deception has been proposed to protect individual privacy on social networks. Here, we introduce and formalize the problem of community detection attack and develop efficient strategies to attack community detection algorithms by rewiring a small number of connections, leading to privacy protection. In particular, we first give two heuristic attack strategies, i.e., Community Detection Attack (CDA) and Degree Based Attack (DBA), as baselines, utilizing the information of detected community structure and node degree, respectively. Then, we propose an attack strategy called “genetic algorithm (GA)-based Q-Attack,” where the modularity Q is used to design the fitness function. We launch community detection attack based on the above three strategies against six community detection algorithms on several social networks. By comparison, our Q-Attack method achieves much better attack effects than CDA and DBA, in terms of the larger reduction of both modularity Q and normalized mutual information (NMI). In addition, we further take transferability tests and find that adversarial networks obtained by Q-Attack on a specific community detection algorithm also show considerable attack effects while generalized to other algorithms.

Recommender systems are becoming more and more important in our daily lives. However, traditional recommendation methods are challenged by data sparsity and efficiency, as the numbers of users, items, and interactions between the two in many real-world applications increase fast. In this paper, we propose a novel clustering recommender system based on node2vec technology and rich information network, namely, N2VSCDNNR, to solve these challenges. In particular, we use a bipartite network to construct the user-item network and represent the interactions among users (or items) by the corresponding one-mode projection network. In order to alleviate the data sparsity problem, we enrich the network structure according to user and item categories, and construct the one-mode projection category network. Then, considering the data sparsity problem in the network, we employ node2vec to capture the complex latent relationships among users (or items) from the corresponding one-mode projection category network. Moreover, considering the dependence on parameter settings and information loss problem in clustering methods, we use a novel spectral clustering method, which is based on dynamic nearest-neighbors (DNNs) and a novel automatically determining cluster number (ADCN) method that determines the cluster centers based on the normal distribution method, to cluster the users and items separately. After clustering, we propose the two-phase personalized recommendation to realize the personalized recommendation of items for each user. A series of experiments validate the outstanding performance of our N2VSCDNNR over several advanced embedding and side information based recommendation algorithms. Meanwhile, N2VSCDNNR seems to have lower time complexity than the baseline methods in online recommendations, indicating its potential to be widely applied in large-scale systems.

Network embedding plays an important role in various network applications, such as node classification and link prediction. Lots of structure-based network embedding methods have been proposed. Yet, they suffer from unsteady embedding performances due to the parameter sensitivity. How to extract valuable attribute information in networks with less parameter influence is still a challenge. In this paper, we propose a novel algorithm, named adaptive network embedding with particle swarm optimization (PSO-ANE), which is based on the second-order dynamic random walk and PSO method, for learning network representations. A second-order dynamic random walk is designed to search a suitable strategy for each node based on the structure-based transition probability, the centrality-based transition probability, and the static link weights. To reduce the parameter dependence, PSO is adopted for key-parameter optimization to get global steady network embedding. The experiments validate that the proposed method outperforms the existing state-of-the-art techniques on multilabel classification, multiclass classification, and link prediction tasks.

With the improvement of living standards, people pay more attention to health, which is significant to analyze people's hospitalizing behaviors. The wide use of mobile devices generates a great deal of data, which contains a lot of travel information about residents. Many people would like to see a doctor through calling an online car hailing for its convenience. Thus, based on big trajectory data generated by the online car hailing, the hospitalizing behaviors of residents are analyzed in this paper. The hospitalizing behaviors are analyzed from two aspects. One is performed from the temporal aspect, in which the daily numbers of trips of hospitalizing behaviors under different modes are analyzed. The other one is performed from the spatial aspect, in which the hot hospitals, popularity, and gravity distribution of hospitals are analyzed. Based on the spatial analysis, the network constructed by the hot hospitals is also analyzed. The results show that the hospitalizing behavior analysis can reflect the hospitalizing behaviors in detail, which can make contributions to the decision-making of infrastructure configuration for institutions, such as urban planning departments and hospitals.

Open source software (OSS) projects and communities are becoming increasingly popular and influential recently. Communications and collaborations are essential for the success of projects. Usually, the most active and productive programmers are awarded with promotion to developers. To more effectively manage and progress the projects, it is important and beneficial to rank the programmers and thus, predict the developer candidates. In this work, we propose to combine machine learning techniques with existing complex network node ranking algorithms to improve the prediction results. Specifically, we have made the following contributions: (1), we have designed a novel machine learning-based classifier with significantly improved prediction performance; (2), we have constructed and tested various networks built based on the programmer email communication information; and (3), we have used real-world project data to compare different techniques and validate our methods. Experimental results demonstrate that our technique reduces the error rate by 25% compared with the second best. Moreover, we discover that the [Formula: see text] nearest neighbor (KNN)-based machine learning algorithm and non-directional temporal network with a time window of 1–3 months give the best prediction results.

Animals search for food based on certain optimal principles and over time form foraging patterns effective for survival in changing environments. Due to the many choices available in modern society, we also face a decision on where to get their food. We call this “modern human food foraging,” since the Internet makes foraging much more convenient than before. People search online for food venues, or restaurants, through websites such as Yelp, and write reviews for the food they tasted, which in turn, facilitate others' searches in the future. These activities make the whole community of restaurant patrons wiser over time. Moreover, the archives of all these choices and evaluations are publicly available, and can help researchers better understand human foraging patterns in modern society. In this paper, we use a Yelp data set to study modern human food foraging patterns, with respect to both geography and cuisine. To understand spatial patterns, we cluster reviewed restaurants geographically and construct a taste similarity network, representing the topology of restaurant cuisine space. We find that people steadily expand their foraging domains from the nearest to them to the distant in geography and from the most familiar to the novel in cuisine. Using longitudinal data of restaurant reviews, we build a geographical foraging network and a taste foraging network for each patron based on which, we propose three kinds of entropies to characterize foraging patterns. We show that the modern foraging patterns of restaurant patrons in both geography and cuisine are of high regularity, indicating that their behaviors are rather predictable. The foraging patterns are also associated with individual social status in the community. Namely, people having a higher variety in the restaurant cuisines they have visited, but fewer actual locations they visited, tend to attract more followers.

Social synchrony (SS) is an emergent phenomenon in human society. People often mimic others which, over time, can result in large groups behaving similarly. Drawing from prior empirical studies of SS in online communities, here we propose a discrete network model of SS based on four attributes: 1) depth of action; 2) breadth of impact, i.e., a large number of actions are performed with a large group of people involved; 3) heterogeneity of role, i.e., people of higher degree play more important roles; and 4) lastly, emergence of phenomenon, i.e., it is far from random. We analyze our model both analytically and with simulations, and find good agreement between the two. We find this model can well explain the four characters of SS, and thus hope it can help researchers better understand human collective behavior.

Real-world networks feature weights of interactions, where link weights often represent some physical attributes. In many situations, to recover the missing data or predict the network evolution, we need to predict link weights in a network. In this paper, we first proposed a series of new centrality indices for links in line graph. Then, utilizing these line graph indices, as well as a number of original graph indices, we designed three supervised learning methods to realize link weight prediction both in the networks of single layer and multiple layers, which perform much better than several recently proposed baseline methods. We found that the resource allocation index (RA) plays a more important role in the weight prediction than other topological properties, and the line graph indices are at least as important as the original graph indices in link weight prediction. In particular, the success application of our methods on Yelp layered network suggests that we can indeed predict the offline co-foraging behaviors of users just based on their online social interactions, which may open a new direction for link weight prediction algorithms, and meanwhile provide insights to design better restaurant recommendation systems.

In this paper, the swarming behavior of multiple Euler-Lagrange systems with cooperation-competition interactions is investigated, where the agents can cooperate or compete with each other and the parameters of the systems are uncertain. The distributed stabilization problem is first studied, by introducing an auxiliary system to each agent, where the common assumption that the cooperation-competition network satisfies the digon sign-symmetry condition is removed. Based on the input-output property of the auxiliary system, it is found that distributed stabilization can be achieved provided that the cooperation subnetwork is strongly connected and the parameters of the auxiliary system are chosen appropriately. Furthermore, as an extension, a distributed consensus tracking problem of the considered multiagent systems is discussed, where the concept of equi-competition is introduced and a new pinning control strategy is proposed based on the designed auxiliary system. Finally, illustrative examples are provided to show the effectiveness of the theoretical analysis.

Passive indoor localization is important. Unlike active localization techniques, it does not require for users to carry measuring devices, e.g., smart phones. Thus, it is widely used in applications such as security, smart housing, object tracking, etc. However, in real-world applications, the passive localization accuracy is limited due to the environment noises, multipath effect, etc. To address those problems, in this paper, we propose to use channel state information (CSI) instead. Specifically, we make the following contributions: 1) we design a CSI-based passive indoor localization system; 2) we develop a Naive Bayes classifier enhanced with confidence level information; and 3) we demonstrate the effectiveness of our technique using real-world deployments. The experimental results show that our technique can achieve more than 86% accuracy on average and at least 15% better than the baseline Naive Bayes classifier.

Software development has always inherently required multitasking: developers switch between coding, reviewing, testing, designing, and meeting with colleagues. The advent of software ecosystems like GitHub has enabled something new: the ability to easily switch between projects. Developers also have social incentives to contribute to many projects; prolific contributors gain social recognition and (eventually) economic rewards. Multitasking, however, comes at a cognitive cost: frequent context-switches can lead to distraction, sub-standard work, and even greater stress. In this paper, we gather ecosystem-level data on a group of programmers working on a large collection of projects. We develop models and methods for measuring the rate and breadth of a developers' context-switching behavior, and we study how context-switching affects their productivity. We also survey developers to understand the reasons for and perceptions of multitasking. We find that the most common reason for multitasking is interrelationships and dependencies between projects. Notably, we find that the rate of switching and breadth (number of projects) of a developer's work matter. Developers who work on many projects have higher productivity if they focus on few projects per day. Developers that switch projects too much during the course of a day have lower productivity as they work on more projects overall. Despite these findings, developers perceptions of the benefits of multitasking are varied.

In distributed software development synchronized actions are important for completion of complex, interleaved tasks that require the abilities of multiple people. Synchronous development is manifested when file commits by two developers are close together in time and modify the same files. Here we propose quantitative methods for identifying synchronized activities in OSS projects, and use them to relate developer synchronization with effective productivity and communication. In particular, we define co-commit bursts and communication bursts, as intervals of time rich in co-commit and correspondence activities, respectively, and construct from them smoothed time series which can be, subsequently, correlated to discover synchrony. We find that synchronized co-commits between developers are associated with their effective productivity and coordination: during co-commit bursts, vs. at other times, the project size grows faster even though the overall coding effort slows down. We also find strong correlation between synchronized co-commits and communication, that is, for pairs of developers, more co-commit bursts are accompanied with more communication bursts, and their relationship follows closely a linear model. In addition, synchronized co-commits and communication activities occur very close together in time, thus, they can also be thought of as synchronizing each other. This study can help with better understanding collaborative mechanisms in OSS and the role communication plays in distributed software engineering.

Developers in complex, self-organized open-source projects often work on many different files, and over time switch focus between them. Shifting focus can have impact on the software quality and productivity, and is thus an important topic of investigation. In this paper, we study focus shifting patterns (FSPs) of developers by comparing trace data from a dozen open source software (OSS) projects of their longitudinal commit activities and file dependencies from the projects call graphs. Using information theoretic measures of network structure, we find that fairly complex focus-shifting patterns emerge, and FSPs in the same project are more similar to each other. We show that developers tend to shift focus along with, rather than away from, software dependency links described by the call graphs. This tendency becomes weaker as either the interval between successive commits, or the organizational distance between committed files (i.e. directory distance), gets larger. Interestingly, this tendency appears stronger with more productive developers. We hope our study will initiate interest in further understanding of FSPs, which can ultimately help to (1) improve current recommender systems to predict the next focus of developers, and (2) provide insight into better call graph design, so as to facilitate developers' work.

Topological structure is considered more and more important in managing a supply network or predicting its development. In this paper, a new framework is proposed to model the topological structure of supply networks, where different types of supply networks can be created just by introducing different supplier-customer connecting rules. Generally, the networks created in the framework are much different from the random networks with the same degree sequences. The revealed phenomenon suggests that real-world supply networks may benefit from its intrinsic mechanism on flexibility, efficiency, and robustness to target attacks. Note to Practitioners-The topological structure of supply networks is considered more and more important in managing a supply network or predicting its development. In this paper, we introduce a framework to model and analyze the topological structure of supply networks. This work aims to characterize supply networks by statistical methods and can help researchers better understand the material dynamics on supply networks and further conveniently create their own supply networks by summarizing practical supplier-customer connecting rules or analyzing real-world supply network data. The work should be further expanded in other aspects, such as simulating material dynamics on supply networks, designing optimal structure by introducing proper supplier-customer connecting rules, rearranging local connections to enhance the competi tiveness and further ensure the long-term benefit of a target firm, and so on, all of which are of much interest for governors, investors, and managers and can be studied in the present framework in the future.