Publisher Copyright: © 2026 Owner/Author.

State-of-the-art single-agent claim verification methods struggle with complex claims that require nuanced analysis of multifaceted evidence. Inspired by real-world professional fact-checkers, we propose DebateCV, the first debate-driven claim verification framework powered by multiple LLM agents. In DebateCV, two Debaters argue opposing stances to surface subtle errors in single-agent assessments. A decisive Moderator is then required to weigh the evidential strength of conflicting arguments to deliver an accurate verdict. Yet, zero-shot Moderators are biased toward neutral judgments, and no datasets exist for training them. To bridge this gap, we propose Debate-SFT, a post-training framework that leverages synthetic data to enhance agents' ability to effectively adjudicate debates for claim verification. Results show that our methods surpass state-of-the-art non-debate approaches in both accuracy (across various evidence conditions) and justification quality.

The increasing popularity of large models and datasets has highlighted the significance of distributed training networks. As gradient synchronization generates substantial traffic, in-network aggregation (INA) has emerged as a solution to offload aggregation onto the switch, alleviating network congestion and accelerating distributed training. However, the limited memory capacity of the INA switch becomes a potential bottleneck as computation shifts into the network, especially in multi-tenant scenarios. To address this bottleneck and enhance network throughput, we propose the Aggregation with Innetwork Resource Pooling (AIRP) framework. Unlike existing approaches that optimize individual switches in a localized manner, AIRP takes a holistic view and efficiently pools switch memory resources across the entire network, allocating them to multiple tenants. Evaluation using the ns-3 simulator and P4 testbed demonstrates that AIRP can accelerate the training of various models, including computer vision and language models. The experimental results show that AIRP outperforms existing INA approaches by up to 7 times in terms of network throughput in multi-tenant scenarios, while also achieving great flexibility and efficiency in deployment.

Detecting social bots is crucial for mitigating the spread of misinformation and preserving online conversation authenticity. State-of-the-art solutions typically leverage graph neural networks (GNNs) to model user representations from social relationships and metadata. However, these approaches overlook two key factors: the similarity of a user and her neighbors, as well as the coordinated behaviors of social bots, resulting in a suboptimal detection performance. To address these issues, we propose HyperScan, a novel representation learning method for social bot detection. Specifically, we introduce three effective learners to capture pair-wise, hop-wise, and group-wise relations. HyperScan learns pair-wise user representations based on social relations and user features. It then enhances user representations by building hop-wise interactions across the learned pair-wise user representations for capturing the structure-level proximity information. Subsequently, it models user representations by constructing higher-order (group-wise) relations derived from user profiles, tweets, and social relations to capture the feature-level proximity knowledge. By leveraging hop-wise interactions and higher-order relations, HyperScan significantly improves bot detection performance. Our extensive experiments demonstrate that HyperScan outperforms state-of-the-art methods on three benchmark datasets. Additional studies validate the robustness and effectiveness of each component of HyperScan.

Recently, decentralized online social networks (DOSNs) such as Mastodon have emerged quickly, bringing new opportunities for studies in user behavior modeling and multi-modal learning. However, their decentralized architecture presents two key challenges: 1) Distributed data and inconsistent access strategies across several individual instances make a unified collection difficult; 2) user-generated content (UGC) contains multiple modalities while lacking standard organization and high-quality annotation. To address these issues, we constructed FediData, a comprehensive multi-modal dataset from Mastodon. Our dataset integrates user profiles, text, images, and social interactions. To validate FediData's usefulness, we designed and analyzed several tasks and systematically evaluated the performance of existing state-of-the-art methods. Our analysis reveals the unique challenges of DOSNs and highlights the value of FediData in DOSN-related studies. We believe FediData could serve as a foundational dataset for advancing user behavior analytics, multi-modal learning, and future decentralized web research. All data and documentation are available in a Zenodo repository at https://zenodo.org/records/15621243 (DOI: 10.5281/zenodo.15621243).

We release a large-scale dataset that captures interactions between human users and CommentRobert, an LLM-based social media agent on Weibo. The dataset contains Weibo posts in which users actively mention the LLM agent account @CommentRobert, indicating that the users are interested in interacting with the platform-empowered LLM agent. The dataset contains 557,645 interactions from 304,400 unique users over 17 months. We detail our data collection methodology, user attributes, and content characteristics, underscoring the dataset's value in examining real-world human-LLM agent interactions. Our analysis offers insights into the demographic and behavioral traits of users interested in the selected LLM agent, interaction dynamics between humans and the agent, and linguistic patterns in comments. These interactions provide a unique lens through which to explore how humans perceive, trust, and communicate with LLMs. This dataset enables further research into modeling human intent understanding, improving LLM agent design, and studying the evolution of human-LLM agent relationships. Potential applications also include long-term user engagement prediction and AI-generated comment detection on social platforms. This constructed dataset is available at https://zenodo.org/records/16921462.

Spiking Neural Network (SNN) simulation involves emulating the activation and firing of spiking neurons on hardware platforms. This is a highly time-sensitive task, requiring the simulation of billions of neurons and their intercommunication within a few milliseconds. Each neuron performs a complex, interdependent multi-stage communication and computation task. We consider the task placement of SNN on supercomputers to accelerate SNN simulation. Existing task placement methods for SNN simulations have two major limitations. First, they lack the capability to handle large-scale SNNs with billions of neurons. Second, they focus primarily on optimizing communication delay, while neglecting multi-stage computation delays in SNN simulations. In this paper, we formalize the SNN Joint Multi-stage Communication and Computation Placement (SJCCP) problem. We demonstrate that SJCCP can be solved using an approximation algorithm with an approximation ratio of $O\left( {{k^2}\sqrt {\log n\log k} } \right)$, where n is the number of voxels in the SNN and k is the number of GPUs. To further reduce the time complexity of solving SJCCP in practice, we propose a novel efficient framework, FastSJP, tailored for large-scale SNN placement. Then we apply the FastSJP framework to a human brain simulation that runs a large-scale SNN model derived from authentic biological data on a supercomputer equipped with 1024 GPUs. Experimental results verify that our framework notably reduces time overhead, ranging from 17.31% to 28.45%, compared to state-of-the-art methods. Leveraging the computational power of the supercomputer, FastSJP maximizes the problem size and processing performance, significantly advancing the development of brain-inspired intelligence.

With the increasing popularity of distributed training applications, the growth in network traffic has become an impediment to the communication among worker nodes in the system. In-network aggregation (INA) has emerged as a solution to improve communication efficiency by offloading gradient aggregation to switches. However, in multi-tenant scenarios, INA switch memory capacity has been identified as a main bottleneck, leading to reduced network throughput and slower training processes. To address this, we propose Adaptive Gradient Quantization (AGQ) on the switch. AGQ reduces the quantization bit-width of gradients, allowing for storage of more gradients within the limited switch memory while maintaining training accuracy. Compared to quantization on hosts, AGQ can swiftly adapt to the available memory on switches and offers an improved balance between minimizing precision loss and enhancing training throughput. We implement AGQ on a P4 switch testbed, and experimental results demonstrate that enabling AGQ can achieve an up to 100% increase in training throughput without explicit drop of training accuracy compared with existing INA solutions like ATP and host-based quantization methods like THC.

Brain simulation is an inherently highly parallel and time-sensitive task, requiring the simulation of billions of neurons and their interactions within just a few milliseconds. With the growing availability of brain data from biological research, more realistic and detailed simulations are becoming feasible. However, this also poses unprecedented challenges for parallel computing due to the extreme sparsity and heterogeneity of the emerging workloads. Efficient deployment of such workloads on modern HPC systems is critical to overcoming these challenges. We propose BMapper, a deployment framework that enables efficient parallel execution of brain simulations on supercomputers. BMapper comprises three synergistic components: BPartitioning, which introduces a novel multi-dimensional hybrid partitioning strategy to balance workloads across GPUs and reduce inter-GPU spike traffic; BPlacement, which applies deterministic spectral partitioning to minimize inter-server communication; and BRelaying, which identifies lightly loaded GPUs to assist the top-k heavily loaded ones by relaying spike traffic. These components work together to balance loads and minimize communication overhead, enabling high-speed simulation of large-scale brain models. BMapper has been deployed to simulate up to 10 billion neurons on a 1000-GPU supercomputer, achieving 25.15%–47.48% faster execution than state-of-the-art methods.

The rising popularity of large language models (LLMs) has raised concerns about potential abuse and harmful content. As a result, developing a highly generalizable and adaptable machine-generated text (MGT) detection system has become an urgent priority. Given that LLMs are most commonly misused in academic writing, this work investigates the generalization and adaptation capabilities of MGT detectors in three key aspects specific to academic writing: First, we construct MGT-Academic, a large-scale dataset comprising over 336M tokens and 749K samples. MGT-Academic focuses on academic writing, featuring human-written texts (HWTs) and MGTs across STEM, Humanities, and Social Sciences, paired with an extensible code framework for efficient benchmarking. Second, we benchmark the performance of various detectors for binary classification and text attribution tasks in both in-domain and cross-domain settings. This benchmark reveals the often-overlooked challenges of text attribution tasks. Third, we introduce a novel text attribution task in which models must adapt to new classes over time, with little or no access to prior training data, spanning both few-shot and many-shot scenarios. We implement a range of adaptation techniques to enhance performance across these settings. Our findings provide new insights into the generalization ability of MGT detectors and lay the foundation for building robust, adaptive detection systems. The code framework is available at https://github.com/Y-L-LIU/MGTBench-2.0.

Predicting future heart rate (HR) not only helps in detecting abnormal heart rhythms but also provides timely support for downstream health monitoring services. Existing methods for HR prediction encounter challenges, especially concerning privacy protection and data heterogeneity. To address these challenges, this paper proposes a novel HR prediction framework, PCFedH, which leverages personalized federated learning and prototypical contrastive learning to achieve stable clustering results and more accurate predictions. PCFedH contains two core modules: a prototypical contrastive learning-based federated clustering module, which characterizes data heterogeneity and enhances HR representation to facilitate more effective clustering, and a two-phase soft clustered federated learning module, which enables personalized performance improvements for each local model based on stable clustering results. Experimental results on two real-world datasets demonstrate the superiority of our approach over state-of-the-art methods, achieving an average reduction of 3.1% in the mean squared error across both datasets. Additionally, we conduct comprehensive experiments to empirically validate the effectiveness of the key components in the proposed method. Among these, the personalization component is identified as the most crucial aspect of our design, indicating its substantial impact on overall performance.

Influential users in online social networks have been given a lot of attention, which could maximize information diffusion in a centralized dissemination process. However, there exist users that can significantly promote diversified information communication among users, which is a pluralistic interaction process. Structural hole (SH) spanners serve as a bridge between social groups, promoting interactions between users and allowing ordinary users to receive diverse information. Filling structural holes, SH spanners have more opportunities to obtain innovative content. Identifying SH spanners is challenging due to their inconspicuous characteristics and privacy policy restrictions. In this article, we are the first to study the SH spanner identification problem using the behavior data generated by users, releasing the dependency on the entire graph structure to find SH spanners. We propose DeepHole to design both the semantic and the sequential behavior analysis modules on users’ generated content, utilizing the TextCNN and Transformer models, respectively. The semantic and sequential embeddings are effective in distinguishing the SH spanners from ordinary users. We conduct comprehensive evaluations using the Yelp and Foursquare datasets simultaneously. Results show that DeepHole can achieve a satisfying performance in detecting SH spanners labeled by both constraint and effective size metrics, with AUC values of 0.923 and 0.933 for Yelp and 0.828 and 0.815 for Foursquare.

HTTP/3 (H3) has experienced significant growth and extensive adoption in various scenarios, especially in Content Delivery Networks (CDNs). Over the past few years, there have been numerous insightful studies on its deployment in industrial CDNs. However, these studies often separately analyze H3 and CDN, overlooking their synergistic integration. In this work, we explore the applicability of H3 in CDN from a holistic perspective. We analyze 325 websites hosted by seven CDN providers and identify three key characteristics where CDN align perfectly with H3's strengths. Firstly, CDN resources dominate the composition of webpages, where enabling H3 can amplify H3's benefits in connection acceleration. Secondly, CDN providers also exhibit a dominant characteristic, with the majority of CDN resources hosted by a few large providers. This phenomenon makes different webpages share the same provider. When browsing consecutively, H3 helps to skip the connection phase by resuming the connections to the same CDN provider across pages. Thirdly, H3 mitigates the congestion problem on webpages serving multiple CDN resources. This work provides a deeper insight into the applicability of H3 in large-scale distributed systems like CDNs, holding promise for informing the development and optimization of industrial H3.

Model partitioning is a promising solution to reduce the high computation load and transmission of high-volume data. Within the scope of Edge AI, the fundamentals of model partitioning involve splitting the model for local computing at the edge and offloading heavy computation tasks to the cloud or server. This approach benefits scenarios with limited computing and battery capacity with low latency requirements, such as connected autonomous vehicles. However, while model partitioning offers advantages in reducing the onboard computation, memory requirements and inference time, it also introduces challenges such as increased energy consumption for partitioned computations and overhead for transferring partitioned data/model. In this work, we explore hybrid model partitioning to optimize computational and communication energy consumption. Our results provide an initial analysis of the tradeoff between energy and accuracy, focusing on the energy-aware model partitioning for future Edge AI applications.

Continuously orchestrating in-network model aggregations for federated learning faces fundamental challenges such as the combinatorial nature of traffic reduction, the dynamic trade-offs between system overhead and model convergence, and the unpredictable inputs from uncertain system environments. In this work, we model a nonlinear mixed-integer program to optimize the long-term total cost of federated learning computation overhead, traffic reduction, network delay, and programmable switch reconfigurations over time. To attack the lexicographic minimax, submodular, and online nature of this problem, we propose a polynomial-time algorithmic framework to judiciously designate the timing of reconfigurations, while designing and invoking a linearized transformation for selecting routing paths, a greedy sub-algorithm for selecting aggregation locations, and an online learning sub-algorithm for controlling federated learning convergence. We demonstrate our rigorous mathematical insights behind our algorithms, and prove the competitive ratio as the performance guarantee. Using trace-driven evaluations, we have validated our approach's superiority over existing methods.

The future of networking must address the connectivity demands of billions of people and trillions of networked devices. The Intelligent Internet of Everything (IoE) is widely considered the next evolution of the Internet, yet it poses significant challenges to the current TCP/IP architecture. To overcome these challenges, the concept of the Cybertwin network has been proposed as a promising future Internet architecture. To facilitate the advancement of Cybertwin-related research, we have developed CTSim, a Cybertwin network simulator. Our demo demonstrates that the Cybertwin network could effectively enhances mobility, availability, and security in Internet of Things scenarios, making it a powerful tool for researchers in these fields.

IPv6, with its larger address space, is well-suited for IoT scenarios, allowing each device to have a unique address for efficient communication. However, its adoption in sensor networks is hindered by underdeveloped IPv6 infrastructure and compatibility issues with IPv4-based networks. To address these challenges, this paper proposes a dual-stack load balancing approach to enhance seamless and reliable connectivity between IPv6 and IPv4. By adaptively distributing traffic between IPv6 and IPv4 channels, our system alleviates performance bottlenecks and improves reliability. A real-world deployment in the campus network of Fudan University demonstrates a 1.9x increase in throughput and a 27% reduction in congestion-related failures with only CPU overhead of 2.44% and a translation delay of 0.58 ms. These results highlight the potential of intelligent load balancing and IPv4/IPv6 translation in supporting scalable and efficient sensor networks and facilitating the transition to IPv6.

The temporal patterns of code submissions, denoted as work rhythms, provide valuable insight into the work habits and productivity in software development. In this paper, we investigate the work rhythms in software development and their effects on technical performance by analyzing the profiles of developers and projects from 110 international organizations and their commit activities on GitHub. Using clustering, we identify four work rhythms among individual developers and three work rhythms among software projects. Strong correlations are found between work rhythms and work regions, seniority, and collaboration roles. We then define practical measures for technical performance and examine the effects of different work rhythms on them. Our findings suggest that moderate overtime is related to good technical performance, whereas fixed office hours are associated with receiving less attention. Furthermore, we survey 92 developers to understand their experience with working overtime and the reasons behind it. The survey reveals that developers often work longer than required. A positive attitude towards extended working hours is associated with situations that require addressing unexpected issues or when clear incentives are provided. In addition to the insights from our quantitative and qualitative studies, this work sheds light on tangible measures for both software companies and individual developers to improve the recruitment process, project planning, and productivity assessment.

Phone number recycling (PNR) refers to the event wherein a mobile operator collects a disconnected number and reassigns it to a new owner. It has posed a threat to the reliability of the existing authentication solution for e-commerce platforms. Specifically, a new owner of a reassigned number can access the application account with which the number is associated, and may perform fraudulent activities. Existing solutions that employ a reassigned number database from mobile operators are costly for e-commerce platforms with large-scale users. Thus, alternative solutions that depend on only the information of the applications are imperative. In this work, we study the problem of detecting accounts that have been compromised owing to the reassignment of phone numbers. Our analysis on Meituan’s real-world dataset shows that compromised accounts have unique statistical features and temporal patterns. Based on the observations, we propose a novel model called temporal pattern and statistical feature fusion model (TSF) to tackle the problem, which integrates a temporal pattern encoder and a statistical feature encoder to capture behavioral evolutionary interaction and significant operation features. Extensive experiments on the Meituan and IEEE-CIS datasets show that TSF significantly outperforms the baselines, demonstrating its effectiveness in detecting compromised accounts due to reassigned numbers.

Human mobility trajectories are fundamental resources for analyzing mobile behaviors in urban computing applications. However, these trajectories, typically collected from location-based services, often suffer from sparsity and irregularity in time. To support the development of mobile applications, there is a need to recover or estimate missing locations of unobserved time slots in these trajectories at a fine-grained spatial-temporal resolution. Existing methods for trajectory recovery rely on either individual user trajectories or collective mobility patterns from all users. The potential to combine individual and collective patterns for precise trajectory recovery remains unexplored. Additionally, current methods are sensitive to the heterogeneous temporal distributions of the observable trajectory segments. In this paper, we propose CLMove (where CL stands for contrastive learning), a novel model designed to capture multilevel mobility patterns and enhance robustness in trajectory recovery. CLMove features a two-stage location encoder that captures collective and individual mobility patterns. The graph neural network based networks in CLMove explore location transition patterns within a single trajectory and across various user trajectories. We also design a trajectory-level contrastive learning task to improve the robustness of the model. Extensive experimental results on three representative real-world datasets demonstrate that our CLMove model consistently outperforms state-of-the-art methods in terms of trajectory recovery accuracy.

CDN is a crucial Internet infrastructure ensuring quick access to Internet content. With the expansion of CDN scenarios, beyond delay, resource types like bandwidth and CPU are also important for CDN performance. Our measurements highlight the distinct impacts of various resource types on different CDN requests. Unfortunately, mainstream CDN server selection schemes only consider a single resource type and are unable to choose the most suitable servers when faced with diverse resource types. To fill this gap, we propose Polygon, a QUIC-powered CDN server selection system that is aware of multiple resource demands. Being an advanced transport layer protocol, QUIC equips Polygon with customizable transport parameters to enable the seamless handling of resource requirements in requests. Its 0-RTT and connection migration mechanisms are also utilized to minimize delays in connection and forwarding. A set of collaborative measurement probes and dispatchers are designed to support Polygon, being responsible for capturing various resource information and forwarding requests to suitable CDN servers. Real-world evaluations on the Google Cloud Platform and extensive simulations demonstrate Polygon’s ability to enhance QoE and optimize resource utilization. The results show up to a 54.8% reduction in job completion time, and resource utilization improvements of 13% in bandwidth and 7% in CPU.

Online Social Networks (OSNs) play a significant role in people's daily life. Increasing OSN traffic promotes the requirement for building self-operated Content Delivery Networks (CDNs) to deliver OSN media data efficiently and reduce traffic costs. OSN data in CDNs is heavily influenced by social connectivity, such as friendships. To reduce CDN network traffic by better using social connectivity information, we propose SocialCache, a pervasive social-aware caching strategy in self-operated CDNs. SocialCache supports several social connectivity metric options that reflect the importance of the OSN users and the popularity of the media files. Through the improvement of the cache replacement algorithm in the CDN node and the communication design between nodes, SocialCache realizes the optimization of network traffic. Worth mentioning, SocialCache can easily integrate into mainstream CDN architectures while protecting user privacy. We implement SocialCache on Mininet, using real-world network measurements for CDN hierarchy and a hill-climbing algorithm for parameter selection. SocialCache outperforms a range of state-of-the-art baselines on three real-world OSN datasets. On the Twitter dataset, SocialCache reduces the network traffic volume by 6.40% and improves the cache hit ratio by 14.22%.

The prediction of health metrics for drivers has become increasingly crucial due to the potential impact of drivers' health conditions on traffic accidents. Heart attack is one of the primary causes of health-related traffic tragedies. However, drivers' heart rate (HR) is considered highly-private data, which should not be collected by a centralized server for training the prediction model. To this end, we contribute FedHeart, a novel privacy-preserving federated learning (FL) system for HR prediction. We observe distinct HR changes when drivers are in steady-state and changing-state conditions, and thus we utilize FL to train two separate models for these states. To enhance the prediction accuracy, we incorporate contrastive learning to extract HR features. Through experiments on two real-world datasets, we validate the efficiency of the proposed system in accurately predicting HR during driving scenarios.

The application of deep learning techniques in software engineering becomes increasingly popular. One key problem is developing high-quality and easy-to-use source code representations for code-related tasks. The research community has acquired impressive results in recent years. However, due to the deployment difficulties and performance bottlenecks, seldom these approaches are applied to the industry. In this paper, we present xASTNN, an eXtreme Abstract Syntax Tree (AST)-based Neural Network for source code representation, aiming to push this technique to industrial practice. The proposed xASTNN has three advantages. First, xASTNN is completely based on widely-used ASTs and does not require complicated data pre-processing, making it applicable to various programming languages and practical scenarios. Second, three closely-related designs are proposed to guarantee the effectiveness of xASTNN, including statement subtree sequence for code naturalness, gated recursive unit for syntactical information, and gated recurrent unit for sequential information. Third, a dynamic batching algorithm is introduced to significantly reduce the time complexity of xASTNN. Two code comprehension downstream tasks, code classification and code clone detection, are adopted for evaluation. The results demonstrate that our xASTNN can improve the state-of-the-art while being faster than the baselines.

Online developer communities like GitHub allow a massive number of developers to collaborate. However, the openness of the communities makes them vulnerable to different types of malicious attacks, since attackers can easily join these communities and interact with legitimate users. In this work, we propose GitSec, a deep learning-based solution for detecting malicious accounts in online developer communities. GitSec distinguishes malicious accounts from legitimate ones based on the account profiles, dynamic activity characteristics, as well as social interactions. First, GitSec introduces two user activity sequences and applies a parallel neural network design with an attention mechanism to process the sequences. Second, GitSec constructs two graphs to represent the interactions between users according to their repository operations. Especially, graph neural networks and structural hole theory are employed to deal with the two constructed graphs. Third, GitSec makes use of the descriptive features to enhance the detection performance. The final judgement is made by a decision maker implemented by a supervised machine learning-based classifier. Based on the real-world data of GitHub users, our comprehensive evaluations show that GitSec achieves a better performance than state-of-the-art solutions, with an AUC value of 0.916.

Outstanding users (OUs) denote the influential, "core" or "bridge" users in the online community. How to accurately detect and rank them is an important problem for third-party online service providers and researchers. Conventional efforts, ranging from early graph-based algorithms to recent machine learning-based approaches, typically rely on an entire network's information or at least ego networks. However, for privacy-conscious users or newly-registered users, such information is not easily accessible. To address this issue, we present DeepPick, a novel framework that considers both the generalization and specialization in the detection task of OUs. For generalization, we introduce deep neural networks to capture nonlinear features. For specialization, we leverage the traditional well-defined metrics to preserve common features. Extensive experiments based on real-world datasets demonstrate that our approach achieves a high efficacy in terms of detection performance against the state-of-the-art.

Despite continuous efforts to build and update mobile network infrastructure, mobile devices in developing regions continue to be constrained by limited bandwidth. Unfortunately, this coincides with a period of unprecedented growth in the sizes of mobile applications. Thus it is becoming prohibitively expensive for users in developing regions to download and update mobile apps critical to their economic and educational development. Unchecked, these trends can further contribute to a large and growing global digital divide. Our goal is to better understand the source of this rapid growth in mobile app code size, whether it is reflective of new functionality, and identify steps that can be taken to make existing mobile apps more friendly to bandwidth constrained mobile networks. We hypothesize that much of this growth in mobile apps is due to poor resource/code management, and do not reflect proportional increases in functionality. Our hypothesis is partially validated by mini-programs, apps with extremely small footprints gaining popularity in Chinese mobile platforms. Here, we use functionally equivalent pairs of mini-programs and Android apps to identify potential sources of “bloat,” i.e., inefficient uses of code or resources that contribute to large package sizes. We analyze a large sample of popular Android apps and quantify instances of code and resource bloat. We develop techniques for automated code and resource trimming, and successfully validate them on a large set of Android apps. We hope our results will lead to continued efforts to streamline mobile apps, making them easier to access and maintain for users in developing regions.

Privacy concerns have long been a nuisance for users when using online social networks (OSNs), and most OSNs require users to configure the privacy settings themselves. Researchers are paying attention to the recommendation of privacy settings for OSN users, as some studies have found it difficult for users to understand and to use. This paper addresses the problem of recommending user-level privacy settings by leveraging the user-generated content (UGC) sequences and account information. We first analyze the differences of users with different privacy settings using the crawled Twitter dataset. Further, we formalize the privacy settings recommendation problem as a multi-label classification task and propose a deep learning-based privacy settings recommendation approach, PrivacyRec. It adopts Recurrent Neural Networks (RNNs) for the UGC sequences modeling to extract deep representations of users' semantic preferences and interaction behavior. In addition, it fuses the above representations with processed account portrait feature through attention mechanism and infers the privacy settings. We evaluate PrivacyRec using two real-world datasets and compare it with several baseline methods. The experimental results demonstrate the effectiveness of PrivacyRec's ability to recommend privacy settings.

Online-to-Offline (O2O) e-commerce service platforms and their users are faced with various fraud risks. Among them, financial identity theft is a widely existing challenge. However, existing methods are insufficient to detect this type of fraud. In this paper, we address the financial identity theft detection problem in e-commerce services by leveraging access environment and behavior sequence. To explore the fraud patterns, we first make a detailed analysis using real cases of identity theft from Meituan, a leading O2O e-commerce platform in China. Our findings are twofold. First, fraudulent accounts sharing the same personal ID would have different access environments, such as devices and IP addresses. Second, a group of fraudulent accounts may have aggregations of devices, IP addresses, and delivery addresses. Based on these observations, we propose a hybrid method termed EnvIT to detect financial identity theft based on the heterogeneous graph and the behavior sequence. EnvIT is able to characterize the access environment and the historical behavior of the accounts. Furthermore, an attentive module is adopted to assign weights to different features automatically. We further evaluate EnvIT via extensive experiments using a real-world dataset from Meituan. Our experimental results demonstrate that EnvIt outperforms several baseline methods in fraudulent account detection and achieves an AUC of 0.9210.

Programmable switches have been proposed in today’s network to enable flexible reconfiguration of devices and reduce time-to-deployment. Buffer sizing, an important factor for network performance, however, has not received enough attention in programmable network. The state-of-the-art buffer sizing solutions usually employ either fixed buffer size or adjust the buffer size heuristically. Without programmability, they suffer from either massive packet drops or large queueing delay in dynamic environment. In this paper, we propose Adaptive Buffer Sizing (ABS), a low-cost and deploy-friendly framework compatible with programmable network. By decoupling the data plane and control plane, ABS-capable switches only need to react to the actions from controller, optimizing network performance in run-time under dynamic traffic. Meanwhile, actions can be programmed by particular Machine Learning (ML) models in the controller to meet different network requirements. In this paper, we address two specific ML models for different scenarios, a reinforcement learning model for relatively stable network with user specific quality requirements, and a supervised learning model for highly dynamic network condition. We implement the ABS framework by integrating the prevalent network simulator NS-2 with ML module. The experiment shows that ABS outperforms state-of-the-art buffer sizing solutions by up to 38.23x under various network environments.

Although IPv6 has been proposed to solve the IP address exhaustion problem for decades, the transition process from IPv4 to IPv6 is rather slow due to the possible loss of users and increased costs for ISPs compared with the potential profits. In order to accelerate this process and make full use of the IPv6 network, in this paper, we propose a user-transparent solution named NetBoost by transferring IPv4 traffic through the IPv6 core network. We also implement a simulator called NetBoostSim to further verify the usefulness and prospective performance gain of NetBoost in different network environments. By deploying our system upon both real and simulated network environments, we showcase that better performance for IPv4 end-to-end connections can be acquired by utilizing the light-loaded IPv6 network to transfer traffic from heavy-loaded IPv4 core network, using stateless IPv4/IPv6 translation techniques. In this way, our system can serve as an incentive for ISPs to upgrade to pure IPv6 networks gradually without concerns for the user churn.

Social networks now connect billions of people around the world, where individuals occupying different positions often represent different social roles and show different characteristics in their behaviors. The structural hole (SH) theory demonstrates that users occupying the bridging positions between different communities have advantages since they control the key information diffusion paths. Users of this type, known as SH spanners, are important when it comes to assimilating social network structures and user behaviors. In this article, we review the use of SHs theory in social network analysis, where SH spanners take advantage of both information and control benefits. We investigate the existing algorithms of SH spanner detection and classify them into information flow-based algorithms and network centrality-based algorithms. For practitioners, we further illustrate the applications of SH theory in various practical scenarios, including enterprise settings, information diffusion in social networks, software development, mobile applications, and machine learning (ML)-based social prediction. Our review provides a comprehensive discussion on the foundation, detection, and practical applications of SHs. The insights can facilitate researchers and service providers to better apply the theory and derive value-added tools with advanced ML techniques. To inspire follow-up research, we identify potential research trends in this area, especially on the dynamics of networks.

Today, Internet service deployment is typically implemented with server replication at multiple locations. Domain name system (DNS), which translates human-readable domain names into network-routable IP addresses, is typically used for distributing users to different server replicas. However, DNS relies on several network-based queries and the queries delay the connection setup process between the client and the server replica. In this article, we propose Artemis, a practical low-latency naming and routing system that supports optimal server (replica) selection based on user-defined policies and provides lower query latencies than DNS. Artemis uses a DNS-like domain name-IP mapping for replica selection and achieves low query latency by combining the name resolution process with the transport layer handshake process. In Artemis, all server replicas at different locations share the same anycast IP address, called Service Address. Clients use the Service Address to establish a transport layer connection with the server. The client's initial handshake packet is routed over an overlay network to reach the optimal server. Then the server migrates the transport layer connection to its original unicast IP address after finishing the handshake process. After that, service discovery is completed, and the client communicates with the server directly via IP addresses. To validate the effectiveness of Artemis, we evaluate its performance via both real trace-driven simulation and real-world deployment. The result shows that Artemis can handle a large number of connections and reduce the connection setup latency compared with state-of-the-art solutions. More specifically, our deployment across 11 Google data centers shows that Artemis reduces the connection setup latency by 39.4% compared with DNS.

As one of the biggest online developer communities, GitHub supports interactions between millions of developers around the world. This paper focuses on detecting and analyzing the interactions on GitHub from several perspectives. First, from a global viewpoint, we build an interaction graph based on GitHub Events to investigate the general structure and attributes of the GitHub interaction network. Second, from the perspective of important users on GitHub, we pay particular attention to those who bridge social circles by spanning across communities. Concretely, we apply the structural hole theory to identify and explore the structural hole spanners in this interaction network. Last but not least, we compare structural hole spanners with ordinary users in several aspects and excavate distinctions between them. To our best knowledge, this is the first study that applies structural hole theory to the interaction graph of a leading social network. It not only provides a unique standpoint on user interactions, but also gives a comprehensive inspection of structural hole spanners.

Dating apps have gained tremendous popularity during the past decade. Compared with traditional offline dating means, dating apps ease the process of partner finding significantly. While bringing convenience to hundreds of millions of users, dating apps are vulnerable to become targets of adversaries. In this article, we focus on malicious user detection in dating apps. Existing methods overlooked the signals hidden in the textual information of user interactions, particularly the interplay of temporal-spatial behaviors and textual information, leading to limited detection performance. To tackle this, we propose DatingSec, a novel malicious user detection system for dating apps. Concretely, DatingSec leverages long short-term memory neural networks (LSTM) and an attentive module to capture the interplay of users' temporal-spatial behaviors and user-generated textual content. We evaluate DatingSec on a real-world dataset collected from Momo, a widely used dating app with more than 180 million users. Experimental results show that DatingSec outperforms state-of-the-art methods and achieves an F1-score of 0.857 and AUC of 0.940.

Online social networks (OSNs) have become a commodity in our daily life. As an important concept in sociology and viral marketing, the study of social influence has received a lot of attentions in academia. Most of the existing proposals work well on dominant OSNs, such as Twitter, since these sites are mature and many users have generated a large amount of data for the calculation of social influence. Unfortunately, cold-start users on emerging OSNs generate much less activity data, which makes it challenging to identify potential influential users among them. In this work, we propose a practical solution to predict whether a cold-start user will become an influential user on an emerging OSN, by opportunistically leveraging the user’s information on dominant OSNs. A supervised machine learning-based approach is adopted, transferring the knowledge of both the descriptive information and dynamic activities on dominant OSNs. Descriptive features are extracted from the public data on a user’s homepage. In particular, to extract useful information from the fine-grained dynamic activities that cannot be represented by the statistical indices, we use deep learning technologies to deal with the sequential activity data. Using the real data of millions of users collected from Twitter (a dominant OSN) and Medium (an emerging OSN), we evaluate the performance of our proposed framework to predict prospective influential users. Our system achieves a high prediction performance based on different social influence definitions.

Accurate traffic speed prediction is an important and challenging topic for transportation planning. Previous studies on traffic speed prediction predominately used spatio-temporal and context features for prediction. However, they have not made good use of the impact of traffic incidents. In this work, we aim to make use of the information of incidents to achieve a better prediction of traffic speed. Our incident-driven prediction framework consists of three processes. First, we propose a critical incident discovery method to discover traffic incidents with high impact on traffic speed. Second, we design a binary classifier, which uses deep learning methods to extract the latent incident impact features. Combining above methods, we propose a Deep Incident-Aware Graph Convolutional Network (DIGC-Net) to effectively incorporate traffic incident, spatio-temporal, periodic and context features for traffic speed prediction. We conduct experiments using two real-world traffic datasets of San Francisco and New York City. The results demonstrate the superior performance of our model compared with the competing benchmarks.

Wearable devices collect user information about their activities and provide insights to improve their daily lifestyles. Smart health applications have achieved great success by training Machine Learning (ML) models on a large quantity of user data from wearables. However, user privacy and scalability are becoming critical challenges for training ML models in a centralized way. Federated learning (FL) is a novel ML paradigm with the goal of training high quality models while distributing training data over a large number of devices. In this demo, we present FL4W, a FL system with wearable devices enabling training a human activity recognition classifier. We also perform preliminary analytics to investigate the model performance with increasing computation of clients.

Being a leading online service providing both local search and social networking functions, Foursquare has attracted tens of millions of users all over the world. Understanding the user behavior of Foursquare is helpful to gain insights for location-based social networks (LBSNs). Most of the existing studies focus on a biased subset of users, which cannot give a representative view of the global user base. Meanwhile, although the user-generated content (UGC) is very important to reflect user behavior, most of the existing UGC studies of Foursquare are based on the check-ins. There is a lack of a thorough study on tips, the primary type of UGC on Foursquare. In this article, by crawling and analyzing the global social graph and all published tips, we conduct the first comprehensive user behavior study of all 60+ million Foursquare users around the world. We have made the following three main contributions. First, we have found several unique and undiscovered features of the Foursquare social graph on a global scale, including a moderate level of reciprocity, a small average clustering coefficient, a giant strongly connected component, and a significant community structure. Besides the singletons, most of the Foursquare users are weakly connected with each other. Second, we undertake a thorough investigation according to all published tips on Foursquare. We start from counting the numbers of tips published by different users and then look into the tip contents from the perspectives of tip venues, temporal patterns, and sentiment. Our results provide an informative picture of the tip publishing patterns of Foursquare users. Last but not least, as a practical scenario to help third-party application providers, we propose a supervised machine learning-based approach to predict whether a user is an influential by referring to the profile and UGC, instead of relying on the social connectivity information. Our data-driven evaluation demonstrates that our approach can reach a good prediction performance with an F1-score of 0.87 and an AUC value of 0.88. Our findings provide a systematic view of the behavior of Foursquare users and are constructive for different relevant entities, including LBSN service providers, Internet service providers, and third-party application providers.

Successfully detecting, analyzing, and reasoning about collective anomalies is important for many real-life application domains (e.g., intrusion detection, fraud analysis, software security). The primary challenges to achieving this goal include the overwhelming number of low-risk events and their multimodal relationships, the diversity of collective anomalies by various data and anomaly types, and the difficulty in incorporating the domain knowledge of experts. In this paper, we propose the novel concept of the faceted High-Order Correlation Graph (HOCG). Compared with previous, low-order correlation graphs, HOCG achieves better user interactivity, computational scalability, and domain generality through synthesizing heterogeneous types of objects, their anomalies, and the multimodal relationships, all in a single graph. We design elaborate visual metaphors, interaction models, and the coordinated multiple view based interface to allow users to fully unleash the visual analytics power of the HOCG. We conduct case studies for three application domains and collect feedback from domain experts who apply our method to these scenarios. The results demonstrate the effectiveness of the HOCG in the overview of point anomalies, the detection of collective anomalies, and the reasoning process of root cause analyses.

Online developer communities like GitHub provide services such as distributed version control and task management, which allow a massive number of developers to collaborate online. However, the openness of the communities makes themselves vulnerable to different types of malicious attacks, since the attackers can easily join and interact with legitimate users. In this work, we formulate the malicious account detection problem in online developer communities, and propose GitSec, a deep learning-based solution to detect malicious accounts. GitSec distinguishes malicious accounts from legitimate ones based on the account profiles as well as dynamic activity characteristics. On one hand, GitSec makes use of users' descriptive features from the profiles. On the other hand, GitSec processes users' dynamic behavioral data by constructing two user activity sequences and applying a parallel neural network design to deal with each of them, respectively. An attention mechanism is used to integrate the information generated by the parallel neural networks. The final judgement is made by a decision maker implemented by a supervised machine learning-based classifier. Based on the real-world data of GitHub users, our extensive evaluations show that GitSec is an accurate detection system, with an F1-score of 0.922 and an AUC value of 0.940.

Today, Internet service deployment is typically implemented with server replication at multiple locations for the purpose of load balancing, failure tolerance, and user experience optimization. Domain name system (DNS) is responsible for translating human-readable domain names into network-routable IP addresses. When multiple replicas exist, upon the arrival of a query, DNS selects one replica and responds with its IP address. Thus, the delay caused by the process of DNS query including the selection of replica is part of the connection setup latency.

Rumor spreading in online social networks can inflict serious damages on individual, organizational, and societal levels. This problem has been addressed via computational approach in recent years. The dominant computational technique for the identification of rumors is the binary classification that uses rumor and non-rumor for the training. In this method, the way of annotating training data points determines how each class is defined for the classifier. Unlike rumor samples that often are annotated similarly, non-rumors get their labels arbitrarily based on annotators' volition. Such an approach leads to unreliable classifiers that cannot distinguish rumor from non-rumor consistently. In this paper, we tackle this problem via a novel classification approach called one-class classification (OCC). In this approach, the classifier is trained with only rumors, which means that we do not need the non-rumor data points at all. For this study, we use two primary Twitter data sets in this field and extract 86 features from each tweet. We then apply seven one-class classifiers from three different paradigms and compare their performance. Our results show that this approach can recognize rumors with a high level of F1-score. This approach may influence the predominant mentality of scholars about computational rumor detection and puts forward a new research path toward dealing with this problem.

Various dedicated web services in the cyberspace, e.g., social networks, e-commerce, and instant communications, play a significant role in people's daily-life. Billions of people around the world access these services through multiple online identifiers (IDs), and interact with each other in both the cyberspace and the physical world. Thanks to the rapid development of wireless and mobile technologies, nowadays these two kinds of interactions are highly relevant with each other. In order to link between the cyberspace and the physical world, we propose a new type of social network, i.e., co-location social network (CLSN). A CLSN contains online IDs describing people's online presence and offline “encountering” events when people come across each other. By analyzing real data collected from a mainstream ISP in China, which contains 32.7 million IDs across the most popular web services, we build a large-scale CLSN, and explore its unique properties from various aspects. The results indicate that the CLSN is quite different from existing online and offline social networks in terms of classic graph metrics. Moreover, we propose a community-based user identification algorithm to find all online IDs belonging to the same physical user. Using some ground-truth data, we demonstrate that our proposed algorithm achieves a high accuracy in user identification. Finally, we perform a user-centric analysis, and we demonstrate the behavioral difference among different types of users.

As the largest microblogging service in China, Sina Weibo, is associated with a huge amount of traffic to and from other major websites. Thanks to a massive set of anonymized data from a major operator covering a large-scale metropolitan area, we extract the click-based transitions between Weibo and other websites. Such transitions have not been looked into in most studies that rely on the visible hyper-links among websites. Based on this unique dataset, we show that 38% of the outgoing transitions (leaving Weibo) go to video clip services. Comparatively, links related to news and long videos are clicked less but posted more. This indicates that photograph and text-sharing in microblog have been gradually replaced by video clip sharing. Also, ignoring video clip sharing, 53% of the transitions go to five of the largest Chinese websites, showing how closely connected Weibo has become in the Chinese Web ecosystem. We further look at users' click behavior in transitions, and find that through co-clustering there do exist a number of prevalent and distinct behavior patterns, suggesting that only seven co-clusters can capture the click behavior of 192 000 studied users. Overall, our work is the first to unveil the significant role of microblogging platforms in the Web ecosystem. Considering the massive user base of Weibo, our findings can aid the design of more personalized link recommendation in Weibo, thus generating more referral traffic, i.e., user visits, from Weibo to other major websites.

In a Website Fingerprinting (WF) attack, a local, passive eavesdropper utilizes network flow information to identify which web pages a user is browsing. Previous researchers have extensively demonstrated the feasibility and effectiveness of WF, but only under the strong Single Page Assumption: the network flow extracted by the adversary always belongs to a single page. In other words, the WF classifier will never be asked to classify a network flow corresponding to more than one page, or part of a page. The Single Page Assumption is unrealistic because people often browse with multiple tabs. When this happens, the network flow induced by multiple tabs will overlap, and current WF attacks fail to classify correctly.

Detecting, analyzing and reasoning collective anomalies is important for many real-life application domains such as facility monitoring, software analysis and security. The main challenges include the overwhelming number of low-risk events and their multifaceted relationships which form the collective anomaly, the diversity in various data and anomaly types, and the difficulty to incorporate domain knowledge in the anomaly analysis process. In this paper, we propose a novel concept of high-order correlation graph (HOCG). Compared with the previous correlation graph definition, HOCG achieves better user interactivity, computational scalability, and domain generality through synthesizing heterogeneous types of nodes, attributes, and multifaceted relationships in a single graph. We design elaborate visual metaphors, interaction models, and the coordinated multiple view based interface to allow users to fully unleash the visual analytics power over HOCG. We conduct case studies in two real-life application domains, i.e., facility monitoring and software analysis. The results demonstrate the effectiveness of HOCG in the overview of point anomalies, detection of collective anomalies, and reasoning process of root cause analysis.

Due to the rapid development of mobile devices and location-based services, location-based social networks (LBSNs) have become very popular in our daily-life. Malicious account detection is very helpful for different kinds of practical applications. In this paper, we explore the malicious account detection problem by introducing a deep learning-based framework. By using the long short-term memory (LSTM) neural network, we are able to build a classifier to achieve the binary classification. By using the real data collected from Momo, a widely used LBSN which has more than 180 million users around the world, we evaluate our framework and the result shows great promise for malicious account detection tasks.

Visual crowdsourcing (VCS) offers an inexpensive method to collect visual data for implementing tasks, such as 3D mapping and place detection, thanks to the prevalence of smartphone cameras. However, without proper guidance, participants may not always collect data from desired locations with a required Quality-of-Information (QoI). This often causes either a lack of data in certain areas, or extra overheads for processing unnecessary redundancy. In this work, we propose SnapTask, a participatory VCS system that aims at creating complete indoor maps by guiding participants to efficiently collect visual data of high QoI. It applies Structure-from-Motion (SfM) techniques to reconstruct 3D models of indoor environments, which are then converted into indoor maps. To increase coverage with minimal redundancy, SnapTask determines locations for the next data collection tasks by analyzing the coverage of the generated 3D model and the camera views of the collected images. In addition, it overcomes the limitations of SfM techniques by utilizing crowdsourced annotations to reconstruct featureless surfaces (e.g. glass walls) in the 3D model. According to a field test in a library, the indoor map generated by SnapTask successfully reconstructs 100% of the library walls and 98.12% of objects and traversal areas within the library. With the same amount of input data our design of guided data collection increases the map coverage by 20.72% and 34.45%, respectively, compared with unguided participatory and opportunistic VCS.

Recently, Edge Computing has emerged as a new computing paradigm dedicated for mobile applications for performance enhancement and energy efficiency purposes. Specifically, it benefits today's interactive applications on power-constrained devices by offloading compute-intensive tasks to the edge nodes which is in close proximity. Meanwhile, Field Programmable Gate Array (FPGA) is well known for its excellence in accelerating compute-intensive tasks such as deep learning algorithms in a high performance and energy efficiency manner due to its hardware-customizable nature. In this paper, we make the first attempt to leverage and combine the advantages of these two, and proposed a new network-assisted computing model, namely FPGA-based edge computing. As a case study, we choose three computer vision (CV)-based interactive mobile applications, and implement their backend computation parts on FPGA. By deploying such application-customized accelerator modules for computation offloading at the network edge, we experimentally demonstrate that this approach can effectively reduce response time for the applications and energy consumption for the entire system in comparison with traditional CPU-based edge/cloud offloading approach.

Today's cities are filled with sensors to get real-time data on all aspects of the city. Each of these data sources reflects one aspect of the city and provides support for city computing. However, these data are owned by different organizations or companies and stored in different locations of the Internet. It takes lots of efforts for researchers to obtained all of these data. At the same time, a single data set can only represent one side of a city's activities, but combining multiple data allows us to stand in a better perspective overlooking the city we study. We designed a city data aggregation platform to aggregate several data sources about city computing to get a more comprehensive understanding of the city, to discover pattern that are not reflected in a single data set, and facilitate researchers to access data from multiple sources.

Nowadays, people spend more and more time with their smart-phones. They want to make good use of smartphones to achieve an efficient time management. However, many people fail to control the time spent on smartphones effectively and end up with a low task completion rate. We design a time management APP targeted to analyze the time management task completion. In this way, a clear picture of the reasons for a low task completion rate is drawn by us, which could be very useful to help people better manage their time and complete their tasks.

As a result of the blooming of online social networks (OSNs), a user often holds accounts on multiple sites. In this article, we study the emerging “cross-site linking” function available on mainstream OSN services including Foursquare, Quora, and Pinterest. We first conduct a data-driven analysis on crawled profiles and social connections of all 61.39 million Foursquare users to obtain a thorough understanding of this function. Our analysis has shown that the cross-site linking function is adopted by 57.10% of all Foursquare users, and the users who have enabled this function are more active than others. We also find that the enablement of cross-site linking might lead to privacy risks. Based on cross-site links between Foursquare and external OSN sites, we formulate cross-site information aggregation as a problem that uses cross-site links to stitch together site-local information fields for OSN users. Using large datasets collected from Foursquare, Facebook, and Twitter, we demonstrate the usefulness and the challenges of cross-site information aggregation. In addition to the measurements, we carry out a survey collecting detailed user feedback on cross-site linking. This survey studies why people choose to or not to enable cross-site linking, as well as the motivation and concerns of enabling this function.

Network Coordinate (NC) systems provide a scalable means for Internet distance prediction and are useful for various Internet-based services, such as cloud or web-based services. Decentralized, matrix factorization-based NC (MFNC) systems have received particular attention recently. They can serve large-scale distributed services (as opposed to centralized NC systems) and do not need to satisfy the triangle inequality (as opposed to Euclidean-based NC systems). However, because of their decentralized nature, MFNC systems are vulnerable to various malicious attacks. In this paper, we provide the first study on attacks toward MFNC systems, and propose a trust and reputation-based approach calledNCShieldto counter such attacks. It is fully decentralized and can easily be customized. Different from previous approaches, NCShield is able to distinguish between legitimate distance variations and malicious distance alterations. Using four representative data sets from the Internet, we show that NCShield can defend against not only the typical disorder, repulsion and isolation attacks, but also more advanced attacks such as frog-boiling attacks. For example, when selecting node pairs with a shorter distance than a predefined threshold in an online game scenario, even if 30 percent of nodes are malicious, NCShield can reduce the false positive rate from 45.5 to 3.7 percent.

Various dedicated web services in the cyberspace, e.g., social networks, e-commerce, and instant communications, play a significant role in people's daily-life. Billions of people around the world access them through multiple online identifiers (IDs), and interact with each other in both the cyberspace and the physical world. These two kinds of interactions are highly relevant to each other. In order to link between the cyberspace and the physical world, we propose a new type of social network, i.e., co-location social network (CLSN). A CLSN contains online IDs describing people's online presence and offline interactions when people come across each other. By analyzing real data collected from a mainstream ISP in China, which contains 32.7 million IDs across most popular web services, we build a large-scale CLSN, and evaluate its unique properties. The results verify that the CLSN is quite different from existing online and offline social networks in terms of different classic graph metrics. This paper is the first research to study CLSN at scale and paves the way for future studies of this new type of social network.

In the past decade, Location-Based Social Networks (LBSNs) have attracted attentions from both the academia and industry. All LBSN sites are venue-centric, i.e., each review or check-in must be associated with a venue. Despite the importance of the venues, there still lacks a systematical study on LBSN sites from the venues' perspective. To fill this gap, we conduct a comprehensive study on Dianping, the largest online review site in China, with data of more than 506K venues. We first characterize the demographics of each venue. We then measure the venue popularity through real-time reviews. Finally, we propose the concept of “venue network” to study the linkage among the venues. Our paper not only provides a clear picture of venues on Dianping, but also sheds light on potential applications in LBSNs.

Summarizing large influence graphs is crucial for many graph visualization and mining tasks. Classical graph clustering and compression algorithms focus on summarizing the nodes by their structural-level or attribute-level similarities, but usually are not designed to characterize the flow-level pattern which is the centerpiece of influence graphs. On the other hand, the social influence analysis has been intensively studied, but little is done on the summarization problem without an explicit focus on social networks. Building on the recent study of the Influence Graph Summarization (IGS), this paper presents a new perspective of the underlying flow-based heuristic. It establishes a direct linkage between the optimal summarization and the classic eigenvector centrality of the graph nodes. Such a theoretic linkage has important implications on numerous aspects in the pursuit of a perfect influence graph summarization. In particular, it enables us to develop a suite of algorithms that can: 1) achieve a near-optimal IGS objective, 2) support dynamic summarizations balancing the IGS objective and the stability of transition in navigating the summarization, and 3) scale to million-node graphs with a near-linear computational complexity. Both quantitative experiments on real-world citation networks and the user studies on the task analysis experience demonstrate the effectiveness of the proposed summarization algorithms.

The Internet was designed with the end-to-end principle where the network layer provided merely the best-effort forwarding service. This design makes it challenging to add new services into the Internet infrastructure. However, as the Internet connectivity becomes a commodity, users and applications increasingly demand new in-network services. This paper proposes PacketCloud, a cloudlet-based open platform to host in-network services. Different from standalone, specialized middleboxes, cloudlets can efficiently share a set of commodity servers among different services, and serve the network traffic in an elastic way. PacketCloud can help both Internet Service Providers (ISPs) and emerging application/content providers deploy their services at strategic network locations. We have implemented a proof-of-concept prototype of PacketCloud. PacketCloud introduces a small additional delay, and can scale well to handle high-throughput data traffic. We have evaluated PacketCloud in both a fully functional emulated environment, and the real Internet.

Traffic from mobile wireless networks has been growing at a fast pace in recent years and is expected to surpass wired traffic very soon. Service providers face significant challenges at such scales including providing seamless mobility, efficient data delivery, security, and provisioning capacity at the wireless edge. In the Mobility First project, we have been exploring clean slate enhancements to the network protocols that can inherently provide support for at-scale mobility and trustworthiness in the Internet. An extensible data plane using pluggable compute-layer services is a key component of this architecture. We believe these extensions can be used to implement in-network services to enhance mobile end-user experience by either off-loading work and/or traffic from mobile devices, or by enabling en-route service-adaptation through context-awareness (e.g., Knowing contemporary access bandwidth). In this work we present details of the architectural support for in-network services within Mobility First, and propose protocol and service-API extensions to flexibly address these pluggable services from end-points. As a demonstrative example, we implement an in network service that does rate adaptation when delivering video streams to mobile devices that experience variable connection quality. We present details of our deployment and evaluation of the non-IP protocols along with compute-layer extensions on the GENI test bed, where we used a set of programmable nodes across 7 distributed sites to configure a Mobility First network with hosts, routers, and in-network compute services.

Microblogging service has emerged to be a dominant web medium for billions of individuals sharing and spreading instant news and information, therefore monitoring the event evolution on microblog sphere is crucial for providing both better user experience and deeper understanding on real-time events. In this paper we explore the problem of generating storylines from microblogs for user input queries. This problem is challenging due to the sparse, dynamic and social nature of microblogs. Given a query of an ongoing event, we propose to sketch the real-time storyline of the event by a two-level solution. We first propose a language model with dynamic pseudo relevance feedback to obtain relevant tweets, and then generate storylines via graph optimization. Comprehensive experiments on Twitter data sets demonstrate the effectiveness of the proposed methods in each level and the overall framework.

In order to minimize user perceived latency while ensuring high data availability, cloud applications desire to select servers from one of the multiple data centers (i.e., server clusters) in different geographical locations, which are able to provide desired services with low latency and low cost. This paper presents CloudGPS, a new server selection scheme of the cloud computing environment that achieves high scalability and ISP-friendliness. CloudGPS proposes a configurable global performance function that allows Internet service providers (ISPs) and cloud service providers (CSPs) to leverage the cost in terms of inter-domain transit traffic and the quality of service in terms of network latency. CloudGPS bounds the overall burden to be linear with the number of end users. Moreover, compared with traditional approaches, CloudGPS significantly reduces network distance measurement cost (i.e., from O(N) to O(1) for each end user in an application using N data centers). Furthermore, CloudGPS achieves ISP-friendliness by significantly decreasing inter-domain transit traffic.

While network coordinate (NC) systems provide scalable Internet distance estimation service and are useful for various Internet applications, decentralized, matrix factorization-based NC (MFNC) systems have received particular attention recently. They can serve large-scale distributed applications (as opposed to centralized NC systems) and do not need to assume triangle inequality (as opposed to Euclidean-based NC systems). However, because of their decentralized nature, MFNC systems are vulnerable to various malicious attacks. In this paper, we provide the first study on attacks toward MFNC systems, and propose a decentralized trust and reputation approach, called NCShield, to counter such attacks. Different from previous approaches, our approach is able to distinguish between legitimate distance variations and malicious distance alterations. Using four representative data sets from the Internet, we show that NCShield can defend against attacks with high accuracy. For example, when selecting node pairs with a shorter distance than a predefined threshold in an online game scenario, even if 30% nodes are malicious, NCShield can reduce the false positive rate from 45.5% to 3.7%.

Network Coordinate (NC) systems provide an efficient and scalable mechanism to estimate latencies among hosts. However, many popular algorithms like Vivaldi suffer greatly from the existence of Triangle Inequality Violations (TIVs). Two-layer systems like Pharos and hierarchical Vivaldi have been proposed to remedy the impact of TIVs. They divide the whole space into several location-based clusters and run NC systems on both global layer and local layer. However, the two-layer model is only able to optimize the intra-cluster links relating to a limited portion of TIV triangles. In this paper, we propose a new NC system, Tarantula, which divides the space in a novel way. By categorizing the TIVs into three classes, we show that Tarantula handles a much larger portion of existing TIVs than two-layer systems. Moreover, we present two techniques to further strengthen the Tarantula system: 1) relate the updating step size in the Vivaldi algorithm used in Tarantula to ground-truth latency so as to improve the prediction for short links; 2) propose Dynamic Cluster Optimization to dynamically adjust clustering of hosts. Our experimental results show that Tarantula outperforms Pharos and Vivaldi significantly in terms of estimation accuracy. When implementing different NC systems in the application of server selection and detour finding, Tarantula again performs the best.

Computing the shortest-path distances between nodes is a key problem in analyzing social graphs. Traditional methods like breadth-first search (BFS) do not scale well with graph size. Recently, a Graph Coordinate System, called Orion, has been proposed to estimate shortest-path distances in a scalable way. Orion uses a landmark-based approach, which does not take account of the shortest-path distances between non-landmark nodes in coordinate calculation. Such biased input for the coordinate system cannot characterize the graph structure well. In this paper, we propose Pomelo, which calculates the graph coordinates in a decentralized manner. Every node in Pomelo computes its shortest-path distances to both nearby neighbors and some random distant neighbors. By introducing the novel partial BFS, the computational overhead of Pomelo is tunable. Our experimental results from different representative social graphs show that Pomelo greatly outperforms Orion in estimation accuracy while maintaining the same computational overhead.

Network coordinate (NC) systems provide a lightweight and scalable way for predicting the distances, i.e., round-trip latencies among Internet hosts. Most existing NC systems embed hosts into a low dimensional Euclidean space. Unfortunately, the persistent occurrence of Triangle Inequality Violation (TIV) on the Internet largely limits the distance prediction accuracy of those NC systems. Some alternative systems aim at handling the persistent TIV, however, they only achieve comparable prediction accuracy with Euclidean distance based NC systems. In this paper, we propose an NC system, so-called Phoenix, which is based on the matrix factorization model. Phoenix introduces a weight to each reference NC and trusts the NCs with higher weight values more than the others. The weight-based mechanism can substantially reduce the impact of the error propagation. Using the representative aggregate data sets and the newly measured dynamic data set collected from the Internet, our simulations show that Phoenix achieves significantly higher prediction accuracy than other NC systems. We also show that Phoenix quickly converges to steady state, performs well under host churn, handles the drift of the NCs successfully by using regularization, and is robust against measurement anomalies. Phoenix achieves a scalable yet accurate end-to-end distances monitoring. In addition, we study how well an NC system can characterize the TIV property on the Internet by introducing two new quantitative metrics, so-called RERPLand AERPL. We show that Phoenix is able to characterize TIV better than other existing NC systems.

Replicated distributed file systems are designed to store large file reliably across lots of machines, and it arouse the problem of selecting the nearest replica for clients. In this paper, we propose Rigel, a Network Coordinates (NC) based nearest replica selection service, which is an effective infrastructure to select the nearest replica for client in a scalable and lightweight way. Our simulation results have demonstrated that Rigel can at least reduce the read latency between clients and replicas by 20% when compared to the replica selection strategy in Hadoop Distributed File System.

Network Coordinate (NC) systems provide a lightweight and useful way for scalable Internet distance prediction while serving as an important component in many Peer-to-Peer applications. Most of the existing NC systems utilize the Euclidean distance model, which is largely impaired by the persistent occurrence of Triangle Inequality Violation (TIV) in the Internet. Recently, matrix factorization (MF) based NC systems, which can completely remove the TIV constraint, provide an alternative approach towards better prediction accuracy. Phoenix, an NC system based on the MF model, well explores the advantage of the MF model and becomes the most accurate NC system so far. However, through experimental study, we find that the prediction accuracy of Phoenix for short links is significantly worse compared with the overall prediction accuracy. Based on the observation, we propose a new NC system, named Pancake, aiming at reducing the high prediction error for short links. By introducing a two-level architecture, Pancake achieves much higher prediction accuracy than other selected existing NC systems. Through extensive experiments, we demonstrate that Pancake reduces the 90th percentile relative error by up to 25.37% from Phoenix. Moreover, Pancake converges very fast and is robust to different dimension values. For further insights, we study the performance of Pancake using a dynamic data set in addition to the widely used aggregate data sets. With varying RTTs over time, Pancake outperforms other NC systems consistently.

Routing policies and the complexity of network give rise to violations of the Triangle Inequality with respect to delay (Round-Trip Time) in the Internet. Most of the network coordinate (NC) systems, for example Vivaldi, suffer from inaccurate distance estimation due to such Triangle Inequality Violations (TIVs). In this abstract, we propose a methodology to overcome TIVs by introducing medium in Euclidean distance based NC model.

Microblogging services, such as Twitter, are among the most important online social networks(OSNs). Different from OSNs such as Facebook, the topology of microblogging service is a directed graph instead of an undirected graph. Recently, due to the explosive increase of population size, graph sampling has started to play a critical role in measurement and characterization studies of such OSNs. However, previous studies have only focused on the unbiased sampling of undirected social graphs. In this paper, we study the unbiased sampling algorithm for directed social graphs. Based on the traditional Metropolis-Hasting Random Walk (MHRW) algorithm, we propose an unbiased sampling method for directed social graphs(USDSG). Using this method, we get the first, to the best of our knowledge, unbiased sample of directed social graphs. Through extensive experiments comparing with the ”ground truth ” (UNI, obtained through uniform sampling of directed graph nodes), we show that our method can achieve excellent performance in directed graph sampling and the error to UNI is less than 10%.

Online microblogging services, as exemplified by Twitter, have become immensely popular during the latest years. However, current microblogging systems severely suffer from performance bottlenecks and malicious attacks due to the centralized architecture. As a result, centralized microblogging systems may threaten the scalability, reliability as well as availability of the offered services, not to mention the high operational and maintenance cost.

Previously, the network must be congested by probing flow before the congestion-related parameters (such as background flow and available bandwidth) can be estimated, which lead to the inaccuracy and inefficiency of today's Internet. In this work, the authors study how to estimate network state without saturating the network. By introducing the queueing theory, this study proposes a novel packet time series analysis (PTSA) framework model, which can be used to estimate the congestion-related parameters without saturating the network. The accuracy and efficiency of PTSA are validated under NS-2 simulation environment. The performance of PTSA methodology is evaluated in Schooner test-bed with a special scenario. The analytical, simulative and experimental results show that PTSA framework is more efficient to estimate network state with less aggression and higher sensitivity than those methods that need to saturate the network.

Broadcatching is a promising mechanism to improve the experience of BitTorrent users by automatically downloading files advertised through RSS feeds. However, though widely used, the mechanism itself has not been well studied. In this paper, we conducted extensive experiments on PlanetLab to evaluate the performance of Broadcatching under different typical scenarios. The results demonstrated the effectiveness of the broadcatching: it reduces the average completion time and downloading failure ratio. It also improves the overall fairness of the system: the subscribers are encouraged to share more while downloading faster, which results in the increased share ratio. Our study is the first work to systematically evaluate the benefit of broadcatching and sheds lights on how to improve performance of BitTorrrent by manipulating peer's behavior like Broadcatching.

Network coordinates (NC) system is an efficient mechanism for Internet distance prediction with scalable measurements. The intrinsical cause for the unsatisfactory accuracy of the simulation-based NC algorithms has been identified. Then Pharos, a fully decentralised and hierarchical scheme, is proposed to solve this problem. Pharos leverages multiple coordinate sets at different distance scales, with the right scale being chosen for prediction each time. We evaluate the performance of Pharos system with the King data set and latency data from PlanetLab, and compare it with the representative NC system, Vivaldi. The experimental results show that Pharos greatly outperforms Vivaldi in Internet distance prediction without adding any significant overhead. Our extensive evaluation results also demonstrate that Pharos can significantly improve the performance in distributed Internet applications, such as overlay multicast and server selection.

Peer-to-peer (P2P) paradigm has provided a disruptive market opportunity to define cost-effective multimedia streaming services, but at the same time network-oblivious P2P applications have been posing substantial technical and social challenges on network efficiency, operator economics and user performance. While end users are concerned with quality upgrade, Internet content provider (ICP) considers more on service scale and Internet service provider (ISP) focuses on operating cost. In taming P2P for a more friendly large-scale application, this paper provides a framework of evaluating the P2P media streaming application performance from perspectives of all entities involved, that is ISP, ICP and end users. Three-level performance metrics are defined, essential concerns of each party are theoretically quantified and bottlenecks in affecting quality service are identified. In handling tussles between P2P performance against ISP traffic, system scale against cost and user QoS against Security, the authors present proposals in defining an unprecedented friendly and cost-effective P2P streaming application to achieve the ideal philosophy of ‘more users=better performance+lower cost’. Based on the explorations in academy and industry, the authors envision that a large-scale streaming system will be built with a synergy of P2P and content distribution networks (CDN), and explore the feasibility of a general peer–server–peer (PSP) structure on the basis of our evaluation framework. With our analytical study and industrial deployment, this paper captures a certain essence of deploying large-scale P2P streaming application from a commercial and realistic point of view and suggests many avenues for addressing the emerging tensions between P2P application and network operators.

A Network Coordinate (NC) system is an efficient mechanism to predict Internet distance with scalable measurements. In this paper, we focus on the node churn problem – the continuous process of nodes arrival and departure – in distributed applications. Studies on Vivaldi, a representative distributed NC system, show that under node churn the prediction accuracy of the NC system will be seriously impaired. In this paper, we focus on how to handle the impact of node churn in Vivaldi. Firstly, we propose a simple solution by directly increasing the measurement frequency. Our experiments have demonstrated that this approach can reduce the harm of node churn. However, it increases the communication overhead as the measurement frequency grows. To avoid such expensive solution, we propose the design and implementation of Myth, a decentralised and fast convergence NC system. It introduces the merit of Landmark-based NC system to shorten convergence time in Vivaldi with slight extra overhead. Our experimental results show that Myth outperforms Vivaldi a lot under node churn, without compromising the performance under stable environment. Moreover, we have found that the use of Myth is a cost-effective way to achieve higher prediction accuracy; it will not only improve the prediction accuracy but also save the communication overhead.

Modeling the Internet delay structure is an important issue in designing large-scale distributed systems. However, linear models fail to characterize Triangle Inequality Violations (TIV), motivating us to research on nonlinear ones. In this paper, we propose the methodology and design of nonlinear modeling by utilizing Kernel Methods(KM), which is demonstrated effective by simulation. Moreover, our nonlinear model is easy to be applied without introducing any measurement overhead.

With the popularity of BitTorrent, improving its performance has been an active research area. Super-seeding, a special upload policy for initial seeds, improves the efficiency in producing multiple seeds and reduces the uploading cost of the initial seeders. However, the overall benefit of super seeding remains a question. In this paper, we conduct an experimental study over the performance of super-seeding scheme of BitTornado. We attempt to answer the following questions: whether and how much super-seeding saves uploading cost, whether the download time of all peers is decreased by super-seeding, and in which scenario super-seeding performs worse. With varying seed bandwidth and peer behavior, we analyze the overall download time and upload cost of super seeding scheme during random period tests over 250 widely distributed PlanetLab nodes. The results show that benefits of super-seeding depend highly on the upload bandwidth of the initial seeds and the behavior of individual peers. Our work not only provides reference for the potential adoption of super-seeding in BitTorrent, but also much insights for the balance of enhancing Quality of Experience (QoE) and saving cost for a large-scale BitTorrent-like P2P commercial application.

Network coordinates (NC) system is an efficient mechanism for Internet distance prediction with limited measurements. In this paper, we identify the intrinsical cause for the inadequate accuracy of the simulation based NC algorithms. We then propose Pharos, a fully decentralized and hierarchical scheme, to remedy this problem. Pharos leverages multiple coordinate sets at different distance scales, with the right scale being chosen for prediction each time. We evaluate the performance of Pharos system with the King data set and latency data from PlanetLab, and compare it with the representative NC system, Vivaldi. The experimental results show that Pharos outperforms Vivaldi much without adding any significant overhead.

Peer-to-peer technology gives novel opportunities to define a cost-effective multimedia streaming application, but at the same time, it brings a set of technical challenges due to its dynamic and heterogeneous nature. To guarantee QoS and facilitate management in large scale high-performance media streaming, we extend the current P2P networking towards a novel Peer-Server-Peer (PSP) architecture for media streaming, in which carefully-deployed servers form a trustworthy and controllable overlay network to stream P2P cluster peers. An analytical model is presented to calculate the quality of experience (QoE) and mapping QoS parameters to prove the effectiveness of Peer-Server-Peer streaming. Joint with the efforts in industry, we explore the feasibility of PSP streaming in the historical context of "demand economy" for media streaming and "best effort" Internet. The value of this paper lies not in its analytical study of this promising PSP concept with practical implementation but also its insight industrial perspective to attract further application.

A large number of overlay protocols have been developed, almost all of which assume each host has two-way communication capability. However, this does not hold as the deployment of firewalls and Network Address Translators (NAT) is widespread in the current Internet, which is a challenge to the design and implementation of overlay models and protocols. In this paper, we present Potential-based Multicast Tree Algorithm (PMTA) to enhance the multicast tree construction in presence of connectivity restricted hosts. We evaluate PMTA and previous multicast tree protocols based on real Internet end-to-end delay datasets. According to evaluation results, PMTA outperforms those protocols in terms of all metrics. PMTA reduces ARDP by 26%, and it also results in 23%-54% reduction in average overlay latencies. As the results suggest, PMTA can build efficient and effective multicast tree and is suitable for Internet multicast applications in the presence of connectivity restricted hosts.