Currently, online social networks such as Facebook, Twitter, Google+, LinkedIn, and Foursquare have become extremely popular all over the world play a significant role in people¿s daily lives. People access OSNs using both traditional desktop PCs new emerging mobile devices. With more than one billion users worldwide, are venue of innovation with many challenging research problems. In this survey, we aim to give comprehensive review state-of-the-art related user behavior from several...

Coronavirus disease 2019 (COVID-19) poses massive challenges for the world. Public sentiment analysis during outbreak provides insightful information in making appropriate public health responses. On Sina Weibo, a popular Chinese social media, posts with negative are valuable analyzing concerns. 999,978 randomly selected COVID-19 related Weibo from 1 January 2020 to 18 February analyzed. Specifically, unsupervised BERT (Bidirectional Encoder Representations Transformers) model is adopted...

Recently, Deepfake has drawn considerable public attention due to security and privacy concerns in social media digital forensics. As the wildly spreading videos on Internet become more realistic, traditional detection techniques have failed distinguishing between real fake. Most existing deep learning methods mainly focus local features relations within face image using convolutional neural networks as a backbone. However, are insufficient for model training learn enough general information...

Bandwidth reservation can effectively improve the service quality for data transfers because of dedicated network resources. However, it is difficult to achieve a desired tradeoff between resource utilization and reliable bandwidth guarantees with time-varying traffic. In this article, we study novel solution based on distributed traffic monitoring control applications that require guarantees. proposed solution, designated allocated an application in advance according its maximum peak, idle...

As a face manipulation technique, the misuse of Deepfakes poses potential threats to state, society, and individuals. Several countermeasures have been proposed reduce negative effects produced by Deepfakes. Current detection methods achieve satisfactory performance in dealing with uncompressed videos. However, videos are generally compressed when spread over social networks because limited bandwidth storage space, which generates compression artifacts inevitably decreases. Hence, how...

Detecting forgery videos is highly desirable due to the abuse of deepfake. Existing detection approaches contribute exploring specific artifacts in deepfake and fit well on certain data. However, growing technique these keeps challenging robustness traditional detectors. As a result, development has reached blockage. In this article, we propose perform from an unexplored voice-face matching view. Our approach founded two supporting points: first, there high degree homogeneity between voice...

Our daily lives have been immersed in widespread location-based social networks (LBSNs). As an open platform, LBSNs typically allow all kinds of users to register accounts. Malicious attackers can easily join and post misleading information, often with the intention influencing users' decisions urban computing environments. To provide reliable information improve experience for legitimate users, we design implement DeepScan, a malicious account detection system LBSNs. Different from existing...

Deepfake brings huge and potential negative impacts to our daily lives. As the real-life videos circulated on Internet become more authentic, most existing detection algorithms have failed since few visual differences can be observed between an authentic video a one. However, forensic traces are always retained within synthesized videos. In this study, we present noise-based model, NoiseDF for short, which focuses underlying noise left behind particular, enhance RIDNet denoiser extract...

By leveraging deep neural networks, recent face swapping techniques have performed admirably in generating faces that maintain consistent identities. Nevertheless, while these methods accurately transfer source identities, they often struggle to preserve important attributes (such as head poses, expressions, and gaze directions) the target faces. As a consequence, current research this domain has not resulted satisfactory performance. In paper, we propose an efficient attribute-preserving...

In order to improve the resource utilization efficiency of heterogeneous multi-robots, minimize execution time multi-type tasks, effectively maintain load balancing robot resources, solve problem multiple resources and difficult find a near-optimal solution for multi-robot collaborative planning, task allocation strategy combining improved particle swarm optimization greedy (IPSO-G) algorithm is proposed. The divided into two steps: First, used search combination tasks robots; after that,...

Conversion Rate (CVR) prediction in modern industrial e-commerce platforms is becoming increasingly important, which directly contributes to the final revenue. In order address well-known sample selection bias (SSB) and data sparsity (DS) issues encountered during CVR modeling, abundant labeled macro behaviors (i.e., user's interactions with items) are used. Nonetheless, we observe that several purchase-related micro specific components on item detail page) can supplement fine-grained cues...

The fast development of Deepfake has brought huge current and potential future negative impacts to our daily lives. As the circulating popular videos have become difficult be distinguished by human eyes, various detection approaches been attempted using deep learning models. However, even though some existing methods achieved reasonable performance with respect statistical evaluation metrics, actual underlying forensic traces barely discussed. In this study, we investigate special noise...

The training of Large Language Models (LLMs) for specialized applications like autonomous driving faces significant data privacy challenges. Federated Learning (FL) offers a solution by enabling local usage while preserving privacy. In this paper, we introduce Gradient Priority-based (FedGPL), novel strategy to enhance the efficiency LLM in vehicles. FedGPL precomputes gradients on server identify critical model layers, allowing vehicles selectively update these layers with data. This...

Numerous new broadband and multimedia applications have emerged in recent years, which streaming video services are particularly of interest. Unlike data communication, requires a subjective measure quality referred to as the Quality Experience (QoE), but evaluation is very time consuming complex. We proposing an objective approach provide mapping between QoE Service (QoS), Video Metric (VQM) utilized indicate QoE. analyzed emulation results derived simple formula estimate from QoS...

Popular Internet services in recent years have shown that remarkable things can be achieved by harnessing the power of masses. However, crowd-sourcing systems also pose a real challenge to existing security mechanisms deployed protect services, particularly those tools identify malicious activity detecting activities automated programs such as CAPTCHAs.

It is often difficult to separate the highly capable “experts” from average worker in crowdsourced systems. This especially true for challenge application domains that require extensive domain knowledge. The problem of stock analysis one such domain, where even paid, well-educated experts are prone make mistakes. As an extremely challenging space, “wisdom crowds” property many applications rely on may not hold. In this article, we study evaluating and identifying context SeekingAlpha...

Popular Internet services in recent years have shown that remarkable things can be achieved by harnessing the power of masses. However, crowd-sourcing systems also pose a real challenge to existing security mechanisms deployed protect services, particularly those tools identify malicious activity detecting activities automated programs such as CAPTCHAs. In this work, we leverage access two large crowdturfing sites gather corpus ground-truth data generated campaigns. We compare and contrast...

As the development of Virtual Reality and Augmented (VR/AR) technology rapidly advances, learning in an artificial immersive environment becomes increasingly feasible. Such emerging not only facilitates promotes efficient process, but also reduces cost access to materials environments. Current research mainly focuses on environments adaptive methods based interactions between trainees environment. However, valuable human biometric data available environments, such as eye gaze controller...