Anomaly detection plays a critical role in ensuring safe, smooth, and efficient operation of machinery equipment industrial environments. With the wide deployment multimodal sensors rapid development Internet Things (IoT), data generated modern production has become increasingly diverse complex. However, traditional methods for anomaly based on single source cannot fully utilize to capture anomalies systems. To address this challenge, we propose new model environments using temporal data....

Influence maximization in social networks is of great importance for marketing new products. Signed with both positive (friends) and negative (foes) relationships pose challenges opportunities, since the influence can be leveraged to promote information propagation. In this paper, we study problem advertisement recommendation signed networks. We propose a framework characterize propagation process networks, which models dynamics individuals' beliefs attitudes towards based on recommendations...

The timely detection of equipment failure can effectively avoid industrial safety accidents. existing fault diagnosis methods based on single-mode signal not only have low accuracy, but also the inherent risk being misled by noise. In this paper, we reveal possibility using multi-modal monitoring data to improve accuracy prediction. main challenge fusion is how fuse We propose a learning framework for low-quality and high-quality data. essence, are used as compensation Firstly, optimized,...

Recently, the practical needs of ``the right to be forgotten'' in federated learning gave birth a paradigm known as unlearning, which enables server forget personal data upon client's removal request. Existing studies on unlearning have primarily focused efficiently eliminating influence requested from model without retraining scratch, however, they rarely doubted reliability global posed by discrepancy between its prediction performance before and after unlearning. To bridge this gap, we...

For ecologists, understanding the reaction of animals to environmental changes is critical. Using networked sensor technology measure wildlife and parameters can provide accurate, real-time comprehensive data for monitoring, research conservation wildlife. This paper reviews: (i) conventional detection technology; (ii) concepts applications Internet Things (IoT) in animal ecology; (iii) advantages disadvantages IoT. The current theoretical limits IoT ecology are also discussed. Although...

Because of their proficiency in capturing the category characteristics graphs, graph neural networks have shown remarkable advantages for graph-level classification tasks, that is, rumor detection and anomaly detection. Due to manipulation special means (e.g. bots) on online media, rumors may spread across whole network at an overwhelming speed. Compared with normal information, popular usually a propagation structure, especially early stage information dissemination. More specifically,...

Crowdsourcing leverages the collective intelligence of massive crowd workers to accomplish tasks in a cost-effective way. On crowdsourcing platform, it is challenging assign an appropriate way due heterogeneity both and workers. In this paper, we explore problem assigning with various skill levels different quality requirements budget constraints. We first formulate task assignment as many-to-one matching problem, which multiple are assigned task, can be successfully completed only if...

Microeconomics-inspired spectrum auctions can effectively improve the utilization for wireless networks to satisfy ever increasing service demands. Considering spatial reuse, bidding nodes without mutual interference are grouped as virtual bidders competing bands, which turns a multi-winner auction into traditional single-winner auction. To make participating bid truthfully, strategy-proof exploited allocate vacant bands. However, how fairly profits of bidder among winning is still an...

To enable dynamic spectrum access, service providers with spare (sellers) trade those who are in need of additional (buyers). In a market, the transaction result is essentially matching between sellers and buyers. Though it tempting to optimize over certain utility functions, stable more desirable, since no participants have incentives deviate from result. Existing algorithms only consider maximum number channels buyer can purchase, but ignore minimum requirement that essential support...

Performance of a collaborative task is mostly dependent on the collective effort from participants. To accomplish participatory effectively and efficiently, team formation problem (TFP) outweighs all other considerations. It even more complicated when social connections among candidates taken into account. As we can imagine, large number tasks require members to be socially close. On contrary, portion tasks, e.g., proposal review, pay attention multidimensional view, should selected variety...

The VM (Virtual Machine)-based software protection technique provides an effective solution to protect software, making it extremely difficult analyze and crack. This has become the research focus of protection. In this paper, we introduce general design ideas technique. However, there exist some vulnerabilities in design. We these detail come up with improvements mitigate them. develop a VM-based system, named NISLVMP, carry out experiments it. results show that are effective.

Social tie prediction is an important issue in social network analysis. Transfer learning often used for to address the problem of insufficient labeled training data, since few users manually annotate their relationships. In this paper, we propose TranGAN, a novel generative adversarial (GAN) based transfer framework prediction, which leverages theories as common knowledge bridge source and target network. GAN helps augment original data set by generating samples that have similar...

Femtocell provides a potential solution for poor coverage and capacity limitation in the indoor environment. For wireless service providers (WSPs), emerging femtocell market brings new opportunities as well challenges. In this paper, we propose an economic framework oligopolistic market, where selfish WSPs try to maximize their net profits by price competition resource allocation. When analyzing optimal strategy each WSP, are facing two major On one hand, it is difficult WSP acquire complete...

Abstract Intelligent monitoring of thermal power plants has been receiving growing interest. As one the promising technologies in intelligent monitoring, machine learning-based fault detections for auxiliary equipment plant have widely investigated. In this work, we constructed normal behavior models to predict current a dual-induced draft fan system by gradient-boosting tree regression. For anomaly detection, evaluated residual calculating mean value and meanwhile added control boundary...

Abstract Career planning of higher vocational students has been an issue concern and focus for colleges universities. The article uses logistic regression models to empirically analyze the correlation between career ideological political education students. results single-factor analysis show that significance psychological quality exercise is largest, its OR 1.427. multifactor Logistic awareness enhancement 0.661. greatly influenced by enhancement. In response study, this paper proposes...

IPTV provides a variety of services to users, e.g., Live TV, Video on Demand (VOD), and Catch-up which allows users freely select from huge pool program genres. A high-quality recommendation system that can well predict users' dynamic preferences is desirable improve user satisfaction. In this paper, we make the first attempt design personalized TV based tensor learning leverages temporal context information implicit feedback. We establish preference model by carefully analyzing influence...

This paper describes and evaluates a predictive algorithm for bandwidth conservation enhanced situation awareness. In such system, knowledge of position intent the participants involved in operation must be readily available, distributed to other units by communications link. objective can achieved predicting tracks each unit ahead time, based on past kinematic history. If predicted is within error requirements, then track estimate not transmitted over communication link, which results...

Nowadays, smartphones have become indispensable for people and many users may browse information on their during walking. This raises severe concerns about pedestrian safety when distracted inadvertently walks from the sidewalk onto street. In this paper, we present WatchOut, a convenient alerting system based fine-grained step classification (e.g., flat, up down ramp) event detection entering street, turn) in an urban environment. contrast to existing systems that rely shoe-mounted sensors,...

The targeted covert collection of Internet data, which can not only effectively hide users' traces but also efficiently navigate to target plays an important role in three-dimensional social security. Due big data generated by online networks, large-scale network analysis usually confront with unbearable computational complexity, resulting failing achieve expected results within the specified response time. To retrieve and analyze most predetermined time, we study problem multiple...