Embedding-based recommender systems rely on historical interactions to model users, which poses challenges for recommending new known as the user cold-start problem. Some approaches incorporate social networks deduce preferences based circles of users solve problem sparse features. However, such methods have difficulty distinguishing between superficial correlations and causal relationships in behaviors, leading inaccuracies predicting preferences. To address aforementioned issues, we...

In the social production system, image data are rapidly generated from almost all fields such as factories, hospitals, and transportation, promoting higher requirements for anomaly detection technologies, including low consumption, adaptability, accuracy. However, existing methods fragile to heterogeneous by complex systems tend require strong computing power resource support. To address above problems, a knowledge-driven framework is proposed, in which local feature enhancement method...

Mobile cloud computing provides powerful and storage capacity on managing GPS big data by offloading vast workloads to remote clouds. For the mobile applications with urgent or communication deadline, it is necessary reduce workload transmission latency between devices This can be technically achieved expanding cloudlets that are moving co-located Access Points (APs). However, not-trivial place such movable efficiently enhance service for dynamic context-aware applications. In view of this...

With the advent of IoT (Internet Things) age, considerable web services are emerging rapidly in service communities, which places a heavy burden on target users' selection decisions. In this situation, various techniques, e.g., collaborative filtering (i.e., CF) is introduced recommendation to alleviate burden. However, traditional CF-based approaches often assume that historical user-service quality data centralized, while neglect distributed situation. Generally, involves inevitable...

Anomaly detection plays an increasingly important role in various fields for critical tasks such as intrusion cybersecurity, financial risk detection, and human health monitoring. A variety of anomaly methods have been proposed, a category based on the isolation forest mechanism stands out due to its simplicity, effectiveness, efficiency, e.g., iForest is often employed state-of-the-art detector real deployment. While majority forests use binary structure, framework LSHiForest has...

Text anomaly detection is crucial for identifying spam, misinformation, and offensive language in natural processing tasks. Despite the growing adoption of embedding-based methods, their effectiveness generalizability across diverse application scenarios remain under-explored. To address this, we present TAD-Bench, a comprehensive benchmark designed to systematically evaluate approaches text detection. TAD-Bench integrates multiple datasets spanning different domains, combining...

Medical time series has been playing a vital role in real-world healthcare systems as valuable information monitoring health conditions of patients. Accurate classification for medical series, e.g., Electrocardiography (ECG) signals, can help early detection and diagnosis. Traditional methods towards rely on handcrafted feature extraction statistical methods; with the recent advancement artificial intelligence, machine learning deep have become more popular. However, existing often fail to...

Personalized federated learning (PFL) has recently gained significant attention for its capability to address the poor convergence performance on highly heterogeneous data and lack of personalized solutions traditional (FL). Existing mainstream approaches either perform aggregation based a specific model architecture leverage global knowledge or achieve personalization by exploiting client similarities. However, former overlooks discrepancies in distributions indiscriminately aggregating all...

Graph Contrastive Learning (GCL), as a primary paradigm of graph self-supervised learning, spurs fruitful line research in tackling the data sparsity issue by maximizing consistency user/item embeddings between different augmented views with random perturbations. However, diversity, crucial metric for recommendation performance and user satisfaction, has received rather little attention. In fact, there exists challenging dilemma balancing accuracy diversity. To address these issues, we...

Graph Neural Networks (GNNs) are widely applied on graph-level tasks, such as node classification, link prediction and graph generation. Existing GNNs mostly adopt a message-passing mechanism to aggregate information with their neighbors, which often makes similar after rounds of aggregations leads oversmoothing. Although recent works have made improvements by combining different message aggregation methods or introducing semantic encodings priors, these based still fail combat oversmoothing...

With the development of Internet Things (IoT) technology, a vast amount IoT data is generated by mobile applications from devices. Cloudlets provide paradigm that allows and to be offloaded devices cloudlets for processing storage through access points (APs) in Wireless Metropolitan Area Networks (WMANs). Since most relevant personal privacy, it necessary pay attention transmission security. However, still challenge realize goal optimizing time, energy consumption resource utilization with...

Anomaly detection is one of the crucial research topics in artificial intelligence, encompassing various fields such as health monitoring, network intrusion detection, and fraud financial transactions. Deep anomaly (DAD) methods are considered effective approaches for addressing complex problems. Among them, deep isolation forest have gained rapid development recently due to their simplicity parameter turning efficiency model training. The existing all based on representation learning, while...

Anomaly detection is a significant but challenging data mining task in wide range of applications. Different domains usually use different ways to measure the characteristics and define anomaly types. As result, it big challenge develop versatile framework that can be universally applied with satisfactory performance most, if not all, In this article, we propose generic isolation forest based ensemble named EDBHiForest, which spaces arbitrary distance measures. It realized through embedding...

Abstract Anomaly detection plays a crucial role in many Internet of Things (IoT) applications such as traffic anomaly for smart transportation and medical diagnosis healthcare. With the explosion IoT data, on data streams raises higher requirements real-time response strong robustness large-scale arriving at same time various application fields. However, existing methods are either slow or application-specific. Inspired by edge computing generic technique, we propose an isolation forest...

With the rapid resource requirements of Internet Things applications, cloud computing technology is regarded as a promising paradigm for provision. To improve efficiency and effectiveness services, it essential to fairness achieve energy savings. However, still challenge schedule virtual machines in an energy-efficient manner while taking into consideration fairness. In view this challenge, fair machine scheduling method applications designed article. Specifically, are analyzed formal way....

Anomaly detection is one of the most important data mining tasks in many real-life applications such as network intrusion for cybersecurity and medical diagnosis healthcare. In big era, these demand fast versatile anomaly capability to handle various types increasingly huge-volume data. However, existing methods are either slow due high computational complexity, or unable deal with complicated anomalies like local anomalies. this paper, we propose a novel method named OPHiForest use order...

Summary With the development of artificial intelligence, cloud‐edge computing and virtual reality, industrial design that originally depends on human imagination power can be transitioned to metaverse applications in smart manufacturing, which offloads services cloud edge platforms for enhancing quality service (QoS), considering inadequate terminal devices like sensors access points (APs). However, large overhead privacy exposure occur during data transmission cloud, while (ECDs) are at...