Mobile-edge computing (MEC), as an emerging paradigm, allows app vendors to deploy their mobile and/or IoT applications on edge servers deliver low-latency services users. However, when server needs serve excessive users concurrently, severe interference is incurred, which immediately reduces users' achievable data rates and, consequently, impacts perceived service quality. This a major challenge the vendor's attempt minimize resources required for serving its with satisfactory To tackle...

Few-shot Named Entity Recognition (NER) spotlights the tag of novel entity types in data-limited scenarios or lower-resource settings. Advances with Pre-trained Language Models (PLMs), including BERT, GPT, and their variants, have driven tremendous strategies to leverage context-dependent representations exploit predefined relational cues, yielding significant gains witnessing unseen entities. Nevertheless, a fundamental issue exists prior efforts regarding susceptibility adversarial attacks...

As a new computing paradigm, crowd-based cooperative aims at effective management and the coordinated use of crowd resources. In task allocation (CBCTA), it is necessary to ensure suitability high-quality collaboration resources for computer supported work. Generally, high matching rate between resource requirements can achieve optimal parameter configuration, whereas ensures quality success rates task. This article proposes methodology optimize model solving CBCTA problems in...

The key to discover potential opportunity information in cross-organisation business processes (COBPs) is identify the primary roles and actors, i.e. how obtain their associations according interactive behaviours within complex social networks. of COBPs commonly considered important explicitly related with activities contained COBPs. In this paper, we define a role as configurable resource model integrating capabilities knowledge required qualified actors. Furthermore, introduce two networks...

Most spatial crowdsourcing systems are designed in a static mode with tasks allocated based on the historical interactions data between crowd participants and tasks. However, these task assignment algorithms usually ignore long-term feedback interactive systems, resulting performance degradation. Though reinforcement learning naturally fits problem of maximizing long term rewards, deep learning-based is still facing challenge crowdsourcing. To address issues, this paper investigates which we...

To cope with time-attribute and variations of event distribution in dynamic evolving process, an streaming process mining based on time series prediction hybrid heuristic miner is proposed. A improved post-task activity logs to optimize the initial particle for Particle Swarm Optimization. Furthermore, "aging factor" attribute also designed adaptive global optimization. Besides, time-related Process Decision Indicator(PDI) defined as a pattern observable identify domain-independent evolution...

Abstract In recent years, the transformative evolution of cloud computing has reshaped organizational practices by enabling outsourcing web service applications. This shift led to emergence environment, characterized involvement Cloud Service Providers (CSPs) and intelligent Composition (CSC) become pivotal in this context, playing a crucial role enhancing efficiency, Quality (QoS), customer satisfaction through aggregation diverse Services (CSs) create composite services. However, vast...

Event Causality Identification (ECI) focuses on extracting causal relations between events in texts. Existing methods for ECI primarily rely features and external knowledge. However, these approaches fall short two dimensions: (1) a text often lack explicit clues, (2) knowledge may introduce bias, while specific problems require tailored analyses. To address issues, we propose SemDI - simple effective Semantic Dependency Inquiry Network ECI. captures semantic dependencies within the context...

Reliable Internet of Things (IoT) service discovery is a significant task in intelligent service-oriented IoT systems. Collaborative filtering (CF) turns out to be an effective solution discovery. However, the traditional CF framework facing following challenges: inefficiency learning high-order interactions between users and services ineffectively making use geographical location information. Moreover, deploying real-world distributed systems poses another challenge terms reliability...

Social network has emerged as an important paradigm in modern business operation. Outsourcing tasks to social helps organisations mitigate the shortage of skill or expertise some domain. Expert team discovery is problem complex collaborative networks. Existing expert models need traverse every candidate until optimal solution found, which would lead high computational cost. In this paper, a formation model proposed outsource order contract search space for seeded candidates, selects...

Purpose This study aims to conduct the aircraft electrical wiring interconnection system (EWIS) safety risk assessment process abundantly and hierarchically establish index considering weights interrelationships of different levels indices. Design/methodology/approach Due failure EWIS being multifactorial, hidden diverse, this paper divides factors influencing into 3 primary indices, 13 secondary indices 38 tertiary Taking open circuit (OCF) short (SCF) as examples, calculate based on...