In this paper, we first introduce the definition of a fine-grained emotion role, which consists of two dimensions: emotion orientation and emotion influence. We then propose a Multi-dimensional Emotion Role Mining model to determine a user's emotion role in online social networks. Specifically, we tend to identify emotion roles by combining a set of features that reflect a user's online emotional status, including emotional macro-micro relations, historical emotion preference, structural factor, temporal factor and emotion change factor. Experiment results on a real-life micro-blog retweeting dataset show that the classification accuracy of the proposed model can achieve up to 90.1%.
Kleinberg proposed a family of small-world networks to explain the navigability of social networks. However, the underlying mechanism driving real networks to be navigable is not yet well understood. In this paper, we model the network formation as a game in which people seek for both high reciprocity and long-distance relationships. We show that the navigable small-world network is a Nash Equilibrium of the game. Moreover, we prove that the navigable small-world equilibrium tolerates collusions of any size and arbitrary deviations of a large random set of nodes, while non-navigable equilibria do not tolerate small group collusions or random perturbations.
The aim of this paper is to propose an innovative grouping approach to enhance the interaction and collaboration among peers by considering the complementary degree of students learning state and their social networks. In order to validate our approach, experiments were designed with a group of students and tested with E-learning system developed using genetic algorithm to produce better grouping results. The outcomes clearly indicate that the proposed approach can generate high heterogeneous grouping results. The technical contribution of this paper can be implemented in different MOOCs platforms with thousands of students, with regard to find peers for collaborative works.
There is a growing number of business process management systems under development both in academia and in practice. At the same time, the advent of big data analytics has changed the scope of such systems. However, reference architectures for business process management systems date back 20 years and, consequently, are not up to date with modern developments. Therefore, this paper proposes an up-to-date reference architecture, called BPMS-RA, for modern business process management system, which is based on recent literature and on existing commercial implementations.
Special Issue on the Economics of Security and Privacy: Guest Editors' Introduction
The widespread adoption of the Internet of Things (IoT) has created a demand for ubiquitous connectivity of IoT devices into the Internet. While end-to-end connectivity for IoT requires in practice IPv6, a vast majority of nodes in Internet are only IPv4-capable. To address this issue, the use of Network Address Translation (NAT) at the IoT network boundary becomes necessary. However, the constrained nature of the IoT devices hinders the integration of traditional NAT traversal architectures through IoT networks. In this paper, we introduce a novel transition mechanism that transparently enables IoT devices behind NATs to connect across different network-layer infrastructures.
TOIT Reviewers for 2017
Cognitive computing is changing the way the world is seen, and indeed cognitive applications are almost deployed in the cloud. However, the attention to security, privacy, network connection and other issues has enabled a paradigm shift, moving the cognitive computing from the cloud to the network's edge. This paper firstly introduces a new network architecture for edge cognitive computing (ECC),then describes the ECC evolution process and design issues in details, and conducts an experiment platform of dynamic service migration based on mobile user behavioral cognition.The experimental results show that proposed ECC architecture has ultra-low latency and high user experience.
Mobile crowdsensing becomes a promising technology for the emerging Internet of Things (IoT) applications in smart environments. In this paper, we introduce a framework enabling mobile crowdsensing in fog environments with a hierarchical scheduling strategy. We first introduce the crowdsensing framework that has a hierarchical structure to organize different resources. Since different positions and performance of fog servers influent the quality of service of IoT applications, we formulate a scheduling problem in the hierarchical fog structure and solve it by using a deep reinforcement learning based strategy. From extensive simulation results, our solution outperforms other scheduling solutions for mobile crowdsensing
Fog computing has drawn significant research interest as it focuses on bringing Cloud-based services closer to IoT users. To fully leverage the capabilities of distributed, resource-constrained and heterogeneous Fog nodes, applications that are decomposed into inter-dependent modules, can be deployed orderly over the nodes based on their latency-sensitivity. Here, we propose a latency-aware Application Module management policy for Fog to meet the diverse latency-driven issues of different applications. It aims to ensure QoS in terms of meeting the deadline and optimizes energy usage in Fog. Simulation experiments of the proposed policy, demonstrate meaningful improvement in performance over alternative latency-aware strategies.
Smart environments (SE) are expected to significantly benefit from integrated edge-fog-cloud (IEFC) paradigm. High-order Bi-Lanczos has emerged as a powerful tool in SE. How to complete data processing without compromising privacy is a challenge in SE. In this work, we propose a novel privacy-preserving high-order Bi-Lanczos (PPHOBL) scheme in IEFC paradigm for SE. We firstly propose a privacy-preserving big data processing model using IEFC paradigm. Subsequently, making use of the model, we present a PPHOBL scheme. Finally, we analyze the scheme based on an intelligent surveillance system. The results demonstrate that the superiority of the scheme for SE.
The massive increase of IoT devices and their collected data raises the question how to analyze all that data. Edge computing may provide a suitable compromise, but the question remains: how much processing should be done locally vs. offloaded to other edge devices? The diverse application requirements and limited resources at the edge extend these challenges. In this article we propose Oops, an optimization framework to decide and adapt the resource management at runtime, and orchestrate the edge devices in a distributed manner. Experimental results show a significantly reduced runtime overhead while even increasing the user utility compared to state-of-the-art.
Online rating systems are often target of manipulation attacks based on posted unfair ratings. In this paper we propose an iterative algorithm to assess rating scores which leverages information about users and score provenance and takes into account the distances between rating options. We prove convergence of our iterative ranking algorithm. We have implemented and tested our rating method on simulated data and world datasets. The experimental results demonstrate that our model provides realistic rating scores even in the presence of massive amount of unfair ratings and outperforms existing ranking algorithms.
Mobile-, edge-, and cloud-computing have the potential to form a computing continuum for disruptive applications. The choice of where in the continuum to execute different functionalities is made at run time, based on context and requirements, with the goal of minimizing latency and battery consumption, and maximizing availability. We propose A3-E, a unified model that exploits the Function-as-a-Service to abstract away the heterogeneity of the continuum. Experiments show that A3-E is capable of dynamically routing the application's requests to the continuum, reducing latency by up to $90$\% when using edge infrastructures, and battery consumption by $74$\% when offloading mobile computation.
We present a new spatio-temporal incentive-based approach to achieve a geographically balanced coverage of crowdsourced services. The proposed approach is based on a new spatio-temporal incentive model that considers multiple parameters including location entropy and spatio-temporal density to encourage the participation of crowdsourced service providers. We present a greedy network flow algorithm that offers incentives to redistribute the crowdsourced service providers to improve the crowdsourced coverage balance. A novel participation probability model is introduced to estimate the expected number of crowdsourced service providers movement based on spatio-temporal features. Experimental results validate the efficiency and effectiveness of the proposed approach.
We propose in this paper a hybrid approach to improve the design structure of Web service interfaces and fix antipatterns as a combination of both deterministic and heuristic-based methods. The first step consists of a deterministic method using graph partitioning-based technique to split the operations of a large service interface into more cohesive interfaces, each one representing a distinct abstraction. Then, the produced interfaces will be checked using a heuristic-based approach based on the non-dominated sorting genetic algorithm (NSGA-II) to correct potential antipatterns while reducing the interface design deviation to avoid taking the service away from its original design.