Search

Part 4: Modelling: Structures and Implementations; International audience; We identify the algebraic structure of the material histories generated by concurrent processes. Specifically, we extend existing categorical theories of resource convertibility to capture concurrent interaction. Our formalism admits an intuitive graphical presentation via string diagrams for proarrow equipments.

Part 5: Invited Paper; International audience; Many interesting systems in several disciplines can be modeled as networks of nodes that can store and exchange data: pervasive systems, edge computing scenarios, and even biological and bio-inspired systems. These systems feature inherent complexity, and often simulation is the preferred (and sometimes the only) way of investigating their behavior; this is true both in the design phase and in the verification and testing phase. In this tutorial paper, we provide a guide to the simulation of such systems by leveraging Alchemist, an existing research tool used in several works in the literature. We introduce its meta-model and its extensible architecture; we discuss reference examples of increasing complexity; and we finally show how to configure the tool to automatically execute multiple repetitions of simulations with different controlled variables, achieving reliable and reproducible results.

Part 1: Tutorial; International audience; Actor languages decouple communication from synchronization, which makes them suitable for distributed and scalable applications with flexible synchronization patterns, but also facilitates analysis. ABS is a timed actor-based modeling language which supports cooperative scheduling and the specification of timing- and resource-sensitive behavior. Cooperative scheduling allows a process which is executing in an actor to be suspended while it is waiting for an event to occur, such that another process in the same actor can execute. Timed semantics allows the specification of the temporal behavior of the modeled system. Resource-sensitive behavior takes a supply and demand perspective of execution, relating cost centers which provision resources to processes which require them. These modeling concepts have been used in ABS to model cloud computing, e.g., data-processing applications running on the Hadoop platform and micro-services running on containers orchestrated by Kubernetes. In this tutorial, we present ABS and its execution environment, and discuss the use of cooperative scheduling and resources in modeling cyber-physical systems and applications deployed on virtualized infrastructure.

Part 2: Communications: Types and Implemenations; International audience; Detecting programming errors and vulnerabilities in software is increasingly important, and building tools that help developers with this task is a crucial area of investigation on which the industry depends. In object-oriented languages, one naturally defines stateful objects where the safe use of methods depends on their internal state; the correct use of objects according to their protocols is then enforced at compile-time by an analysis based on behavioral types.We present Java Typestate Checker (JATYC), a tool based on the Checker Framework that verifies Java programs with respect to typestates. These define the object’s states, the methods that can be called in each state, and the states resulting from the calls. The tool offers the following strong guarantees: sequences of method calls obey to object’s protocols; completion of objects’ protocols; detection of null-pointer exceptions; and control of the sharing of resources through access permissions.To the best of our knowledge, there are no research or industrial tools that offer all these features. In particular, the implementation of sharing control in a typestate-based tool seems to be novel, and has an important impact on programming flexibility, since, for most programs, the linear discipline imposed by behavioral types is too strict.Sharing of objects is enabled by means of an assertion language incorporating fractional permissions; to lift from programmers the burden of writing the assertions, JATYC infers all of these by building a constraint system and solving it with Z3, producing general assertions sufficient to accept the code, if these exist.

Part 3: Large-Scale Decentalised Systems; International audience; Business processes, data, and run-time control state are all key elements in the design of enterprise applications. However, the different layers for processes, data, and control are usually represented using different technologies that must be explicitly combined and kept in sync.We introduce ReGraDa, a process definition and programming language inspired by the declarative business process language Dynamic Condition Response Graphs and targeting the integrated description of business processes, data and run-time state as reactive graph data. ReGraDa processes compile directly to a data-centric graph-based system (neo4j/cypher), allowing for the database system to manage a process without the need for an external execution engine.The underlying graph structure allows for the definition of native data relations between data elements that subsumes the integrity guaranties of relational and the semi-structured data models. Graph relationships are also used to represent data-dependency and control-flow in the database.We present the reactive semantics of the language, a translation from ReGraDa to cypher, evaluate its performance, and briefly discuss future work and applications.

Part 1: Cloud and Fog Computing; International audience; Multi-tenancy enables cost-effective SaaS through resource consolidation. Multiple customers, or tenants, are served by a single application instance, and isolation is enforced at the application level. Service load for different tenants can vary over time, requiring applications to scale in and out. A large class of SaaS providers operates legacy applications structured around a relational (SQL) database. These applications achieve tenant isolation through dedicated fields in their relational schema and are not designed to support scaling operations. We present a novel solution for scaling in or out such applications through the migration of a tenant’s data to new application and database instances. Our solution requires no change to the application and incurs no service downtime for non-migrated tenants. It leverages external tables and foreign data wrappers, as supported by major relational databases. We evaluate the approach using two multi-tenant applications: Iomad, an extension of the Moodle Learning Management System, and Camunda, a business process management platform. Our results show the usability of the method, minimally impacting performance for other tenants during migration and leading to increased service capacity after migration.

Part 2: Poster Papers; International audience; In this work we investigate the coupling of Flex-E and X-Ethernet technologies as a means to realize interface slicing and fast switching respectively. Flex-E is investigated as a key technology that is able to split a physical interface into isolated sub-channels, decouple MAC rate from PHY rate and achieve hard bandwidth isolation. X-Ethernet technology introduces Ethernet PCS switching, eliminates table lookup and buffer queuing and will be investigated as a fast switching mechanism that is exploiting the PCS layer relay. We present evaluation results from implementation activities that showcase the ability of an integrated Flex-E/X-Ethernet solution to enable network slicing features in transport networks. Furthermore we present a technical approach on the way the control plane of Flex-E/X-Ethernet network can be integrated with an orchestration and management solution.

Part 4: Business Processes; International audience; Corporate governance has been severely condemned, as a result of company failures around the world. Regulatory and legislative measure have been introduced in response to these failures. However, arguments by sceptics advocate that it is time consuming and costly to comply these legislative measures that cause overregulation. These measures further do not always add value to business initiatives and adherence to these measures cannot be enforced or guaranteed. This paper argues for the use of business process management systems (BPMSs) to improve corporate governance. The “dynamics capabilities theory model of IT-enabled organisational performance” [1] is applied in this study as theoretical underpinning. A theoretical framework is proposed for IT-enabled and IT-enforced corporate governance compliance using a BPMS, after collecting data from seven BPMS user companies and a BPMS vendor company in South Africa.

Part 2: Cloud and Systems; International audience; Coping with failures in modern distributed storage systems that handle massive volumes of heterogeneous and potentially rapidly changing data, has become a very important challenge. A common practice is to utilize fault tolerance methods like Replication and Erasure Coding for maximizing data availability. However, while erasure codes provide better fault tolerance compared to replication with a more affordable storage overhead, they frequently suffer from high reconstruction cost as they require to access all available nodes when a data block needs to be repaired, and also can repair up to a limited number of unavailable data blocks, depending on the number of the code’s parity block capabilities. Furthermore, storing and placing the encoded data in the federated storage system also remains a challenge. In this paper we present Fed-DIC, a framework which combines Diagonally Interleaved Coding on client devices at the edge of the network with organized storage of encoded data in a federated cloud system comprised of multiple independent storage clusters. The erasure coding operations are performed on client devices at the edge while they interact with the federated cloud to store the encoded data. We describe how our solution integrates the functionality of federated clouds alongside erasure coding implemented on edge devices for maximizing data availability and we evaluate the working and benefits of our approach in terms of read access cost, data availability, storage overhead, load balancing and network bandwidth rate compared to popular Replication and Erasure Coding schemes.

Part 5: Big Data and Machine Learning; International audience; Researchers have been experimenting with various drivers of the diffusion rate like sentiment analysis which only considers the presence of certain words in a tweet. We theorize that the diffusion of particular content on Twitter can be driven by a sequence of nouns, adjectives, adverbs forming a sentence. We exhibit that the proposed approach is coherent with the intrinsic disposition of tweets to a common choice of words while constructing a sentence to express an opinion or sentiment. Through this paper, we propose a Custom Weighted Word Embedding (CWWE) to study the degree of diffusion of content (retweet on Twitter). Our framework first extracts the words, create a matrix of these words using the sequences in the tweet text. To this sequence matrix we further multiply custom weights basis the presence index in a sentence wherein higher weights are given if the impactful class of tokens/words like nouns, adjectives are used at the beginning of the sentence than at last. We then try to predict the possibility of diffusion of information using Long-Short Term Memory Deep Neural Network architecture, which in turn is further optimized on the accuracy and training execution time by a Convolutional Neural Network architecture. The results of the proposed CWWE are compared to a pre-trained glove word embedding. For experimentation, we created a corpus of size 230,000 tweets posted by more than 45,000 users in 6 months. Research experimentations reveal that using the proposed framework of Custom Weighted Word Embedding (CWWE) from the tweet there is a significant improvement in the overall accuracy of Deep Learning framework model in predicting information diffusion through tweets.

Part 3: eBusiness; International audience; Technologically, competition is fierce in the banking industry, as such, banks need to offer cutting-edge technology and still make sure the banking solutions provided keep customers satisfied, as customers could switch between banks with minimum inconvenience. Electronic service quality measures the level of satisfaction deliver to bank customers through online banking digital solutions. This paper provides an investigation of electronic service quality based on customers’ perceptions in South Africa. The paper presents findings based on quantitative data from 184 online banking customers using E-S-QUAL measurement scale. The results show that banks are meeting customer expectations of service quality delivery, although there is room for improvement. The study has practical implications to financial institutions in South Africa, as it highlights areas of attention to improve service quality delivered to online banking customers.

Part 1: Regular Papers; International audience; Next generation networks – spanning from wide area networks to intra-datacenters – are expected to encompass devices and sub-devices from different vendors. Inter operability and vendor neutrality are now key topics for researchers and companies as a way to reduce costs avoiding to be tied to a single vendor. Also the management system is expected to be disaggregated from the hardware and possibly developed by a different company. An example is the white box, which disaggregates the software from the hardware and can be composed of modules from different vendors. Currently, NETCONF – based on YANG – is the protocol considered for the (re)configuration of white boxes. Discussions are being carried on to identify commonly agreed information models to guarantee vendor-neutral configuration and management of next generation networks.In addition, emerging services (e.g., 5G) will impose the involvement of a variety of technologies (e.g., networks, datacenters) and resources (e.g., compute, storage, radio). Such heterogeneity will imply the cooperation of several business actors to bring services to the users. In such a scenario, the management of resources will require their abstraction in order to satisfy confidentiality as well as to increase scalability and, again, to provide information that is neutral with respect to the related vendor and technology.In this paper, we propose to use NETCONF also for the exchange of resource information among different business actors considering the several network segments (wide area network, data centers, radio segment) involved in a 5G vertical. To this purpose, resource abstraction is proposed based on a YANG data model. An experiment will involve the (re)configuration of a wide area network composed of white boxes. Moreover, resource information (virtual links) will be exchanged through NETCONF with the objective of orchestrating resources.

International audience; As ubiquitous computer and sensor systems become abundant, the potential for automatic identification and tracking of human behaviours becomes all the more evident. Annotating complex human behaviour datasets to achieve ground truth for supervised training can however be extremely labour-intensive, and error prone. One possible solution to this problem is activity discovery: the identification of activities in an unlabelled dataset by means of an unsupervised algorithm. This paper presents a novel approach to activity discovery that utilises deep learning based language production models to construct a hierarchical, tree-like structure over a sequential vector of sensor events. Our approach differs from previous work in that it explicitly aims to deal with interleaving (switching back and forth between activities) in a principled manner, by utilising the long-term memory capabilities of a recurrent neural network cell. We present our approach and test it on a realistic dataset to evaluate its performance. Our results show the viability of the approach and that it shows promise for further investigation. We believe this is a useful direction to consider in accounting for the continually changing nature of behaviours.

Part 2: Security Testing; International audience; Penetration testing (PT) is nowadays one of the most common and used activities to evaluate a given asset’s security status. Penetration testing aims to secure networks and highlights the security issues of such networks. More precisely, PT, which is used for proactive defense and information systems protection, is a structured process, made up of various phases that typically needs to be carried out within a limited period.In this work, we first define a modular semi-automatic approach, which allows us to collect and integrate data from various exploit repositories. These data are then used to provide the penetration tester (i.e., the pentester) with information on the best available tools (i.e., exploits) to conduct the exploitation phase effectively. Also, the proposed approach has been implemented through a proof of concept based on the Nmap Scripting Engine (NSE), which integrates the features provided by the Nmap Vulscan vulnerability scanner, and allows, for each vulnerability detected, to find the most suitable exploits for this vulnerability. We remark that the proposed approach is not focused on the vulnerability mapping phase, which is carried out through Vulscan. Instead, it is focused on the automatic finding of the exploits that can be used to take advantage of the results achieved by such a phase.

Part 1: Block Chain; International audience; Public procurement constitutes a core government function for providing goods and services to citizens. The overall success of a digitized public-procurement function yields progress and economic growth for the nation. In this research, we analyze the potential of blockchain-based systems to enhance effectiveness, ease, and transparency in public procurement in the case of Nigeria and identify the current challenges facing public procurement, i.e., lack of trust and transparency among critical stakeholders in the procurement process, systems that only weakly support transaction recording and documentation, complex process structures, corruption in institutions involved in the procurement process. To address these issues, a blockchain-based framework is developed to enable interoperability of information-systems involved in the procurement process, increase citizen participation in eliciting project requirements and to enable a more transparent project monitoring and auditing. We apply the framework to a case study with respect to identifying on-chain activities that enable system interoperability, e-participation and project auditability.

International audience

Part 1: Model-Based Testing; International audience; Domain-specific Languages (DSLs) are widely used in model-based testing to make the benefits of modeling available to test engineers while avoiding the problem of excessive learning effort. Complex DSLs benefit from a formal definition of their semantics for model processing as well as consistency checking. A formal semantics can be established by mapping the model domain to a well-known formalism. In this paper, we present an industrial use case which includes a mapping from domain-specific models to Moore Machines, based on a Partial MAX-SAT problem, encoding a predicative semantics for the model-to-model mapping. We show how Partial MAX-SAT solves the frame problem for a non-trivial DSL in which the non-effect on variables cannot be determined statically. We evaluated the performance of our model-transformation algorithm based on models from our industrial use case.

International audience; Deep autoencoding Gaussian mixture model (DAGMM) employs dimensionality reduction and density estimation and jointly optimizes them for unsupervised anomaly detection tasks. However, the absence of large amount of training data greatly compromises DAGMM’s performance. Due to rising concerns for privacy, a worse situation can be expected. By aggregating only parameters from local training on clients for obtaining knowledge from more private data, federated learning is proposed to enhance model performance. Meanwhile, privacy is properly protected. Inspired by the aforementioned, this paper presents a federated deep autoencoding Gaussian mixture model (FDAGMM) to improve the disappointing performance of DAGMM caused by limited data amount. The superiority of our proposed FDAGMM is empirically demonstrated with extensive experiments.

Part 4: Business Processes; International audience; The Task-Technology Fit (TTF) theory provides a means of quantifying the effectiveness of technology in a system by assessing the relationship between the technology and the tasks the technology aims to support. The theory is widely recognized and has been applied in various ways, but little work has been done to summarize and synthesize the application of TTF in literature. The aim of this study is to identify and summarize the focus areas of studies that applied TTF, the environment in which it was applied, and the technologies which were considered by conducting a scoping review. It was found that applied studies focused primarily on generating theory or assessing certain real-world phenomena; was applied in a wide range of environments with the majority being in healthcare; and considered various technologies, with an increasing number of studies focusing on mobile technology. The findings of this study contribute to the understanding of TTF applications and assists in framing future research to further analyze TTF studies.

Part 1: Regular Papers; International audience; Spectrally-spatially flexible optical networks (SS-FONs) are proposed as a solution for future traffic requirements in optical backbone networks. As SS-FONs operate within flex-grid, the provisioning of lightpaths spanning multiple frequency slots results in spectrum fragmentation, especially in presence of dynamic traffic. Fragmentation, in turn, may lead to blocking of dynamic requests due to the lack of sufficiently-large free spectral windows. In this paper, to reach a better understanding of fragmentation in SS-FON, we extend several metrics used in (single-core) elastic optical networks to measure the fragmentation in SS-FONs. Next, we apply these metrics to a dynamic-routing algorithm with the goal of minimizing bandwidth blocking. Finally, we analyze the impact of spatial continuity constraint (SCC) on the network fragmentation. Simulations run on two representative network topologies show that the root mean square factor metric yields the best performance in terms of blocking when compared to other analyzed metrics.

Part 1: Regular Papers; International audience; The FASTER cable system has been developed as the first trans-pacific optical submarine cable system designed for digital-coherent transmission at the initial state. With this significant change for submarine cables, the design capacity is continuously being upgraded following the improvement of the state-of-the-art modulation format to maximize spectral efficiency even at the limited optical signal-to-noise ratio (OSNR). For the next generation, the novel technologies are expected to increase the capacity per cable under the conditions of limited feeding power and space in a cable. This report reviews some technologies from the current to the promising future submarine cable systems such as the introduction of space-division-multiplexing (SDM) technologies.

Part 1: Model-Based Testing; International audience; Chatbots, i.e., systems that communicate in natural language, have been of increasing importance over the last few years. These virtual agents provide specific services or products to clients on a 24/7 basis. Chatbots provide a simple and intuitive interface, i.e., natural language processing, which makes them increasingly attractive for various applications. In fact, chatbots are used as substitutes for repetitive tasks or user inquiries that can be automated. However, these advantages always are accompanied with concerns, e.g., whether security and privacy can be assured. These concerns become more and more important, because in contrast to simple requests, more sophisticated chatbots are able to utilize personalized services to users. In such cases, sensitive user data are processed and exchanged. Hence, such systems become natural targets for cyber-attacks with unforeseen consequences. For this reason, assuring information security of chatbots is an important challenge in practice. In this paper, we contribute to this challenge and introduce an automated security testing approach for chatbots. The presented framework is able to generate and run tests in order to detect intrinsic software weaknesses leading to the XSS vulnerability. We assume a vulnerability to be triggered when obtaining critical information from or crashing the virtual agent, regardless of its purpose. We discuss the underlying basic foundations and demonstrate the testing approach using several real-world chatbots.

Part 1: Regular Papers; International audience; Network resources are imperfect and vulnerable to failure from a wide variety of sources. Survivability is a well-researched field that focuses on strategies to prevent or reduce the harm inflicted when network elements fail. Solutions tend to either provision resources, such as backup paths, proactively so that traffic can be switched to the alternative route after a failure, or quickly find new resources to provision after failure event occurs. Current survivability solutions guarantee protection against these failures, but there is no mathematical model to calculate the network blocking probability for survivability solutions of anycast communication. In this paper, we developed new analytical models to calculate network-wide blocking performance for anycast survivability approaches. Performance results show that our models are accurate and are verified by extensive simulation results.

In this paper, we revisit proportional fair channel allocation in IEEE 802.11 networks. Traditional approaches are either based on the explicit solution of the optimization problem or use iterative solvers to converge to the optimum. Instead, we propose an algorithm able to learn the optimal slot transmission probability only by monitoring the throughput of the network. We have evaluated this algorithm both (i) using the true value of the function to optimize and (ii) considering estimation errors. We provide a comprehensive performance evaluation that includes assessing the sensitivity of the algorithm to different learning and network parameters as well as its reaction to network dynamics. We also evaluate the effect of noisy estimates on the convergence rate and propose a method to alleviate them. We believe our approach is a practical solution to improve the performance of wireless networks as it does not require knowing the network parameters in advance. Yet, we conclude that the setting of the parameters of the algorithm is crucial to guarantee fast convergence.

International audience; AI has turned into a focal piece of our life – as buyers, clients, and, ideally, as scientists and professionals! Regardless of whether we are applying prescient displaying systems to our examination or business issues, accept we make them thing in like manner: We need to make “great” forecasts! Fitting a model to our preparation information would one say one is a thing, however how would we realize that it sums up well to concealed information? How would we realize that it does not only retain the information we sustained it and neglects to make high forecasts on future examples, tests that it has not seen previously? Additionally, how would we select an appropriate model in any case? Perhaps an alternate learning calculation could be more qualified for the current issue? The right utilization of model assessment, model choice, and calculation choice systems is indispensable in scholarly AI examine just as in numerous mechanical settings. This article audits various systems that can be utilized for every one of these three subtasks and talks about the primary focal points and drawbacks of every method with references to theoretical and observational investigations. Further, suggestions are given to empower best yet plausible practices in research and uses of AI. In this article, we have used applications like Drowsiness detection, Oil price prediction, Election result evaluation as examples to explain algorithm selection and model evaluation.

Part 2: Security Testing; International audience; During this new world pandemic, a lot of technical issues have come to life, forcing all industries to reinvent themselves and how the day to day operations are performed. Remote working has become not only a trend but a necessity. Starting from online lecturing to online meetings on a need to know basis, people are concerned about their privacy and security in the online environment. Thus, in this paper, we focus our attention to the concepts of trust and reputation models in Wireless Mobile Networks (WMN), Wireless Sensor Networks (WSN) more accurately, for secure routing of large data packets with respect to Quality of Service (QoS). This paper analyses the current context in wireless mobile networks, more specifically in wireless sensor networks, and proposes a novel adaptive method to evaluate it by defining a Markov based trust function and a new model to manage it.

Part 2: Poster Papers; International audience; It is generally very difficult to optimize the routing policies in optical networks with dynamic traffic. Most widely-used routing policies, e.g., shortest path routing and least congested path (LCP) routing, are heuristic policies. Although the LCP is often regarded as the best-performing adaptive routing policy, we are often eager to know whether there exist better routing policies that surpass these heuristics in performance. In this paper, we propose a framework of reinforcement learning (RL) based routing scheme, that learns routing decisions during the interactions with the environment. With a proposed self-learning method, the RL agent can improve its routing policy continuously. Simulations on a ring-topology metro optical network demonstrate that, the proposed scheme outperforms the LCP routing policy.

Part 3: eBusiness; International audience; Social commerce is a rapidly growing platform of e-commerce that utilises social media and online social interaction to build brand awareness and increase sales. Buying and selling through social media can create a reliable and sustainable platform for buyers and vendors, offering an alternative platform to traditional online approaches. Research on social commerce began to achieve traction in 2006 and has grown since with a significant focus from academics who have offered new insight to many of the key topics. This study seeks to offer an additional contribution to the literature by analysing the predictors of consumer adoption of social commerce from existing studies by employing a weight analysis technique. The analysis considered seven dependent variables (along with their best and worst predictors) that are most frequently examined and are relevant to consumer adoption. The review presented in this study suggests that the intention to purchase is the most frequently examined dependent variable and that trust in the social commerce context is a key factor.

Part 5: Big Data and Machine Learning; International audience; The benefits of Big Data Analytics (BDA) are substantial in instances where organisations manage to successfully implement analytical capabilities. These benefits include improved, data driven decision-making, which can lead to deeper insight into business operations and as a result better performing organisations. Not surprisingly, an increased number of organisations are researching best implementation practices for BDA projects.Similar to software projects, research has shown that many BDA projects fail or do not deliver the business value as promised. To address this issue, the main objective of this research is to identify BDA implementation considerations for new BDA endeavors that will help organisations to align their BDA efforts with their overall business strategy to maximize business value.Based on a Benefit Dependency Network (BDN) model as main theoretical underpinning, a structured literature review was conducted focusing on investment objectives, business benefits, enabling changes and IT enablers when implementing BDA. A BDA implementation requires a holistic approach by considering aspects such as the skills of people which will have an impact on the structure of the organisation, business processes and technology changes to deliver benefits and investment objectives. Each of the domains of the BDN should be considered prior to BDA implementations.The research offers a guideline to organisations implementing BDA, based on the foundation of BDN.

Part 2: Poster Papers; International audience; Advanced interconnect technologies play a prominent role in scaling up the performance of data center networks. Spatial Division Multiplexing (SDM) photonic interconnects have been suggested as means to overcome capacity upgrade requirements and enable disaggregation in future data centers. Here, Holographic Optical Elements (HOE) are proposed for SDM photonic interconnects. The proposed coupling scheme entails the use of holograms as fan out components for coupling light from a source to multicore fibers (MCF). The scheme is versatile and can be adopted to any SDM fiber core arrangement. Appropriate Computer-Generated Holograms (CGHs) are designed for two kind of MCF and the HOE interconnect is evaluated in terms of performance variations against system and fabrication related parameters. Furthermore, the interconnect design is optimized for loss-sensitive applications.

Part 6: ICT and Education; International audience; It has been reported that WhatsApp, a social media application, had approximately 1.6 billion active users globally as of July 2019, almost one-fifth of the total world’s population. Thus, research about WhatsApp’s influence in general and especially its influence in education was relevant and significant. While there was much research involving WhatsApp and learning, it was not conclusive about the effects of WhatsApp on student learning. Specifically, research focusing on collaborative learning using WhatsApp was lacking, including research instruments for measuring collaboration on WhatsApp. Consequently, the paper’s research problem was the lack of research instruments for measuring collaboration on WhatsApp in relation to academic achievement. To address the research problem, the study followed the important initial and conceptual steps of the instrument development process to develop a research instrument to measure collaboration on WhatsApp in relation to academic achievement. The result of the paper was a developed instrument that provides researchers with a basis to measure the explanatory constructs involved in mobile collaborative learning (MCL) processes on WhatsApp and potentially other social media platforms. Therefore, the paper made an appropriately theoretical contribution, which was grounded in the scientific literature. The study facilitated positivistic research and epistemology for acquiring objective and precise scientific knowledge. Such deductive research promotes theory testing and development and presents educators and students with scientific evidence about learning with MCL applications such as WhatsApp from which both curriculum and learning design can be informed and benefited. In the age of connected mobility this is a necessity.

Part 2: Poster Papers; International audience; In this paper, we propose a stochastic failure and optical network topology (SFONT) model that can be used to comprehensively analyze the resilience of optical networks against a large number of possible earthquakes. We study an optical network densification problem, where dense network topologies are generated using the proposed SFONT model. Further, a seismic-risk aware optical network densification (SRA-OND) scheme is proposed with a view to design the future optical networks robust against earthquakes. The proposed SFONT model has been evaluated at various stages of network densification. To validate the capability of the proposed SFONT model to emulate real-world networking and failure scenario, we also perform a similar analysis based on RailTel optical network topology, seismic hazard maps, and real past earthquake data from India. Simulations indicate that the proposed SFONT model can be used to estimate and analyze the impact of network-resilience schemes on optical networks for a large number of possible earthquakes.

International audience; Although Random Forests (RFs) are an effective and scalable ensemble machine learning approach, they are highly dependent on the discriminative ability of the available individual features. Since most data mining problems occur in the context of pre-existing data, there is little room to choose the original input features. Individual RF decision trees follow a greedy algorithm that iteratively selects the feature with the highest potential for achieving subsample purity. Common heuristics for ranking this potential include the gini-index and information gain metrics. This study seeks to improve the effectiveness of RFs through an adapted gini-index splitting function and a feature engineering technique. Using a structured framework for comparative evaluation of RFs, the study demonstrates that the effectiveness of the proposed methods is comparable with conventional gini-index based RFs. Improvements in the minimum accuracy recorded over some UCI data sets, demonstrate the potential for a hybrid set of splitting functions.

Part 6: ICT and Education; International audience; Despite the growing importance of digitization in all aspects of teaching and learning, digital moderation (eModeration) has received little attention in research or practice. Considering the secondary school environment, no evidence-based requirements could be found for the development of a digital moderation system. This finding provides the rationale for this investigation into the requirements for an efficient eModeration system for IT and CAT assessments at grade 12 level in South Africa (SA). This study draws on eModeration literature as well as technology adoption literature to identify a set of pre-adoption eModeration requirements against the background of the challenges and benefits of eModeration. The research design involved a single exploratory case study with IT and CAT moderators as units of analysis. Qualitative data analysis using Atlas.ti V8 was conducted on the 61 responses received from an online survey and 4 responses received from a focus group interview. The findings confirmed the absence of customized eModeration systems. The main contribution is the eModeration requirements identified from literature and then refined by triangulation with the empirical findings of the survey. The research has practical value in guiding the design of eModeration systems for the school environment.

Fault-tolerance is a core feature in distributed database systems, particularly the ones deployed in cloud environments. The dependability of these systems often relies in middleware components that abstract the DBMS logic from the replication itself. The highly configurable nature of these systems makes their throughput very dependent on the correct tuning for a given workload. Given the high complexity involved, machine learning techniques are often considered to guide the tuning process and decompose the relations established between tuning variables. This paper presents a machine learning mechanism based on reinforcement learning that attaches to a hybrid replication middleware connected to a DBMS to dynamically live-tune the configuration of the middleware according to the workload being processed. Along with the vision for the system, we present a study conducted over a prototype of the self-tuned replication middleware, showcasing the achieved performance improvements and showing that we were able to achieve an improvement of 370.99% on some of the considered metrics., The research leading to these results has received funding from the European Union’s Horizon 2020 - The EU Framework Programme for Research and Innovation 2014–2020, under grant agreement No. 731218.

The purpose of the paper is to present the first self-stabilizing algorithm for finding \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n-1$$\end{document} one-to-many node-disjoint paths in message passing model. Two paths in a network are said to be node disjoint if they do not share any nodes except for the endpoints. Our proposed algorithm works on n-dimensional star networks \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$S_n$$\end{document}. Given a source node s and a set of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$D = \{d_1, d_2, ...,d_{n-1} \}$$\end{document} of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n-1$$\end{document} destination nodes in the n-dimensional star network, our algorithm constructs \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n-1$$\end{document} node-disjoints paths \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_1, P_2,...,P_{n-1}$$\end{document}, where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_i$$\end{document} is a path from s to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$d_i$$\end{document}, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1 \le i \le n-1$$\end{document}. Since the proposed solution is self-stabilizing [7], it does not require initialization and withstands transient faults. The stabilization time of our algorithm is \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O(n^2)$$\end{document} rounds.

International audience; Due to the lack of adequate public datasets, the proponents of many existing cloud intrusion detection systems (IDS) have relied on the DARPA dataset to design and evaluate their models. In the current paper, we show empirically that the DARPA dataset by failing to meet important statistical characteristics of real world cloud traffic data center is inadequate for evaluating cloud IDS. We present, as alternative, a new public dataset collected through a cooperation between our lab and a non-profit cloud service provider, which contains benign data and a wide variety of attack data. We present a new hypervisor-based cloud IDS using instance-oriented feature model and supervised machine learning techniques. We investigate 3 different classifiers: Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM) algorithms. Experimental evaluation on a diversified dataset yields a detection rate of 92.08% and a false positive rate of 1.49% for random forest, the best performing of the three classifiers.

Part 1: Regular Papers; International audience; In optical transport networks, the Quality of Transmission (QoT) using a physical layer model (PLM) is estimated before establishing new or reconfiguring established optical connections. Traditionally, high margins are added to account for the model’s inaccuracy and the uncertainty in the current and evolving physical layer conditions, targeting uninterrupted operation for several years, until the end-of-life (EOL). Reducing the margins increases network efficiency but requires accurate QoT estimation. We present two machine learning (ML) assisted QoT estimators that leverage monitoring data of existing connections to understand the actual physical layer conditions and achieve high estimation accuracy. We then quantify the benefits of planning/upgrading a network over multiple periods with accurate QoT estimation as opposed to planning with EOL margins.

Part 2: Poster Papers; International audience; Machine learning techniques for optimization purposes in the optical domain have been reviewed extensively in recent years. While several studies are pointing in the right direction towards building enhanced transport network control systems including estimation algorithms, the physical effects encountered in the optical domain raise several challenges that are hard to learn from and mitigate. In this paper, we provide a performance analysis of various supervised learning algorithms when predicting the Quality of Transmission (QoT), in terms of signal to noise ratio (OSNR), of lightpaths when erbium doped fiber amplifier (EDFA) power excursions and fiber nonlinearities are taken into account. The analysis considers F1-scores and computational training times as the main comparison metrics. A customized optical data network simulator was used for the generation of synthetic labeled data samples. Our results depict similar performance among groups of classifiers, and a correlation between the data sample size and the prediction accuracy.

Part 3: eBusiness; International audience; The characteristics of digital entrepreneurship and digital transformation and how they are related, is complex and important to understand in this digital age. Such an understanding of digital entrepreneurship is perceived as a key pillar for economic growth, job creation and innovation. However, a number of issues regarding digital entrepreneurship and digital transformation are prevalent, inhibiting digital entrepreneurs to optimise the advantages that digital entrepreneurship contributes towards business value. Therefore, the aim of this research paper is to investigate digital entrepreneurship and digital transformation, their characteristics and inter-relationships. Data extracted and analysed through a structured analysis process, recognises and discusses the characteristics of digital entrepreneurship and digital transformation. The characteristics are reported by employing the Dynamic Capabilities Theory as the structure. By understanding the characteristics of digital entrepreneurship and digital transformation, individuals and organisations may either create new business ventures or transform existing businesses through the development of novel digital technologies or the innovative application of such technologies.

Part 1: Regular Papers; International audience; The transmission reach of multi-core fiber (MCF) communication systems is severely affected by inter-core crosstalk (IC-XT), which limits its application for long-reach core optical network. One of the major factors limiting the transmission reach of MCF is nonlinear IC-XT interference, which makes the overall system nonlinear, thereby resulting in a poor bit error rate (BER) performance. Conventional Volterra series based nonlinear equalizer are computationally complex, and impaired by modeling error due to the truncation of polynomial kernel. In this paper, for the first time, we propose multivariate kernel least mean square (KLMS) based adaptive nonlinear equalizer for mitigating IC-XT impairments in MCF communication systems. The proposed scheme is inspired from reproducing kernel Hilbert space (RKHS) based machine learning algorithms. Simulations are performed for different multi-core structures, fiber lengths, and modulation schemes, which show that the proposed KLMS algorithm exhibit superior BER performance over the existing Volterra series equalizer.

International audience; The Domain Name System (DNS) is today a fundamental part of Internet’s working. Considering that Internet has grown in the last decades as part of human’s culture, user patterns regarding their behavior are present in the network data. As a consequence, some of these human behavior patterns are present as well in DNS data. With real data from the ‘.cl’ ccTLD, this work seeks to detect those human patterns by using Machine Learning techniques. As DNS traffic is described by a time series, particular and complex techniques have to be used in order to process the data and extract this information. The procedure that we apply in order to achieve this goal is divided in two stages. The first one consists of using clustering to group DNS domains basing on the similarity between their users’ activity. The second stage establishes a comparison between the obtained groups by using Association Rules. Finding human patterns in the data could be of high interest to researchers that analyze the human behavior regarding Internet’s usage. The procedure was able to detect some trends and patterns in the data that are discussed along with proper evaluation measures for further comparison.

Part 3: eBusiness; International audience; While online price comparison websites have burgeoned, there is scant understanding of how they influence online consumer behavior. This study addresses this gap in the literature by investigating the influence of price comparison websites on online switching behavior, and also suggests some additional factors that may be considered when looking at this relationship. We argue that shopper innovativeness, their perceived usefulness of the ad, and their customer service experience consciousness are important factors to consider when evaluating the impact of price comparison websites on eSwitching behavior. We also argue that the most appropriate theoretical lens through which to investigate this relationship is that of the consumer empowerment paradigm. A conceptual model is proposed and tested. Our analysis of 345 sample respondents finds that perceived usefulness of ads and customer service experience expectations are important enablers to price comparison websites use. Similarly, we find that shopper innovativeness and customer service expectations, in addition to price comparison websites use, are significant enablers to eSwitching behaviour. However, contrary to prediction, we find that shopper innovativeness has little to no influence on shoppers’ use of price comparison websites; we also observed similar patterns concerning the link between the perceived usefulness of online ads and eSwitching behaviour. In conclusion, our research contributes to better understanding the influence of price comparison websites on online switching behavior, and the factors that might influence this relationship.

Part 2: Fourth Industrial Revolution; International audience; Smart city initiatives are widely becoming part of the world agenda to address crises and to identify new initiatives for countries to manage resources while providing better living conditions for all citizens. The purpose of this study was to design a model to support the alignment of value in Smart City initiatives. To address this purpose, a systematic literature review (SLR) was conducted to find what Smart City initiatives have been addressed in empirical studies, and what dimensions and factors are linked to these initiatives. The SLR also identified the stakeholders in a Smart City, and what their roles should be linked to these initiatives. Six theories were identified and used to undergird the researcher’s understanding of the domains of Smart Cities, value and alignment. The concepts from these theories were then used with the SLR findings to design a conceptual model for Smart City initiatives. The proposed Value Alignment Smart City Model (VASC) can be used to plan or assess Smart City initiatives. The main contribution is the alignment of value amongst stakeholders to support the success of such initiatives. Further research is required to investigate adopting the model and empirically evaluate it.

Part 1: Regular Papers; International audience; We review the most important WDM DCI system innovations. State-of-the-art coherent transmission has already exceeded 6 b/s/Hz, using subcarrier modulation. The adoption of software innovations in automation and programmability, that DCI pioneered in transport networks, has also simplified operations and enables the emergence of “open” transport architectures. Combining these advancements with emerging network analytics frameworks allows exciting innovations in network design and management optimization.

Part 1: Regular Papers; International audience; Parallel computing allows us to process incredible amounts of data in a timely manner by distributing the workload across multiple nodes and executing computation simultaneously. However, the performance of this parallelism usually suffers from network bottleneck. In optical switching enabled data center networks (DCNs), to satisfy the complex and time-varying bandwidth demands from the parallel computing, it is critical to fully exploit the flexibility of optical networks and meanwhile reasonably schedule the optical resources. Considering that the traffic flows generated by different applications in DCNs usually exhibit different statistical or correlative features, it is promising to schedule the optical resources with the assistance of machine learning. In this paper, we introduce a framework called intelligent optical resources scheduling system, and discuss how this framework can assist resource scheduling based on machine learning approaches. We also present our recent simulation results to verify the performance of the framework.

Part 3: eBusiness; International audience; Cybersecurity readiness is a challenging issue for online retail businesses which are losing billions of dollars due to cyber-crimes and a lack of readiness to manage these. Therefore, research into cybersecurity readiness in the online retail industry is needed. Technical tools are the foremost measures of defence against these attacks. This study investigates cybersecurity readiness from the technical perspective in some UK online retailers. This research adopted a qualitative case study approach with semi-structured interviews for collecting data. A total of 15 interviews were conducted with an online retail company’s staff and management who had responsibility for managing cybersecurity. A thematic analysis method was used to analyse the qualitative data. The research findings show that the company is facing internal and external threats to their information systems and their technical defences are not very effective at present. The company should consider investing more resources in the technical controls to prevent these attacks.

Part 2: Fourth Industrial Revolution; International audience; Rapid development of advanced technologies and their use bear crucial influence on Smart City development. Smart economy ranks among key Smart City components. The rising awareness of the importance of Economy in Smart cities becomes a widely discussed issue and gets reflected in professional literature. The aim of the paper is to map and analyse the state of usage of current topics and terms “SMART city” and “economy” and their bibliographic coupling and co-occurrence on Web of Science. The analysis focuses on published documents and their citation, use of journals for publishing of articles, authors that have the most articles, number of published documents in countries. The reoccurrence of the most often used key words in the articles will be analysed. Analysis was done in March 2019 via VOSviewer software. The highest number of articles is presented in Smart economy in smart cities, Journal of cleaner production and Cities. USA and India are the countries with highest number of articles. Smart city, Cities, Smart cities, Governance, Innovation, Big data, Economy and Urbanization are the most often used keywords.

Part 5: Big Data and Machine Learning; International audience; Healthcare fraud is considered a challenge for many societies. Health care funding that could be spent on medicine, care for the elderly or emergency room visits are instead lost to fraudulent activities by materialistic practitioners or patients. With rising healthcare costs, healthcare fraud is a major contributor to these increasing healthcare costs. This study evaluates previous anomaly detection machine learning models and proposes an unsupervised framework to identify anomalies using a Generative Adversarial Network (GANs) model. The GANs anomaly detection (GAN-AD) model was applied on two different healthcare provider data sets. The anomalous healthcare providers were further analysed through the application of classification models with the logistic regression and extreme gradient boosting models showing good performance. Results from the SHapley Additive exPlanation (SHAP) also signifies that the predictors used explain the anomalous healthcare providers.

Part 4: Business Processes; International audience; Enterprise designers need to continuously re-design their enterprise, re-evaluating the technologies that are available to digitize their operations. Although light-weight agile software development approaches are favored by software development service providers, additional requirements elicitation practices should be incorporated when scaling factors apply, since design team members need to have a shared understanding of the operating context and high-level requirements. Research indicated that the organization construction diagram (OCD) could be useful to create a shared context for enterprise operation, linking detailed functional requirements to this shared context during software development. Although the OCD is concise, its associated concepts are abstract and an additional story-card method (SCM) is needed to transform existing enterprise implementations into an OCD. Since additional evaluation of the SCM was required, this study focused on a real-world demonstration of the SCM at a Fintech company where an agile software development approach is applied. The results indicate that the SCM is useful when incorporated within an agile software development approach.