This paper aims to solve the path following problem for an underactuated unmanned-surface-vessel (USV) based on deep reinforcement learning (DRL). A smoothly-convergent DRL (SCDRL) method is proposed based on the deep Q network (DQN) and reinforcement learning. In this new method, an improved DQN structure was developed as a decision-making network to reduce the complexity of the control law for the path following of a three-degree of freedom USV model. An exploring function was proposed based on the adaptive gradient descent to extract the training knowledge for the DQN from the empirical data. In addition, a new reward function was designed to evaluate the output decisions of the DQN, and hence, to reinforce the decision-making network in controlling the USV path following. Numerical simulations were conducted to evaluate the performance of the proposed method. The analysis results demonstrate that the proposed SCDRL converges more smoothly than the traditional deep Q learning while the path following error of the SCDRL is comparable to existing methods. Thanks to good usability and generality of the proposed method for USV path following, it can be applied to practical applications.

This paper addresses the co-design problem of a fault detection filter and controller for a networked-based unmanned surface vehicle (USV) system subject to communication delays, external disturbance, faults, and aperiodic denial-of-service (DoS) jamming attacks. First, an event-triggering communication scheme is proposed to enhance the efficiency of network resource utilization while counteracting the impact of aperiodic DoS attacks on the USV control system performance. Second, an event-based switched USV control system is presented to account for the simultaneous presence of communication delays, disturbance, faults, and DoS jamming attacks. Third, by using the piecewise Lyapunov functional (PLF) approach, criteria for exponential stability analysis and co-design of a desired observer-based fault detection filter and an event-triggered controller are derived and expressed in terms of linear matrix inequalities (LMIs). Finally, the simulation results verify the effectiveness of the proposed co-design method. The results show that this method not only ensures the safe and stable operation of the USV but also reduces the amount of data transmissions.

The issue of using vehicles to wirelessly recharge nodes for energy supplement in Wireless Sensor Networks has become a research hotspot in recent works. Unfortunately, most of the researches did not consider the rationality of the recharging request threshold and also overlooked the difference of node’s power consumption, which may lead to premature death of nodes as well as low efficiency of Wireless Charging Vehicles(WCVs). In order to solve the above problems, a Periodic and Distributed Energy Supplement Method based on maximum recharging benefit (PDESM) is proposed in this paper. Firstly, to avoid frequent recharging requests from nodes, we put forward an annuluses based cost-balanced data uploading strategy under deterministic deployment. Then, one WCV in each annulus periodically selects and recharges nodes located in this region which send the energy supplement requests. In addition, the predicted value of power consumption of nodes are calculated out according to the real-time energy consumption rate, and thus the most appropriate recharging request threshold is obtained. Finally, a moving path optimization scheme based on Minimum Spanning Tree is constructed for distributed recharging. Simulation results show that, PDESM performs well on enhancing the proportion of the alive nodes as well as the wireless recharging efficiency compared with NFAOC and FCFS. Moreover, it also has advantage in balancing the energy consumption of WCVs.

The growth and presence of heterogeneous sensor-equipped Internet-connected devices inside the home can increase efficiency and quality of life for the residents. Simultaneously, these devices continuously collect, process, and transmit data about the residents and their daily lifestyle activities to unknown parties outside the home. Such data can be sensitive and personal, leading to increasingly intimate insights into private lives. This data allows for the implementation of services, personalization support, and benefits offered by smart home technologies. Alas, there has been a surge of cyberattacks on connected home devices that essentially compromise privacy and security of the residents.

Providing privacy and security is a critical issue in smart connected homes. Many residents are concerned about unauthorized access into their homes and about the privacy of their data. However, it is typically challenging to implement privacy and security in a smart connected home because of its heterogeneity of devices, the dynamic nature of the home network, and the fact that it is always connected to the Internet, amongst other things. As the numbers and types of smart home devices are increasing rapidly, so are the risks with these devices. Concurrently, it is also becoming increasingly challenging to gain a deeper understand- ing of the smart home. Such understanding is necessary to build a more privacy-preserving and secure smart connected home. Likewise, it is needed as a precursor to perform a comprehensive privacy and security analysis of the smart home.

In this dissertation, we render a comprehensive description and account of the smart connected home that can be used for conducting risk analysis. In doing so, we organize the underlying smart home devices ac- cording to their functionality, identify their data-collecting capabilities, and survey the data types being collected by them. Such is done using the technical specification of commercial devices, including their privacy policies. This description is then leveraged for identifying threats and for analyzing risks present in smart connected homes. Such is done by analyzing both scholarly literature and examples from the industry, and leveraging formal modeling. Additionally, we identify malicious threat agents and mitigations that are relevant to smart connected homes. This is performed without limiting the research and results to a particular configuration and type of smart home.

This research led to three main findings. First, the majority of the surveyed commercial devices are collecting instances of sensitive and personal data but are prone to critical vulnerabilities. Second, there is a shortage of scientific models that capture the complexity and heterogeneity of real-world smart home deployments, especially those intended for privacy risk analysis. Finally, despite the increasing regulations and attention to privacy and security, there is a lack of proactive and integrative approaches intended to safeguard privacy and security of the residents. We contributed to addressing these three findings by developing a framework and models that enable early identification of threats, better planning for risk management scenarios, and mitigation of potential impacts caused by attacks before they reach the homes and compromise the lives of the residents.

Overall, the scientific contributions presented in this dissertation help deepen the understanding and reasoning about privacy and security concerns affecting smart connected homes, and contributes to advancing the research in the area of risk analysis as applied to such systems.

Using a face validity approach, this paper provides a validation of the Database Forensic Investigation Metamodel (DBFIM). The DBFIM was developed to solve interoperability, heterogeneity, complexity, and ambiguity in the database forensic investigation (DBFI) field, where severalmodels were identified, collected, and reviewed to develop DBFIM. However, the developedDBFIM lacked the face validity-based approach that could ensure DBFIM’s applicability in the DBFIfield. The completeness, usefulness, and logic of the developed DBFIM needed to be validated byexperts. Therefore, the objective of this paper is to perform the validation of the developed DBFIMusing the qualitative face validity approach. The face validity method is a common way of validating metamodels through subject expert inquiry on the domain application of the metamodel to assess whether the metamodel is reasonable and compatible based on the outcomes. For this purpose,six experts were nominated and selected to validate the developed DBFIM. From the expert review,the developed DBFIM was found to be complete, coherent, logical, scalable, interoperable, and useful for the DBFI field. 

Many Internet of Things (IoT) systems generate a massive amount of data needing to be processed and stored efficiently. Cloud computing solutions are often used to handle these tasks. However, the increasing availability of computational resources close to the edge has prompted the idea of using these for distributed computing and storage. Edge computing may help to improve IoT systems regarding important quality attributes like latency, energy consumption, privacy, and bandwidth utilization. However, deciding where to deploy the various application components is not a straightforward task. This is largely due to the trade-offs between the quality attributes relevant for the application. We have performed a systematic mapping study of 98 articles to investigate which quality attributes have been used in the literature for assessing IoT systems using edge computing. The analysis shows that time behavior and resource utilization are the most frequently used quality attributes; further, response time, turnaround time, and energy consumption are the most used metrics for quantifying these quality attributes. Moreover, simulation is the main tool used for the assessments, and the studied trade-offs are mainly between only two qualities. Finally, we identified a number of research gaps that need further study.

BACKGROUND: Telemedicine innovations are rarely adopted into routine health care, the reasons for which are not well understood. Teleguidance, a promising service for remote surgical guidance during endoscopic retrograde cholangiopancreatography (ERCP) was due to be scaled up, but there were concerns that user attitudes might influence adoption.

OBJECTIVE: Our objective was to gain a deeper understanding of ERCP practitioners' attitudes toward teleguidance. These findings could inform the implementation process and future evaluations.

METHODS: We conducted semistructured interviews with ERCP staff about challenges during work and beliefs about teleguidance. Theoretical constructs from the technology acceptance model (TAM) guided the thematic analysis. Our findings became input to a 16-item questionnaire, investigating surgeons' beliefs about teleguidance's contribution to performance and factors that might interact with implementation.

RESULTS: Results from 20 interviews with ERCP staff from 5 hospitals were used to adapt a TAM questionnaire, exchanging the standard "Ease of Use" items for "Compatibility and Implementation Climate." In total, 23 ERCP specialists from 15 ERCP clinics responded to the questionnaire: 9 novices (<500 ERCP procedures) and 14 experts (>500 ERCP procedures). The average agreement ratings for usefulness items were 64% (~9/14) among experts and 75% (~7/9) among novices. The average agreement ratings for compatibility items were somewhat lower (experts 64% [~9/14], novices 69% [~6/9]). The averages have been calculated from the sum of several items and therefore, they only approximate the actual values. While 11 of the 14 experts (79%) and 8 of the 9 novices (89%) agreed that teleguidance could improve overall quality and patient safety during ERCP procedures, only 8 of the 14 experts (57%) and 6 of the 9 novices (67%) agreed that teleguidance would not create new patient safety risks. Only 5 of the 14 experts (36%) and 3 of the 9 novices (33%) were convinced that video and image transmission would function well. Similarly, only 6 of the 14 experts (43%) and 6 of the 9 novices (67%) agreed that administration would work smoothly. There were no statistically significant differences between the experts and novices on any of the 16 items (P<.05).

CONCLUSIONS: Both novices and experts in ERCP procedures had concerns that teleguidance might disrupt existing work practices. However, novices were generally more positive toward teleguidance than experts, especially with regard to the possibility of developing technical skills and work practices. While newly trained specialists were the main target for teleguidance, the experts were also intended users. As experts are more likely to be key decision makers, their attitudes may have a greater relative impact on adoption. We present suggestions to address these concerns. We conclude that using the TAM as a conceptual framework can support user-centered inquiry into telemedicine design and implementation by connecting qualitative findings to well-known analytical themes.

For the third time in a row, HICSS is hosting the minitrack Making Digital Transformation Real. We received six submissions from 16 authors. Mainly, authors are affiliated with European research institutions. Also, we received submissions from authors affiliated with South African and Canadian research institutions. This year, with the COVID-19 pandemic, we received fewer submissions than expected. Topic- wise, the focus is clearly on the development of frameworks and the setting-up of propositions. We use a framework by Vial to classify the three selected submissions. Through the lens of the framework, there is an orientation towards research on structure and value creation. For the future, there is the need to develop more specific hypotheses for testing. Research on other areas of Vial’s framework, such as strategy and impact, is needed.

The phenomenon of network vigilantism is autonomously attributed to how anomalies and obscure activities from adversaries can be tracked in real-time. Needless to say, in today’s dynamic, virtualized, and complex network environments, it has become undeniably necessary for network administrators, analysts as well as engineers to practice network vigilantism, on traffic as well as other network events in real-time. The reason is to understand the exact security posture of an organization’s network environment at any given time. This is driven by the fact that modern network environments do, not only present new opportunities to organizations but also a different set of new and complex cybersecurity challenges that need to be resolved daily. The growing size, scope, complexity, and volume of networked devices in our modern network environments also makes it hard even for the most experienced network administrators to independently provide the breadth and depth of knowledge needed to oversee or diagnose complex network problems. Besides, with the growing number of Cyber Security Threats (CSTs) in the world today, many organisations have been forced to change the way they plan, develop and implement cybersecurity strategies as a way to reinforce their ability to respond to cybersecurity incidents. This paper, therefore, examines the relevance of Real-Time Monitoring (RTM) as a supplementary security component of vigilantism in modern network environments, more especially for proper planning, preparedness, and mitigation in case of a cybersecurity incident. Additionally, this paper also investigates some of the key issues and challenges surrounding the implementation of RTM for security vigilantism in our modern network environments.

The rapidly evolving Internet of Things (IoT) systems demands addressing new requirements. This particularly needs efficient deployment of IoT systems to meet the quality requirements such as latency, energy consumption, privacy, and bandwidth utilization. The increasing availability of computational resources close to the edge has prompted the idea of using these for distributed computing and storage, known as edge computing. Edge computing may help and complement cloud computing to facilitate deployment of IoT systems and improve their quality. However, deciding where to deploy the various application components is not a straightforward task, and IoT system designer should be supported for the decision.

To support the designers, in this thesis we focused on the system qualities, and aimed for three main contributions. First, by reviewing the literature, we identified the relevant and most used qualities and metrics. Moreover, to analyse how computer simulation can be used as a supporting tool, we investigated the edge computing simulators, and in particular the metrics they provide for modeling and analyzing IoT systems in edge computing. Finally, we introduced a method to represent how multiple qualities can be considered in the decision. In particular, we considered distributing Deep Neural Network layers as a use case and raked the deployment options by measuring the relevant metrics via simulation.

Care robots are often portrayed as an exciting new technology for improving care practices. Whether these robots will be accepted and integrated into care work or not, is likely to be affected by the assumptions, expectations and understandings held by potential end users, such as frontline staff and the people that are cared for. This paper describes how the conceptual framework of technological frames was used to identify the nature of care robots, care robots in use and care robot strategy as shared group level assumptions, expectations and understandings of care robots among care staff and potential care receivers. Focus groups were conducted with 94 participants. These groups consisted of line managers, frontline care staff, older people and students training to become carers. The technological frame of the nature of care robots revealed two complementary components: care robots as a threat to the quality of care, and care robots as substitute for humans and human care, held together by imaginaries of care robots. The technological frame of care robots in use revealed aspects of prospective end-users' uncertainty of their ability to handle care robots, and their own perceived lack of competence and knowledge about care robots. In addition, the following potential criteria for successful use of care robots were identified: adequate training, incentives for usage (needs and motives), usability, accessibility and finances. The technological frame of care robot strategy was revealed as believed cost savings and staff reduction. The novelty of the results, and their relevance for science and practice, is derived from the theoretical framework which indicates that adoption of care robots will be dependent on how well societies succeed in collectively shaping congruent technological frames among different stakeholders and aligning technological development accordingly.

With the rise of digital technologies, digital transformation (DT) has become an issue in the field of higher education. In higher education institutions and enterprises alike, DT means to digitalize internal processes and offer digital services and products. There are barriers that must be overcome to master this challenge. Our study follows an explorative mixed methods design. We identify the barriers to DT and transfer them to a research model. We examine the influence of individually perceived barriers on the DT process, thus contributing to the theoretical foundation of DT barriers in higher education institutions. This paper offers an approved scale to measure barriers to DT and a valid operationalization of DT barriers. The identified predictors can explain over 50% of the alteration problems of the DT process. Results indicate that management students have a significantly low level of concern regarding their privacy and traceability but demand more commitment at the organizational level.

We present new results on two types of guarding problems for polygons. For the first problem, we present an optimal linear time algorithm for computing a smallest set of points that guard a given shortest path in a simple polygon having 𝑛n edges. We also prove that in polygons with holes, there is a constant 𝑐>0c>0 such that no polynomial-time algorithm can solve the problem within an approximation factor of 𝑐log𝑛clogn, unless P=NP. For the second problem, we present a (𝑘+ℎ)(k+h)-FPT algorithm for computing a shortest tour that sees 𝑘k specified points in a polygon with ℎh holes. We also present a 𝑘k-FPT approximation algorithm for this problem having approximation factor 2‾√2. In addition, we prove that the general problem cannot be polynomially approximated better than by a factor of 𝑐log𝑛clogn, for some constant 𝑐>0c>0, unless P=NP.

The Internet of Things has become a central and exciting research area encompassing many fields in information and communication technologies and adjacent domains. IoT systems involve interactions with heterogeneous, distributed, and intelligent things, both from the digital and physical worlds including the human in the loop. Thanks to the increasingly wide spectrum of applications and cheap availability of both network connectivity and devices, a number of different stakeholders from industry, academia, society and government are part of the IoT ecosystem.

We seek to investigate possibilities of extending the emotional and cognitive experience of using products or services through the cross-modality of vision with other senses (synesthesia). Through multi-sensory, interactive environments, consumers become more engaged in the use of a product or service and may, in fact, participate as co-creators of their own experiences. To achieve the highest level of spontaneity and provoke human activation to enable us to study this, we suggest an experimental context based on interactive technologies, aesthetics and design. For this purpose, we use an artistic environment in the form of an interactive installation. Two examples of such experimental interactive art installations, Art Machine: Mindcatcher and Re-Digital, are described in this paper.

Data pipelines involve a complex chain of interconnected activities that starts with a data source and ends in a data sink. Data pipelines are important for data-driven organizations since a data pipeline can process data in multiple formats from distributed data sources with minimal human intervention, accelerate data life cycle activities, and enhance productivity in data-driven enterprises. However, there are challenges and opportunities in implementing data pipelines but practical industry experiences are seldom reported. The findings of this study are derived by conducting a qualitative multiple-case study and interviews with the representatives of three companies. The challenges include data quality issues, infrastructure maintenance problems, and organizational barriers. On the other hand, data pipelines are implemented to enable traceability, fault-tolerance, and reduce human errors through maximizing automation thereby producing high-quality data. Based on multiple-case study research with five use cases from three case companies, this paper identifies the key challenges and benefits associated with the implementation and use of data pipelines.

Labeling is a cornerstone of supervised machine learning. However, in industrial applications, data is often not labeled, which complicates using this data for machine learning. Although there are well-established labeling techniques such as crowdsourcing, active learning, and semi-supervised learning, these still do not provide accurate and reliable labels for every machine learning use case in the industry. In this context, the industry still relies heavily on manually annotating and labeling their data. This study investigates the challenges that companies experience when annotating and labeling their data. We performed a case study using a semi-structured interview with data scientists at two companies to explore their problems when labeling and annotating their data. This paper provides two contributions. We identify industry challenges in the labeling process, and then we propose mitigation strategies for these challenges.

The discussion of the digital divide and digital inclusion has extended to older adults. Although knowledge on the digital divide and digital inclusion among native older adults has increased substantially, little is known about the situations of older immigrants in relation to the digital divide. This paper employed the scoping review approach to map the situations and research methods of the digital divide among older immigrants from recent empirical studies. The initial search identified 997 articles, of which 13 articles were selected for this review. The results showed that socioeconomic status, language proficiency, degree of acculturation, level of education, and digital literacy are the most common factors leading to the disparities between native older adults and older immigrants. Although the results showed a narrowing gap as concerns access to the Internet, interventions are needed to reduce the divide among individuals of different ethnicities due to disparities in digital skills and knowledge. The included studies applied quantitative, qualitative, and mixed-method approaches. The homogeneity of the findings of some included studies implied the need to develop more methods and models to study the digital divide among older immigrants. This review suggested that future research incorporate ethnic characteristics in the research design to provide in-depth knowledge about the ethnic group. This knowledge could potentially be utilized for future interventions aimed at narrowing the remaining gap of the digital divide.

We do not yet know how the robotization of eldercare will unfold, but one thing is clear: technology mediates human practices and experiences [1]. As such, care robots will co-shape the actions of care givers and older people and influence the perceptions and experiences of old age. The robotization of eldercare means that it is essential for developers, policy makers, and researchers to become increasingly aware of the intertwined and implicit expectations that older people impose on care robots. This paper both zooms in towards older people’s individual expectations and zooms out towards expectation configurations at a group level and the expectation imagery of care robots in future eldercare.

Nowadays, machine learning (ML) is an integral component in a wide range of areas, including software analytics (SA) and business intelligence (BI). As a result, the interest in custom ML-based software analytics and business intelligence solutions is rising. In practice, however, such solutions often get stuck in a prototypical stage because setting up an infrastructure for deployment and maintenance is considered complex and time-consuming. For this reason, we aim at structuring the entire process and making it more transparent by deriving an end-to-end framework from existing literature for building and deploying ML-based software analytics and business intelligence solutions. The framework is structured in three iterative cycles representing different stages in a model’s lifecycle: prototyping, deployment, update. As a result, the framework specifically supports the transitions between these stages while also covering all important activities from data collection to retraining deployed ML models. To validate the applicability of the framework in practice, we compare it to and apply it in a real-world ML-based SA/BI solution.

Biometric solutions for access control is an active line of research. Specifically, when it comes to facial identification for access control, these systems can pose privacy concerns. For instance, identifying people that do not want to use the facial identification module. This work focuses on implementing an intent-aware system, which uses a hand gesture trigger to initiate the identification process. In order to evaluate the system, test cases were performed to verify accuracy of each hand gesture. Thereafter, a scenario was created to simulate an activation of the prototype system. The evaluation was used to determine the convenience and guidance when implementing intent-aware systems.  

Continuous experimentation (CE) refers to a group of practices used by software companies to rapidly assess the usage, value and performance of deployed software using data collected from customers and the deployed system. Despite its increasing popularity in the development of web-facing applications, CE has not been discussed in the development process of business-to-business (B2B) mission-critical systems.

We investigated in a case study the use of CE practices within several products, teams and areas inside Ericsson. By observing the CE practices of different teams, we were able to identify the key activities in four main areas and inductively derive an experimentation process, the HURRIER process, that addresses the deployment of experiments with customers in the B2B and with mission-critical systems. We illustrate this process with a case study in the development of a large mission-critical functionality in the Long Term Evolution (4G) product. In this case study, the HURRIER process is not only used to validate the value delivered by the solution but to increase the quality and the confidence from both the customers and the R&D organization in the deployed solution. Additionally, we discuss the challenges, opportunities and lessons learned from applying CE and the HURRIER process in B2B mission-critical systems.

The collection of high-quality data provides a key competitive advantage to companies in their decision-making process. It helps to understand customer behavior and enables the usage and deployment of new technologies based on machine learning. However, the process from collecting the data, to clean and process it to be used by data scientists and applications is often manual, non-optimized and error-prone. This increases the time that the data takes to deliver value for the business. To reduce this time companies are looking into automation and validation of the data processes. Data processes are the operational side of data analytic workflow.

DataOps, a recently coined term by data scientists, data analysts and data engineers refer to a general process aimed to shorten the end-to-end data analytic life-cycle time by introducing automation in the data collection, validation, and verification process. Despite its increasing popularity among practitioners, research on this topic has been limited and does not provide a clear definition for the term or how a data analytic process evolves from ad-hoc data collection to fully automated data analytics as envisioned by DataOps.

This research provides three main contributions. First, utilizing multi-vocal literature we provide a definition and a scope for the general process referred to as DataOps. Second, based on a case study with a large mobile telecommunication organization, we analyze how multiple data analytic teams evolve their infrastructure and processes towards DataOps. Also, we provide a stairway showing the different stages of the evolution process. With this evolution model, companies can identify the stage which they belong to and also, can try to move to the next stage by overcoming the challenges they encounter in the current stage.

As tabletop games are ported to digital versions to increase their accessibility, the expected User Experience (UX) might be degraded in the transition. This paper aims to understand how and why playing tabletop games differentiates depending on the platform. Seven tabletop games have been chosen from different genres with an official digital adaptation. Our approach has been to do a comparative analysis of both versions followed by a user study to analyze and measure the UX differences, measuring five key factors, Usability, Engagement, Social Connectivity, Aesthetics, and Enjoyment. Our results indicate that games that rely on imperfect information offer a much higher social connectivity and engagement when played around a table. Meanwhile, games relying on tile-placement offers higher usability and engagement when played digitally due to the assistance provided by the game. However, the physical versions got, in general, a higher rating than the digital versions in all key factors except slightly in the usability. Physical versions are the preferred options, but the digital versions' benefits, such as accessibility and in-game assistance, makes them relevant for further analysis.  

With the growth of procedural content generation in game development, there is a need for a viable generative method to give context and make sense of the content within game space. We propose procedural narrative as context through objectives, as a useful means to structure content in games. In this paper, we present and describe an artifact developed as a sub-system to the Evolutionary Dungeon Designer (EDD) that procedurally generates objectives for the dungeons created with the tool. The quality of the content within rooms is used to generate objectives, and together with the distributions and design of the dungeon, main and side objectives are formed to maximize the usage of game space and create a proper context.  

The relationship between negative and positive connotations with regard to malware in the cloud is rarely investigated according to the prevailing literature. However, there is a significant relationship between the use of positive and negative connotations. A clear distinction between the two emanates when we use the originally considered malicious code, for positive connotation like in the case of capturing keystrokes in a proactive forensic purpose. This is done during the collection of digital evidence for Digital Forensic Readiness (DFR) purposes, in preparation of a Digital Forensic Investigation (DFI) process. The paper explores the problem of having to use the keystrokes for positive reasons as a piece of potential evidence through extraction and digitally preserving it as highlighted in ISO/IEC 27037: 2012 (security approaches) and ISO/IEC 27043: 2015 (legal connotations). In this paper, therefore, the authors present a technique of how DFR can be achieved through the collection of digital information from the originally considered malicious code. This is achieved without modifying the cloud operations or the infrastructure thereof, while preserving the integrity of digital information and possibly maintain the chain of custody at the same time. The paper proposes that the threshold of malicious code intrusion in the cloud can be transformed to an efficacious process of DFR through logical acquisition and digitally preserving keystrokes. The experiment-tested keystrokes have shown a significant approach that could achieve proactive forensics.

The digital media landscape has been exposed in recent years to an increasing number of deliberately misleading news and disinformation campaigns, a phenomenon popularly referred as fake news. In an effort to combat the dissemination of fake news, designing machine learning models that can classify text as fake or not has become an active line of research. While new models are continuously being developed, the focus so far has mainly been aimed at improving the accuracy of the models for given datasets. Hence, there is little research done in the direction of explainability of the deep learning (DL) models constructed for the task of fake news detection.In order to add a level of explainability, several aspects have to be taken into consideration. For instance, the pre-processing phase, or the length and complexity of the text play an important role in achieving a successful classification. These aspects need to be considered in conjunction with the model's architecture. All of these issues are addressed and analyzed in this paper. Visualizations are further employed to grasp a better understanding how different models distribute their attention when classifying fake news texts. In addition, statistical data is gathered to deepen the analysis and to provide insights with respect to the model's interpretability.

Internet of Things (IoT) are becoming commonplace in homes, buildings, cities, and nations, and IoT networks are also getting more complex and interconnected. The complexity, interconnectivity, and heterogeneity of IoT systems, however, complicate digital (forensic) investigations. The challenge is compounded due to the lack of holistic and standardized approaches. Hence, building on the ISO/IEC 27043 international standard, we present a holistic digital forensic readiness (DFR) framework. We also qualitatively evaluate the utility of the proposed DFR framework.

The process of collecting traffic data is a key component to evaluate the current state of a transportation network and to analyze movements of vehicles. In this paper, we argue that both active stationary and mobile measurement devices should be taken into account for high-quality traffic data with sufficient geographic coverage. Stationary devices are able to collect data over time at certain locations in the network and mobile devices are able to gather data over large geographic regions. Hence, the two types of measurement devices have complementary properties and should be used in conjunction with each other in the data collection process. To evaluate the complementary characteristics of stationary and mobile devices for traffic data collection, we present a traffic simulation model, which we use to study the share of successfully identified vehicles when using both types of devices with varying identification rate. The results from our simulation study, using freight transport in southern Sweden, shows that the share of successfully identified vehicles can be significantly improved by using both stationary and mobile measurement devices. 

With the steady rise of home and building automation management system, it is becoming paramount to gain access to information that reflects consumption patterns with devicelevel granularity. Various application-level services can then makes use of this data for monitoring and controlling purposes in an efficient manner. In this paper we report on the design and development of an Internet of Things (IoT) end-to-end solution for electric appliance recognition that can operate in real-time and entails low hardware cost. For the task of identifying various appliance signatures we also provide a comparative analysis, where on the one hand, we investigate the suitability of several machine learning approaches given publicly available datasets, that generally provide months worth of data with a relatively low sampling frequency. On the other hand, we proceed to evaluate their discriminative effectiveness for our particular scenario, where the goal is to provide rapid identification of the appliance signature in real-time based on a reduced training dataset (few-shot learning). This is particularly important in the context of appliance recognition, where due to the high variance in consumption patterns within each class, in order to achieve high accuracy, data points often need to be collected for each individual appliance or device that would need to be later identified. Clearly, this data collection process is often expensive and difficult to perform, especially in large-scale settings, hence few-shot learning is key. Besides presenting our end-to-end IoT solution that meets the abovementioned desiderata, the paper also provides an analysis of the computational demand of such an approach with regard to cost and real-time performance, which is often critical to low-powered IoT solutions. (C) 2020 The Authors. Published by Elsevier B.V.

Whereas data labelling in IoT applications is costly, it is also time consuming to train a supervised Machine Learning (ML) algorithm. Hence, a human oracle is required to gradually annotate the data patterns at run-time to improve the models’ learning behavior, through an active learning strategy in form of User Feedback Process (UFP). Consequently, it is worth to note that during UFP there may exist malicious content that may subject the learning model to be vulnerable to adversarial attacks, more so, manipulative attacks. We argue in this position paper, that there are instances during incremental learning, where the local data model may present wrong output, if retraining is done using data that has already been subjected to adversarial attack. We propose a Distributed Interactive Secure Federated Learning (DISFL) framework that utilizes UFP in the edge and fog node, that subsequently increases the amount of labelled personal local data for the ML model during incremental training. Furthermore, the DISFL framework addresses data privacy by leveraging federated learning, where only the model's knowledge is moved to a global unit, herein referred to as Collective Intelligence Node (CIN). During incremental learning, this would then allow the creation of an immutable chain of data that has to be trained, which in its entirety is tamper-free while increasing trust between parties. With a degree of certainty, this approach counters adversarial manipulation during incremental learning in active learning context at the same time strengthens data privacy, while reducing the computation costs.

Data is the new currency and key to success. However, collecting high-quality data from multiple distributed sources requires much effort. In addition, there are several other challenges involved while transporting data from its source to the destination. Data pipelines are implemented in order to increase the overall efficiency of data-flow from the source to the destination since it is automated and reduces the human involvement which is required otherwise. Despite existing research on ETL (Extract-Transform-Load) and ELT (Extract-Load-Transform) pipelines, the research on this topic is limited. ETL/ELT pipelines are abstract representations of the end-to-end data pipelines. To utilize the full potential of the data pipeline, we should understand the activities in it and how they are connected in an end-to-end data pipeline. This study gives an overview of how to design a conceptual model of data pipeline which can be further used as a language of communication between different data teams. Furthermore, it can be used for automation of monitoring, fault detection, mitigation and alarming at different steps of data pipeline.

In order to distinguish themselves from their competitors, software service providers constantly try to assess and improve customer satisfaction. However, measuring customer satisfaction in a continuous way is often time and cost intensive, or requires effort on the customer side. Especially in B2B contexts, a continuous assessment of customer satisfaction is difficult to achieve due to potential restrictions and complex provider-customer-end user setups. While concepts such as web usage mining enable software providers to get a deep understanding of how their products are used, its application to quantitatively measure customer satisfaction has not yet been studied in greater detail. For that reason, our study aims at combining existing knowledge on customer satisfaction, web usage mining, and B2B service characteristics to derive a model that enables an automated calculation of quantitative customer satisfaction scores. We apply web usage mining to validate these scores and to compare the usage behavior of satisfied and dissatisfied customers. This approach is based on domain-specific service quality and web usage metrics and is, therefore, suitable for continuous measurements without requiring active customer participation. The applicability of the model is validated by instantiating it in a real-world B2B online platform.

Data pipelines play an important role throughout the data management process. It automates the steps ranging from data generation to data reception thereby reducing the human intervention. A failure or fault in a single step of a data pipeline has cascading effects that might result in hours of manual intervention and clean-up. Data pipeline failure due to faults at different stages of data pipelines is a common challenge that eventually leads to significant performance degradation of data-intensive systems. To ensure early detection of these faults and to increase the quality of the data products, continuous monitoring and fault detection mechanism should be included in the data pipeline. In this study, we have explored the need for incorporating automated fault detection mechanisms and mitigation strategies at different stages of the data pipeline. Further, we identified faults at different stages of the data pipeline and possible mitigation strategies that can be adopted for reducing the impact of data pipeline faults thereby improving the quality of data products. The idea of incorporating fault detection and mitigation strategies is validated by realizing a small part of the data pipeline using action research in the analytics team at a large software-intensive organization within the telecommunication domain.

Machine Learning (ML) and Artificial Intelligence (AI) have increasingly gained attention in research and industry. Federated Learning, as an approach to distributed learning, shows its potential with the increasing number of devices on the edge and the development of computing power. However, most of the current Federated Learning systems apply a single-server centralized architecture, which may cause several critical problems, such as the single-point of failure as well as scaling and performance problems. In this paper, we propose and compare four architecture alternatives for a Federated Learning system, i.e. centralized, hierarchical, regional and decentralized architectures. We conduct the study by using two well-known data sets and measuring several system performance metrics for all four alternatives. Our results suggest scenarios and use cases which are suitable for each alternative. In addition, we investigate the trade-off between communication latency, model evolution time and the model classification performance, which is crucial to applying the results into real-world industrial systems.

Even though the cloud paradigm and its associated services has been adopted in various enterprise applications, there has been major issues with regard to authenticating users' critical data. Single Sign on (SSO) is a user authentication technique through which a server authenticates and allows a user to use a single aspect of login credentials, for example, to access multiple services in the cloud. Even though SSO reduces the number of logins that are needed over heterogeneous environments, the risk that might be associated with the security of SSO might be detrimental if, for example, a Man-in-the Middle (MITM) attacker manages to gain control of the SSO credentials. It is also possible to get the identity of the users who have logged into Active Directory or intranet and this identity can easily be used to log into other web-based applications, and this requires the use of the Security Assertion Mark-up Language (SAML). SAML is basically a standard that allows users to be logged into applications as per their sessions. The problem that this paper addresses is the lack of a proactive technique of hardening cloud-based SAML while combining SSO with a Multi-Factor Authentication (MFA) at the time of writing this paper. The authors have, therefore, proposed an effective approach that unifies SSO with MFA in this context. Based on the base score index conducted over Common Vulnerability Scoring System (CVSS), the architecture proves to be reliable, feasible and with better performance.

Companies across domains are rapidly engaged in shifting computational power and intelligence from centralized cloud to fully decentralized edges to maximize value delivery, strengthen security and reduce latency. However, most companies have only recently started pursuing this opportunity and are therefore at the early stage of the cloud-to-edge transition. To provide an overview of AI deployment in the context of edge/cloud/hybrid architectures, we conduct a systematic literature review and a grey literature review. To advance understanding of how to integrate, deploy, operationalize and evolve AI models, we derive a framework from existing literature to accelerate the end-to-end deployment process. The framework is organized into five phases: Design, Integration, Deployment, Operation and Evolution. We make an attempt to analyze the extracted results by comparing and contrasting them to derive insights. The contribution of the paper is threefold. First, we conduct a systematic literature review in which we review the contemporary scientific literature and provide a detailed overview of the state-of-the-art of AI deployment. Second, we review the grey literature and present the state-of-practice and experience of practitioners while deploying AI models. Third, we present a framework derived from existing literature for the end-to-end deployment process and attempt to compare and contrast SLR and GLR results.

Machine learning and deep learning techniques are becoming increasingly popular and critical for companies as part of their systems. However, although the development and prototyping of ML/DL systems are common across companies, the transition from prototype to production-quality deployment models are challenging. One of the key challenges is how to determine the selection of an optimal architecture for AI deployment. Based on our previous research, and to offer support and guidance to practitioners, we developed a framework in which we present five architectural alternatives for AI deployment ranging from centralized to fully decentralized edge architectures. As part of our research, we validated the framework in software-intensive embedded system companies and identified key challenges they face when deploying ML/DL models. In this paper, and to further advance our research on this topic, we identify factors that help practitioners determine what architecture to select for the ML/D L model deployment. For this, we conducted a follow-up study involving interviews and workshops in seven case companies in the embedded systems domain. Based on our findings, we identify three key factors and develop a framework in which we outline how prioritization and trade-offs between these result in certain architecture. The contribution of the paper is threefold. First, we identify key factors critical for AI system deployment. Second, we present the architecture selection framework that explains how prioritization and trade-offs between key factors result in the selection of a certain architecture. Third, we discuss additional factors that may or may not influence the selection of an optimal architecture.

This paper addresses the problem of hazard recognition and risk assessment in open and non-predictive environments to support decision making and action selection. Decision making and action selection incorporate decreasing situational risks and maintain safety as operational constraints. Commonly, neither existing application-related safety standards nor the situation modeling or knowledge representation is considered in that context. This contribution introduces a novel approach denoted as a Safety-Driven Behavior Management focusing on situation modeling and the problem of knowledge representation in its sub-functions in the context of situational risks. It combines the safety standards-oriented hazard analysis and the risk assessment approach with the machine learning-based situation recognition. An example illustrating the approach is presented in this paper.

Deep neural networks for positioning can improve accuracy by adapting to inhomogeneous environments. However, they are still susceptible to noisy data, often resulting in invalid positions. A related task, map matching, can be used for reducing geographical invalid positions by aligning observations to a model of the real world. In this paper, we propose an approach for positioning, enhanced with map matching, within a single deep neural network model. We introduce a novel way of reducing the number of invalid position estimates by adding map information to the input of the model and using a map-based loss function. Evaluating on real-world Received Signal Strength Indicator data from an asset tracking application, we show that our approach gives both increased position accuracy and a decrease of one order of magnitude in the number of invalid positions.

The potential of IoT in contributing towards sustainable economic development in Sub-Saharan Africa (SSA) through digital transformation and effective service delivery is widely accepted. However, the unreliability/unavailability of connectivity and power grid infrastructure as well as the unaffordability of the overall system hinders the implementation of a multi-layered IoT architecture for rural societal services in SSA. In this work, affordable IoT architecture that operates without reliance on broadband connectivity and power grid is developed. The architecture employs energy harvesting system and performs data processing, actuation decisions and network management locally by integrating a customized low- cost computationally capable device with the gateway. The sharing of this device among the water resource and quality management, healthcare and agriculture applications further reduces the overall system cost. The evaluation of LPWAN technologies reveals that LoRaWAN has lower cost with added benefits of adaptive data rate and largest community support while providing comparable performance and communication range with the other technologies. The relevant results of the analysis is communicated to end-users’ mobile device via 2G/3G GPRS. Hence, the proposed IoT architecture enables the implementation of IoT systems for improving efficiency in three key application areas at low cost.

Deep neural networks have the ability to generalize beyond observed training data. However, for some applications they may produce output that apriori is known to be invalid. If prior knowledge of valid output regions is available, one way of imposing constraints on deep neural networks is by introducing these priors in a loss function. In this paper, we introduce a novel way of constraining neural network output by using encoded regions with a loss function based on gradient interpolation. We evaluate our method in a positioning task where a region map is used in order to reduce invalid position estimates. Results show that our approach is effective in decreasing invalid outputs for several geometrically complex environments.

We report and review a visualization system capable of displaying gaze and speech data elicited from pairs of subjects interacting in a discussion. We elicit such conversation data in our first experiment, where two participants are given the task of reaching a consensus about questions involving images. We validate the system in a second experiment where the purpose is to see if a person could determine which question had elicited a certain visualization. The visualization system allows users to explore reasoning behavior and participation during multimodal dialogue interactions.

Traffic flow estimates play a key role in traffic network management and planning of transportation networks. Commonly it is the average daily traffic (ADT) flow for different road segments that constitute the data. This paper shows how an advanced and detailed analysis based on hourly flow measurements over the day can contribute to a deeper understanding of how hourly flows together reflect the vehicles’ routes. The proposed method identifies the shortest travel time paths between all possible origins and destinations in a transportation network, and thereafter it identifies the most significant routes in the network by performing statistical tests. For this purpose, the paper presents a mathematical model, a vehicle simulator based on this model, and a statistical framework that is able to find the most probable underlying routes. The paper contains a real test scenario based on 24-hour traffic flows (hour by hour) to demonstrate the applicability of the method.

Data driven and experimental development practices provide effective means for companies to adopt a customer and market-centric way-of-working. In online companies, controlled experimentation is the primary technique to measure how customers respond to variants of deployed software. Over the recent years, and due to increasing connectivity and data collection from products in the field, these practices are being adopted also in software-intensive embedded systems companies. In these companies, experiments are run on selected instances of the system or as comparisons of previously computed data to ensure value delivery to customers, improve quality and explore new value propositions. However, to utilize the benefits of data- driven and experimental development practices, companies need to define what value factors to optimize for. For highly complex embedded systems with thousands of parameters, and with people at different levels in the organization having different opinions about the value of features, this is a challenging task. In this paper, we report on longitudinal multi-case study research in which we explore value modeling as a technique to help people in development, in product management and on the business level to align interests and agree on value factors. Based on this work, we identify five purposes of value modeling and how this technique helps accelerate critical activities in an organization. The contribution of this paper is three-fold. First, we provide empirical evidence for how value modeling is an effective technique to help companies define what to optimize for. Second, we identify five purposes of value modeling. Third, we identify the key challenges that the case companies experience when applying value modeling.

Fog computing is an architecture that is used to distribute resources such as computing, storage, and memory closer to end-user to improve applications and service deployment. The idea behind fog computing is to improve cloud computing and IoT infrastructures by reducing compute power, network bandwidth, and latency as well as storage requirements. This paper presents an overview of what fog computing is, related concepts, algorithms that are present to improve fog computing infrastructure as well as challenges that exist. This paper shows that there is a great advantage of using fog computing to support cloud and IoT systems.

Over the past 15 years, A/B testing has been a critical tool for accurate prioritization of development efforts in online and web-facing companies. As automotive companies progress on their digitalization process, A/B testing and other experimentation techniques start to be adopted. However, specific characteristics of the automotive software industry create additional challenges to the successful adoption of A/B testing. Recently, research has been conducted to investigate the challenges and opportunities for experimentation techniques in the automotive and more generally in the embedded systems domain. However, despite the collaboration with industry, previous research was based on either hypothesized or toy scenarios in companies seeking, but not yet running experimentation. Utilizing a case study method, we investigate the challenges of adopting A/B testing in two large-scale automotive companies that are currently running or preparing for their first A/B testing. The contribution of this paper is two-fold. First, we present our main findings in terms of the challenges of real A/B testing iterations in automotive vehicles. Second, we present the current, potential solutions and lessons learned from applying A/B testing in the automotive domain.

In recent years, the application of artificial intelligence (AI) has become an integral part of a wide range of areas, including software engineering. By analyzing various data sources generated in software engineering, it can provide valuable insights into customer behavior, product performance, bugs and errors, and many more. In practice, however, AI for software analytics and business intelligence often gets stuck in a prototypical stage and the results are rarely used to make decisions based on data. To understand the underlying root causes of this phenomenon, we conduct both an explanatory case study and a survey on the challenges of realizing and utilizing artificial intelligence in the context of software-intensive businesses. As a result, we identify a vicious circle that prevents practitioners from moving from prototypical analytics to continuous and productively usable software analytics and business intelligence based on AI.