The papers in this special section focus on hybrid human-artificial intelligene (AI) for multimedia computing. Multimedia computing has experienced a tremendous growth in the last decades, with applications ranging from multimedia information retrieval and analysis to multimedia compression and communication. However, the increasing volume and complexity of multimedia data driven by the large-scale spread of various new devices and sensors is posing a serious challenge to traditional multimedia computing algorithms. Artificial intelligence (AI), in particular deep learning techniques, has improved the performance of multimedia computing algorithms for many tasks, including computer vision and natural language processing. But unlike humans, AI is poor at solving tasks across multiple domains or in dealing with an uncontrolled dynamic environment. Hybrid Human-Artificial Intelligence (HH-AI) is an emerging field that aims at combining the benefits of human intelligence, such as semantic association, inference, and generalization with the computing power of AI.

Digital transformation (DT) has become a buzzword, triggering different disciplines in research and influencing practice, which leads to independent research streams. Scholars investigate the antecedents, contingencies, and consequences of these disruptive technologies by examining the use of single technologies or of digitization, in general. Approaches are often very specialized and restricted to their domains. Thus, the immense breadth of technologies and their possible applications conditions a fragmentation of research, impeding a holistic view. With this systematic literature review, we aim to fill this gap in providing an overview of the different disciplines of DT research from a holistic business perspective. We identified the major research streams and clustered them with co-citation network analysis in nine main areas. Our research shows the main fields of interest in digital transformation research, overlaps of the research areas and fields that are still underrepresented. Within the business research areas, we identified three dominant areas in literature: finance, marketing, and innovation management. However, research streams also arise in terms of single branches like manufacturing or tourism. This study highlights these diverse research streams with the aim of deepening the understanding of digital transformation in research. Yet, research on DT still lacks in the areas of accounting, human resource management, and sustainability. The findings were distilled into a framework of the nine main areas for assisting the implications on potential research gaps on DT from a business perspective.

This chapter discusses the digital reading condition and multisensory reading from the perspective of journalism, which implies other implications and dimensions than educational or literary reading. Global as well as local news organizations around the world are experimenting with new digital opportunities for presence and engagement of users beyond the written word, and the chapter gives some examples. Framed by the term “immersive journalism,” the chapter presents examples of the use of immersive technologies like 3D models, Augmented and Virtual Reality, and 360° photography and videography. It also suggests to include digital audio formats as examples of journalistic products that are able to create different kinds of sensory and social experiences of presence.

The success of neural networks largely builds on their ability to create internal knowledge representations from real-world high-dimensional data, such as images, sound, or text. Approaches to extract and present these representations, in order to explain a neural network’s decision, is an active and multifaceted research field. To gain a deeper understanding of a central aspect of this field, we performed a targeted literature review focusing on research that aims to associate internal representations with human understandable concepts. By using deductive nomological explanations combined with causality theories as an analytical lens, we analyse nine carefully selected research papers. We find our analytical lens, the explanation structure and causality, useful to understand what can be expected, and not expected, from explanations inferred from neural networks. The analysis additionally uncovers an ambiguity in the reviewed literature related to the goal: is it (a) understanding the ML model, (b) the training data or (c) actionable explanations that are true-to-the-domain?

The currently dominating artificial intelligence and machine learning technology, neural networks, builds on inductive statistical learning processes. Being void of knowledge that can be used deductively these systems cannot distinguish exemplars part of the target domain from those not part of it. This ability is critical when the aim is to build human trust in real-world settings and essential to avoid usage in domains wherein a system cannot be trusted. In the work presented here, we conduct two qualitative contextual user studies and one controlled experiment to uncover research paths and design openings for the sought distinction. Through our experiments, we find a need to refocus from average case metrics and benchmarking datasets toward systems that can be falsified. The work uncovers and lays bare the need to incorporate and internalise a domain ontology in the systems and/or present evidence for a decision in a fashion that allows a human to use our unique knowledge and reasoning capability. Additional material and code to reproduce our experiments can be found at https://github.com/k3larra/ood.

Machine Learning (ML) and Artificial Intelligence (AI) impact many aspects of human life, from recommending a significant other to assist the search for extraterrestrial life. The area develops rapidly and exiting unexplored design spaces are constantly laid bare. The focus in this work is one of these areas; ML systems where decisions concerning ML model training, usage and selection of target domain lay in the hands of domain experts. 

This work is then on ML systems that function as a tool that augments and/or enhance human capabilities. The approach presented is denoted Human In Command ML (HIC-ML) systems. To enquire into this research domain design experiments of varying fidelity were used. Two of these experiments focus on augmenting human capabilities and targets the domains commuting and sorting batteries. One experiment focuses on enhancing human capabilities by identifying similar hand-painted plates. The experiments are used as illustrative examples to explore settings where domain experts potentially can: independently train an ML model and in an iterative fashion, interact with it and interpret and understand its decisions. 

HIC-ML should be seen as a governance principle that focuses on adding value and meaning to users. In this work, concrete application areas are presented and discussed. To open up for designing ML-based products for the area an abstract model for HIC-ML is constructed and design guidelines are proposed. In addition, terminology and abstractions useful when designing for explicability are presented by imposing structure and rigidity derived from scientific explanations. Together, this opens up for a contextual shift in ML and makes new application areas probable, areas that naturally couples the usage of AI technology to human virtues and potentially, as a consequence, can result in a democratisation of the usage and knowledge concerning this powerful technology.

Mobile apps are an increasingly important part of public transport, and can be seen as part of the journey experience. Personalisation of the app is then one aspect of the experience that, for example, can give travellers a possibility to save favourite journeys for easy access. Such a list of journeys can be extensive and inaccurate if it doesn’t consider the traveller’s context. Making an app context aware and present upcoming journeys transforms the app experience in a personal direction, especially for commuters. By using historical personal contextual data, a travel app can present probable journeys or accurately predict and present an upcoming journey with departure times. The predictions can take place when the app is started or be used to remind a commuter when it is time to leave in order to catch a regularly travelled bus or train.

To address this research opportunity we have created an individually trained Machine Learning (ML) agent that we added to a publicly available commuter app. The added part of the app uses weekday, time, user activity and location to predict a user’s upcoming journey. Predictions are made when the app starts and departure times for the most probable transport are presented to the traveller. In our case a commuter only makes a few journey searches in the app every day which implies that, based on our contextual parameters, it will take at least some weeks to create journey patterns that can give acceptable accuracy for the predictions. In the work we present here, we focus on how to handle this cold start problem e.g. the situation when no or inaccurate historical data is available for the Machine Learning agent to train from. These situations will occur both initially when no data exists and due to concept drift originating from changes in travel patterns. In these situations, no predictions or only inaccurate predictions of upcoming journeys can be made.    

We present experiences and evaluate results gathered when designing the interactions needed for the MT session as well as design decisions for the ML pipeline and the ML agent. The user’s interaction with the ML agent during the teaching session is a crucial factor for the success. During the teaching session, information on what the agent already has learnt has to be presented to the user as well as possibilities to unlearn obsolete commute patterns and to teach new. We present a baseline that shows an idealised situation and the amount of training data that the user needs to add in a MT session to reach acceptable accuracy in predictions. Our main contribution is user evaluated design proposals for the MT session.

Using individually trained ML agents opens up opportunities to protect personal data and this approach can be used to create mobile applications that is independent of local transport providers and thus act on open data on a global scale.

Artificiell intelligens och i synnerhet Maskininlärning (ML) påverkar i hög grad människors liv genom de kan skapa monetärt värde från data. Denna produktifiering av insamlad data påverkar på många sätt våra liv, från val av partner till att rekommendera nästa produkt att konsumera. ML-baserade system fungerar väl i denna roll eftersom de kan förutsäga människors beteende baserat på genomsnittliga prestandamått, men deras användbarhet är mer begränsad i situationer där det är viktigt med transparens visavi de kunskapsrepresentationer ett enskilt beslut baseras på.

 Målet med detta arbete är att kombinera människors och maskiners styrkor via en tydlig maktrelation där en slutanvändare har kommandot. Denna maktrelation bygger på användning av ML-system som är transparenta med bakomliggande orsaker för ett föreslaget beslut. Artificiella neurala nätverk är ett intressant val av ML-teknik för denna uppgift eftersom de kan bygga interna kunskapsrepresentationer från rå data och därför tränas utan specialiserad ML kunskap. Detta innebär att ett neuralt nätverk kan tränas genom att exponeras för data från en måldomän och i denna process internalisera relevanta kunskapsrepresentationer. Därefter kan nätet presentera kontextuella förslag på beslut baserat på dessa representationer. I icke-statiska situationer behöver det fragment av den verkliga världen som internaliseras i ML-systemet kontextualiseras av en människa för att systemet skall vara användbart och tillförlitligt.

 I detta arbete utforskas det ovan beskrivna området via en övergripande forskningsfråga: Vilka utmaningar och möjligheter kan uppstå när en slutanvändare använder neurala nätverk som stöd för enstaka beslut i ett väldefinierat sammanhang?

 För att besvara forskningsfrågan ovan används metodologin forskning genom design, detta på grund av att den valda metodologin matchar öppenheten i forskningsfrågan. Genom sex designexperiment utforskas utmaningar och möjligheter i situationer där enskilda kontextuella beslut är viktiga. De initiala designexperimenten fokuserar främst på möjligheter i situationer där neurala nätverk presterar i paritet med människors kognitiva förmågor och de senare experimenten utforskar utmaningar i situationer där neurala nätverk överträffar människans kognitiva förmågor.  Den andra delen fokuserar främst på metoder som syftar till att förklara beslut föreslagna av det neurala nätverket.

 Detta arbete bidrar till existerande kunskap på tre sätt: (1) utforskande av lärande relaterat till neurala nätverk med målet att presentera en terminologi användbar för kontextuellt beslutsfattande understött av ML-system, den framtagna terminologin inkluderar generativa begrepp som: sann-i-relation-till-domänen, koncept, utanför-distributionen och generalisering, (2) ett antal designriktlinjer, (3) behovet av att justera interna kunskapsrepresentationer i neurala nätverk så att de överensstämmer med koncept vilket skulle kunna medföra att neurala nätverk kan producera förklaringsbara beslut. Jag föreslår även att en framkomlig forskningsstrategi är att träna neurala nätverk med utgångspunkt från grundläggande koncept, som former och färger. Denna strategi innebär att nätverken kan generalisera utifrån dessa generella koncept i olika domäner. Den föreslagna forskningsriktning syftar till att producera mer komplexa förklaringar från neurala nätverk baserat på grundläggande generaliserbara koncept.

Central to deep learning is an ability to generalise within a target domain consistent with human beliefs within the same domain. A label inferred by the neural network then maps to a human mental representation of a, to the label, corresponding concept. If an explanation concerning why a specific decision is promoted it is important that we move from average case performance metrics towards interpretable explanations that build on human understandable concepts connected to the promoted label. In this work, we use Explainable Artificial Intelligence (XAI) methods to investigate if internal knowledge representations in trained neural networks are aligned and generalise in correspondence to human mental representations. Our findings indicate an, in neural networks, epistemic misalignment between machine and human knowledge representations. Consequently, if the goal is classifications explainable for en users we can question the usefulness of neural networks trained without considering concept alignment. 

Contemporary Machine Learning (ML) often focuseson large existing and labeled datasets and metrics aroundaccuracy and performance. In pervasive online systems, conditionschange constantly and there is a need for systems thatcan adapt. In Machine Teaching (MT) a human domain expertis responsible for the knowledge transfer and can thus addressthis. In my work, I focus on domain experts and the importanceof, for the ML system, available features and the space they span.This space confines the, to the ML systems, observable fragmentof the physical world. My investigation of the feature space isgrounded in a conducted study and related theories. The resultof this work is applicable when designing systems where domainexperts have a key role as teachers.

Building interpretable machine learning agents is a challenge that needs to be addressed to make the agents trustworthy and align the usage of the technology with human values. In this work, we focus on how to evaluate interpretability in a machine teaching setting, a settingthat involves a human domain expert as a teacher in relation to a machine learning agent. By using a prototype in a study, we discuss theinterpretability denition and show how interpretability can be evaluatedon a functional-, human- and application level. We end the paperby discussing open questions and suggestions on how our results can be transferable to other domains.

Machine learning research today is dominated by atechnocentric perspective and in many cases disconnected fromthe users of the technology. The machine teaching paradigm insteadshifts the focus from machine learning experts towards thedomain experts and users of machine learning technology. Thisshift opens up for new perspectives on the current use of machinelearning as well as new usage areas to explore. In this study,we apply and map existing machine teaching principles ontoa contextual machine teaching implementation in a commutingsetting. The aim is to highlight areas in machine teaching theorythat requires more attention. The main contribution of this workis an increased focus on available features, the features space andthe potential to transfer some of the domain expert’s explanatorypowers to the machine learning system.

In this paper, we study explanations in a setting where human capabilities are in parity with Machine Learning (ML) capabilities. If an ML system is to be trusted in this situation, limitations in the trained ML model’s abilities have to be exposed to the end-user. A majority of current approaches focus on the task of creating explanations for a proposed decision, but less attention is given to the equally important task of exposing limitations in the ML model’s capabilities, limitations that in turn affect the validity of created explanations. Using a small-scale design experiment we compare human explanations with explanations created by an ML system. This paper explores and presents how the structure and terminology of scientific explanations can expose limitations in the ML models knowledge and be used as an approach for research and design in the area of explainable artificial intelligence.

Concepts are powerful human mental representations used to explain, reason and understand. In this work, we use theories on concepts as an analytical lens to compare internal knowledge representations in neural networks to human concepts. In two image classification studies we find an unclear alignment between these, but more pronounced, we find the need to further develop explanation methods that incorporate concept ontologies. 

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. 

In this paper, we study the complementary characteristics of stationary and mobile devices for traffic data collection. Since stationary devices continuously collect traffic data at fixed locations in a network, they can give insight of the traffic at particular locations over a longer period of time. Mobile devices have wider range and are able to collect traffic data over a larger geographic region. Thus, we argue that both types of technology should be considered to obtain high-quality information about vehicle movements. We present a traffic simulation model, which we use to study the share of successfully identified vehicles when considering both stationary and mobile technologies with varying identification rate. The results of our study, where we focus on freight transport in southern Sweden, confirms that it is possible to identify the majority of vehicles, even when the identification rate is low, and that the share of identified vehicles can be increased by using both stationary and mobile measurement devices.

We propose an optimization model to tackle the problem of determining how projects are assigned to student groups based on a bidding procedure. In order to improve student experience in project-based learning we resort to actively involving them in a transparent and unbiased project allocation process. To evaluate our work, we collected information about the students' own views on how our approach influenced their level of learning and overall learning experience and provide a detailed analysis of the results. The results of our evaluation show that the large majority of students (i.e., 91%) increased or maintained their satisfaction ratings with the proposed procedure after the assignment was concluded, as compared to their attitude towards the process before the project assignment occurred.

As an alternative to the car, the bicycle is considered important for obtaining more sustainable urban transport. The bicycle has many positive effects; however, bicyclists are more vulnerable than users of other transport modes, and the number of bicycle related injuries and fatalities are too high. We present a clustering analysis aiming to support the identification of the locations of bicyclists' perceived unsafety in an urban traffic network, so-called bicycle impediments. In particular, we used an iterative k-means clustering approach, which is a contribution of the current paper, and DBSCAN. In contrast to standard k-means clustering, our iterative k-means clustering approach enables to remove outliers from the data set. In our study, we used data collected by bicyclists travelling in the city of Lund, Sweden, where each data point defines a location and time of a bicyclist's perceived unsafety. The results of our study show that 1) clustering is a useful approach in order to support the identification of perceived unsafe locations for bicyclists in an urban traffic network and 2) it might be beneficial to combine different types of clustering to support the identification process. (C) 2020 The Authors. Published by Elsevier B.V.

The bicycle has many positive effects; however, bicyclists are more vulnerablethan users of other transport modes, andthe number of bicycle related injuries and fatalities are toohigh.We present a clustering analysis aiming to support the identification of the locations ofbicyclists' perceived unsafety in an urban trafficnetwork, so-called bicycle impediments.In  particular,  we presentan  iterative  k-means  clustering approach,  which  in  contrast  to  standard  k-means  clustering, enables to remove outliers and solitary points from the data set. In our study, we used data collected by bicyclists travelling inthe city of Lund, Sweden, where each data point defines a location andtime of a bicyclist's perceived unsafety.The results of our study show that 1) clustering is a usefulapproach in order to support the identification of perceived unsafelocations forbicyclists in an urban traffic networkand2) it might bebeneficial to combine different types of clustering to support theidentification process. Furthermore, using the adjusted Rand index, our results indicate highrobustness of our iterative k-means clustering approach.

It has been shown in previous research that regression modeling can be used in order to predict the number of bicycles registered by a bicycle counter. To improve the prediction accuracy, it has also been suggested that a long-term trend curve estimate can be incorporated in a regression problem formulation. A long-term trend curve estimate aims to capture those factors that are difficult, or even impossible, to explicitly model as input variables in the regression model. In the current paper, we present a regression-based approach for evaluating long-term trend curve estimates regarding their possibility to improve the regression prediction accuracy of bicycle counter data. We illustrate our approach by applying it on a time series recorded by a bicycle counter in Malmö, Sweden. For the considered data set, our experimental results indicate that a polynomial of degree two, which has been fitted to the time series, gives the best prediction.

This article investigates the event-triggered adaptive fuzzy output feedback setpoint regulation control for the surface vessels. The vessel velocities are noisy and small in the setpoint regulation operation and the thrusters have saturation constraints. A high-gain filter is constructed to obtain the vessel velocity estimations from noisy position and heading. An auxiliary dynamic filter with control deviation as the input is adopted to reduce thruster saturation effects. The adaptive fuzzy logic systems approximate vessel's uncertain dynamics. The adaptive dynamic surface control is employed to derive the event-triggered adaptive fuzzy setpoint regulation control depending only on noisy position and heading measurements. By the virtue of the event-triggering, the vessel's thruster acting frequencies are reduced such that the thruster excessive wear is avoided. The computational burden is reduced due to the differentiation avoidance for virtual stabilizing functions required in the traditional backstepping. It is analyzed that the event-triggered adaptive fuzzy setpoint regulation control maintains position and heading at desired points and ensures the closed-loop semi-global stability. Both theoretical analyses and simulations with comparisons validate the effectiveness and the superiority of the control scheme. 

This paper investigates the dynamic event-triggered adaptive neural coordinated disturbance rejection for marine vehicles with external disturbances as the sinusoidal superpositions with unknown frequencies, amplitudes and phases. The vehicle movement mathematical models are transformed into parameterized expressions with the neural networks approximating nonlinear dynamics. The parametric exogenous systems are exploited to express external disturbances, which are converted into the linear canonical models with coordinated changes. The adaptive technique together with disturbance filters realize the disturbance estimation and rejection. By using the vectorial backstepping, the dynamic event-triggered adaptive neural coordinated disturbance rejection controller is derived with the dynamic event-triggering conditions being incorporated to reduce execution frequencies of vehicle's propulsion systems. The coordinated formation control can be achieved with the closed-loop semi-global stability. The dynamic event-triggered adaptive disturbance rejection scheme achieves the disturbance estimation and cancellation without requiring the a priori marine vehicle's model dynamics. Illustrative simulations and comparisons validate the proposed scheme.

With the increasing applications of magnetic robots in medical instruments, the research on different structures and locomotion approaches of magnetic robots has become a hotspot in recent years. A ferrofluid rolling robot (FRR) with magnetic actuation is proposed and enabled to realize a novel locomotion approach in this article. The drive performance of ferrofluid is elaborated, which is characterized by the magnetic torque of a rectangle-shaped object filled with ferrofluid under magnetic field. First, the proposed structure and locomotion mechanism of the FRR are detailed. Moreover, based on the established mathematical models of the FRR, the simulations with straight and turning locomotion are carried out, respectively. Finally, the FRR prototype is manufactured by 3-D printing, and experimental results demonstrate that the feasibility of straight and turning locomotion is verified. The locomotion performance of the FRR is in good agreement with the theoretical models where the root mean square (rms) value of displacement for experiments and simulations is 1.2 mm. In this work, the proposed FRR can automatically switch from straight to turning locomotion with a fast response in an external magnetic field, and does not has magnetism when without a magnetic field. 

In this paper, we investigate the dynamic task scheduling problem with stochastic task arrival times and due dates in the UAV-based intelligent on-demand meal delivery system (UIOMDS) to improve the efficiency. The objective is to minimize the total tardiness. The new constraints and characteristics introduced by UAVs in the problem model are fully studied. An iterated heuristic framework SES (Stochastic Event Scheduling) is proposed to periodically schedule tasks, which consists of a task collection and a dynamic task scheduling phases. Two task collection strategies are introduced and three Roulette-based flight dispatching approaches are employed. A simulated annealing based local search method is integrated to optimize the solutions. The experimental results show that the proposed algorithm is robust and more effective compared with other two existing algorithms.

Named data networking (NDN) is an emerging information-centric networking paradigm, in which the Internet of Things (IoT) achieves excellent scalability. Recent literature proposes the concept of NDN-IoT, which maximizes the expansion of IoT applications by deploying NDN in the IoT. In the NDN, the security is built into the network by embedding a public signature in each data package to verify the authenticity and integrity of the content. However, signature schemes in the NDN-IoT environment are facing several challenges, such as signing security challenge for resource-constrained IoT end devices (EDs) and verification efficiency challenge for NDN routers. This article mainly studies the data package authentication scheme in the package-level security mechanism. Based on mobile edge computing (MEC), an efficient certificateless group signature scheme featured with anonymity, unforgeability, traceability, and key escrow resilience is proposed. The regional and edge architecture is utilized to solve the device management problem of IoT, reducing the risks of content pollution attacks from the data source. By offloading signature pressure to MEC servers, the contradiction between heavy overhead and shortage of ED resources is avoided. Moreover, the verification efficiency in NDN router is much improved via batch verification in the proposed scheme. Both security analysis and experimental simulations show that the proposed MEC-based certificateless group signature scheme is provably secure and practical.

Insider and outsider database threats have more often than not posed a greater challenge as far as integrity and investigation of databases is concerned. Database forensic investigation is a process through which scientifically proven methods can be used to create a hypothesis that can prove or disprove the occurrence of a potential security incident. This paper explores the techniques that can be used to conduct forensic investigations of a compromised MySQL database. The authors have simulated investigative scenarios that have aided to conduct forensic investigative processes and the results are promising.

Data has been claimed to be the new oil of the 21st century as it has seen to be able both to improve the existing products and services as well as to create new revenue streams for its utilizing company with a secondary customers base. However, while there is active streams of research for developing machine learning and data science methods, considerably less has been done to understand and characterize data business activities in the software-intensive companies. This study uses a multiple case study approach in the software-intensive embedded system domain. Four large international embedded system companies were selected as the case study subjects. The objective is to understand how the case companies are developing their activities for successful utilization of the data. The study identifies six distinct stages with their own challenges. In addition, this study serves as a starting for further work for supporting software-intensive embedded system companies to start data business.

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.

This paper addresses the problem of hazard recognition and risk assessment in open and non-predictive environments to support decision making and action selection for UAS. Decision making and action selection incorporate decreasing situational risks and maintain safety as operational constraints. Commonly, neither existing safety standards nor the situation modeling or knowledge representation is considered in that context. This contribution applies a novel approach denoted as a Safety-Driven Behavior Management for UAS focusing on situation modeling, and the problem of knowledge representation in the context of situational risks. It combines the safety standards-oriented hazards analysis and the risk assessment approach with the machine learning-based situation recognition. The illustrative scenario and first experimental results underline the feasibility of the novel approach.

Continuous experimentation (CE) refers to a set of practices used by software companies to rapidly assess the usage, value, and performance of deployed software using data collected from customers and systems in the field using an experimental methodology. However, despite its increasing popularity in developing web-facing applications, CE has not been studied in the development process of business-to-business (B2B) mission-critical systems. By observing the CE practices of different teams, with a case study methodology inside Ericsson, we were able to identify the different practices and techniques used in B2B mission-critical systems and a description and classification of the four possible types of experiments. We present and analyze each of the four types of experiments with examples in the context of the mission-critical long-term evolution (4G) product. These examples show the general experimentation process followed by the teams and the use of the different CE practices and techniques. Based on these examples and the empirical data, we derived the HURRIER process to deliver high-quality solutions that the customers value. Finally, we discuss the challenges, opportunities, and lessons learned from applying CE and the HURRIER process in B2B mission-critical systems. 

In this paper, we present our body-and-mind-centric approach for the design of wearable personal assistants (WPAs) motivated by the fact that such devices are likely to play an increasing role in everyday life. We also report on the utility of such a device for orthopedic surgeons in hospitals. A prototype of the WPA was developed on Google Glass for supporting surgeons in three di↵erent scenarios: (1) touch-less interaction with medical images, (2) tele-presence during surgeries, and (3) mobile access to Electronic Patient Records (EPR) during ward rounds. We evaluated the system in a clinical simulation facility and found that while the WPA can be a viable solution for touch-less interaction and remote collaborations during surgeries, using the WPA in the ward rounds might interfere with social interaction between clinicians and patients. Finally, we present our ongoing exploration of gaze and gesture as alternative input modalities for WPAs inspired by the hospital study.

In this paper, we treat VR as a new writing space in the long tradition of inscription. Constructing Virtual Reality (VR) narratives can then be understood as a process of inscribing text in space, and consuming them as a process of "reading" the space. Our research objective is to explore the meaning-making process afforded by spatial narratives-to test whether VR facilitates traditional ways of weaving complex, multiple narrative strands and provides new opportunities for leveraging space. We argue that, as opposed to the linear space of a printed book, a VR narrative space is similar to the physical space of a museum and can be analyzed on three distinct levels: (1) the architecture of the space itself, (2) the collection, and (3) the individual artifacts. To provide a deeper context for designing VR narratives, we designed and implemented a testbed called RealityMedia to explore digital remediations of traditional narrative devices and the spatial, immersive, and interactive affordances of VR. We conducted task-based user study using a VR headset and follow-up qualitative interviews with 20 participants. Our results highlight how the three semantic levels (space, collection, and artifacts) can work together to constitute meaningful narrative experiences in VR.

The general idea of tracking and tracing apps is that they track the contacts of users so that in case a user tests positive for COVID-19, all the other users that she has been in contact with get a warning signal that they have potentially been in contact with the COVID-19 virus. This is, to quarantine potential carriers of the virus even before they show symptoms. We set up a scenario in which we test the effects the introduction of such an app has on the dynamics of infection with varying amounts of app users. Running the experiments resulted in a slightly lower peak of infections for higher app usages and the total amount of infected individuals over the course of the whole run decreased not more than 10% in any case. The app seems mainly effective in decreasing contacts and infections in public spaces (except hospitals) while increasing the contacts and infections at home.

Den här studien undersöker effekterna av att ersätta en traditionell salstentamen med en open book online-tentamen innehållande frågor av mer fördjupande diskussions- och problemlösningskaraktär. Den nya tentamensformen innebär således både att ett nytt medium för frågor och svar tillämpas, och att strukturen på frågorna i tentamen förändras. Studien fokuserar dels på vad som testas  i relation till lärandemålen (baserat på tentamensfrågor, svar och resultat) och dels studenternas attityder till de olika tentamensformerna (baserat på enkätsvar). Resultaten visar som väntat att den traditionella salstentamen i större utsträckning avslöjar studenternas faktakunskaper medan en djupare förståelse är mer central i den nya tentamensformen. Faktakunskaper kan dock i viss utsträckning även testas i den nya tentamensformen. Studenternas tentamensresultat visar att studenterna har klarat övergången mellan de båda tentamensformerna på ett bra sätt, medan resultaten från enkäterna visar att studenterna är övervägande positiva till den nya tentamensformen men att frågorna upplevdes som svåra och tentamenstiden alltför knapp. Den här typen av frågor introducerar därmed en ny typ av svårighet för studenterna. Studien pekar även på viss problematik med rättssäkerheten i den nya tentamensformen.

Different storage and handling conditions in cold supply chains often cause variations in the remaining shelf life of perishable foods. In particular, the actual shelf life may differ from the expiration date printed on the primary package. Based on temperature sensors placed on or close to the food products, a remaining shelf-life prediction (RSLP) service can be developed, which estimates the remaining shelf life of individual products, in real-time. This type of service may lead to decreased food waste and is used for discovering supply chain inefficiencies and ensuring food quality. Depending on the system architecture, different service qualities can be obtained in terms of usability, accuracy, security, etc. This article presents a novel approach for how to identify and select the most suitable system architectures for RSLP services. The approach is illustrated by ranking different architectures for a RSLP service directed towards the supply chain managers. As a proof of concept, some of the most highly ranked architectures have been implemented and tested in food cold supply chains.

This paper explores the possibilities and challenges of realizing a context-aware travel planner with bidirectional information exchange between the actor and the traveller during unplanned traffic disturbances. A prototype app is implemented and tested to identify potential benefits. The app uses data from open APIs, and beacons to detect the traveller context (which train or train platform the traveller is currently on). Alternative travel paths are presented to the user, and each alternative is associated with a certainty factor reflecting the reliability of the travel time prognoses. The paper also presents an interview study that investigates PT actors’ views on the potential use for actors and travellers of new information about certainty factors and travellers’ contexts, during unplanned traffic disturbances. The results show that this type of travel planner can be realized and that it enables travellers to find ways to reach their destination, in situations where the public t ravel planner only suggests infeasible travel paths. The value for the traveller of the certainty factors are also illustrated. Additionally, the results show that providing actors with information about traveller context and certainty factors opens up for the possibility of more advanced support for both the PT actor and the traveller.

With the increasing diffusion of Internet of Things (IoT) technologies, the transportation of goods sector is in a position to adopt novel intelligent services that cut across the otherwise highly fragmented and heterogeneous market, which today consists of a myriad of actors. Legacy systems that rely upon direct integration between all actors involved in the transportation ecosystem face considerable challenges for information sharing. Meanwhile, IoT based services, which are designed as devices that follow goods and communicate directly to cloud-based backend systems, may provide services that previously were not available. For the purposes of this paper, we present a theoretical framework for classification of such intelligent goods systems based on a literature study. The framework, labelled as the Intelligent Goods Service (IGS) framework, aims at increasing the understanding of the actors, agents, and services involved in an intelligent goods system, and to facilitate system comparisons and the development of new innovative solutions. As an illustration of how the IGS framework can be used and contribute to research in this area, we provide an example from a direct industry-academia collaboration.

In order for people to choose public transport over private car usage, public transport systems must be both reliable and accessible, which is not always the case today. Based on interviews with public transport actors, this paper investigates the missing information and communication flows during unplanned disturbances in the public transport system of southern Sweden. Two potential solution approaches to supply the missing information are also identified: an information system common for all public transport actors in the region, and a traveler check-in system, providing traveler specific information to the actors. The information requirements of both systems, and their potential benefits, are presented. The primary objective of the study is to improve the possibilities for both actors and travelers to act during unplanned disturbances by more efficient information sharing and better traveler information.

New and emerging technologies, such as connected sensors, smartphones and smart cards, offer new possibilities to collect rich real-time information about travelers. Moreover, smartphones also enable travelers to actively share information, for instance, about their intended travel plans. This type of information can be used to improve public transport disturbance management. In this paper, the potential gain of collecting different types of information about travelers is explored to support action decisions made by public transport actors, during unplanned disturbances. Based on interviews and workshops, the paper provides a mapping between different information types and possible action decisions that can be supported. Furthermore, based on a literature review focused on current and potential technical solutions, a guidance to which solutions support which type of action decisions, is also provided. Amongst others, the results show that automated fare collection, which is one of the most commonly implemented systems providing real-time information about the traveler, can support a large number of action decisions relevant in unplanned disturbance scenarios. The technical solution providing the most extensive information, and thereby providing the best support for the action decisions, involves smartphone apps delivering user-generated information. The drawback with this solution is that it might violate privacy, and that it typically relies on the travelers providing relevant information voluntarily.

Travel support for public transport today usually takes no or little account of the traveler’s personal needs and current context. Thereby, travelers are often suggested irrelevant travel plans, which may force them to search for information from other sources. In particular, this is a problem during unplanned disturbances. By incorporating the traveler’s context information into the travel support, travelers could be provided with individually tailored information. This would especially benefit travelers who find it more difficult than others to navigate the public transport system. Furthermore, it might raise the accessibility and general attractiveness of public transport. This paper contributes with an understanding of how information about the traveler’s context can enhance the support provided by travel planners, in the case of disturbances in public transport. In particular, the paper includes a high-level analysis of how and in which situations context information can be useful. The analysis shows how information about the traveler’s context can improve travel planners, as well as highlights some risks in relation to some identified scenarios. Several technologies for retrieving information about the physical context of the traveler are also identified. The study is based on a literature review, a workshop, and interviews with domain experts.

A fuzzy Petri net (FPN) is a powerful tool to model and analyze knowledge-based systems containing vague information. This paper systematically reviews recent developments of the FPN model from the following three perspectives: knowledge representation using FPN, reasoning mechanisms using an FPN framework, and the latest industrial applications using FPN. In addition, some specific modeling and reasoning approaches to FPN to solve the 'state-explosion problem' are illustrated. Furthermore, detailed analysis of the discussed aspects are shown to reveal some interesting findings, as well as their developmental history. Finally, we present conclusions and suggestions for future research directions.

Context: With the advent of Machine Learning (ML) and especially Deep Learning (DL) technology, companies are increasingly using Artificial Intelligence (AI) in systems, along with electronics and software. Nevertheless, the end-to-end process of developing, deploying and evolving ML and DL models in companies brings some challenges related to the design and scaling of these models. For example, access to and availability of data is often challenging, and activities such as collecting, cleaning, preprocessing, and storing data, as well as training, deploying and monitoring the model(s) are complex. Regardless of the level of expertise and/or access to data scientists, companies in all embedded systems domain struggle to build high-performing models due to a lack of established and systematic design methods and processes.

Objective: The overall objective is to establish systematic and structured design methods and processes for the end-to-end process of developing, deploying and successfully evolving ML/DL models.

Method: To achieve the objective, we conducted our research in close collaboration with companies in the embedded systems domain using different empirical research methods such as case study, action research and literature review.

Results and Conclusions: This research provides six main results: First, it identifies the activities that companies undertake in parallel to develop, deploy and evolve ML/DL models, and the challenges associated with them. Second, it presents a conceptual framework for the continuous delivery of ML/DL models to accelerate AI-driven business in companies. Third, it presents a framework based on current literature to accelerate the end-to-end deployment process and advance knowledge on how to integrate, deploy and operationalize ML/DL models. Fourth, it develops a generic framework with five architectural alternatives for deploying ML/DL models at the edge. These architectural alternatives range from a centralized architecture that prioritizes (re)training in the cloud to a decentralized architecture that prioritizes (re)training at the edge. Fifth, it identifies key factors to help companies decide which architecture to choose for deploying ML/DL models. Finally, it explores how MLOps, as a practice that brings together data scientist teams and operations, ensures the continuous delivery and evolution of models. 

The adoption of continuous software engineering practices such as DevOps (Development and Operations) in business operations has contributed to significantly shorter software development and deployment cycles. Recently, the term MLOps (Machine Learning Operations) has gained increasing interest as a practice that brings together data scientists and operations teams. However, the adoption of MLOps in practice is still in its infancy and there are few common guidelines on how to effectively integrate it into existing software development practices. In this paper, we conduct a systematic literature review and a grey literature review to derive a framework that identifies the activities involved in the adoption of MLOps and the stages in which companies evolve as they become more mature and advanced. We validate this framework in three case companies and show how they have managed to adopt and integrate MLOps in their large-scale software development companies. The contribution of this paper is threefold. First, we review contemporary literature to provide an overview of the state-of-the-art in MLOps. Based on this review, we derive an MLOps framework that details the activities involved in the continuous development of machine learning models. Second, we present a maturity model in which we outline the different stages that companies go through in evolving their MLOps practices. Third, we validate our framework in three embedded systems case companies and map the companies to the stages in the maturity model. 

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.