This study examines physical activity patterns among residents of the High North using smartwatch data and machine learning models. We collected 90 d of Apple smartwatch data from 15 Northern Norwegian residents both native and non-native and applied thirteen regression algorithms to predict total energy expenditure. CatBoostRegressor showed better performance (R ^2 = 0.9999, Mean Absolute Error = 0.480), and through SHAP analysis, workout energy, and duration were shown to be the most important predictive variables. While non-native residents showed lower, more variable activity levels with daily step counts of 0–8,000 and energy expenditure mostly below 500 kcal/day, native residents maintained more consistent daily step counts (5,000–17,500) and higher active energy expenditure (1,000-4,000 kcal/day). These results show the need to consider cultural background while developing technology-based physical activity interventions for High North populations and imply that machine learning models can efficiently predict activity patterns in High North latitudes.

One of the key promises of Mobile Edge Computing (MEC) is its low latency. Current large-scale IoT deployments rely on cloud for their reliability, low cost, and ease of use. For outdoor IoT deployments, 5G cellular networks offer significantly enhanced bandwidth and dramatically reduced latency compared to previous generations, enabling real-time data processing and control. Therefore, leveraging 5G connectivity is crucial for outdoor IoT applications requiring responsiveness and complex data handling. Combining MEC with 5G has the potential to provide the ease of cloud computing alongside low latency. We investigate the latency performance on a 5G cellular network with an experimental MEC setup. In our proof-of-concept, we demonstrate the benefits of using an edge-based compute server for real-time power transmission line analytics. We compare our solution with state-of-the-art multi-region cloud deployments and discuss the advantages of mobile edge computing (MEC). Our real-world evaluation demonstrates a low latency of 44.62 ms for MEC compared to cloud regions; however, the gap is narrowing. While such low latencies can benefit real-world deployments, they remain insufficient to meet the stringent requirements of smart power grid operations ( ∼ 8 ms).

A nudge represents a gentle push, intended to help people change behavior in a desirable direction. Personalization and contextawareness make it possible to tailor a nudge to the user's specific needs and the current situation, and thus make it relevant and meaningful to the user. To be effective, a nudge must be given at a time when the user needs to be pushed and is able to follow the nudge. This paper presents how triggers, based on the event-condition-action (ECA) rule, enable timely nudging by starting the nudge design process at the right time so that a nudge is available when the user is susceptible to nudging. The triggers are tailored to the user's situation and need for nudging, by personalizing both event detection and the evaluation of the condition for starting the nudge design action.

Many media providers offer complementary products on different platforms to target a diverse consumer base. Online sports coverage, for instance, may include professionally produced audio and video channels, as well as Web pages and native apps offering live statistics, maps, data visualizations, social commentary and more. Many consumers also engage in parallel usage, setting up streaming products and interactive interfaces on available screens, laptops and handheld devices. This ability to combine products holds great promise, yet, with no coordination, cross-platform user experiences often appear inconsistent and disconnected. We present Control-driven Media (CdM), an extension of the current media model that adds support for coordination and consistency across interfaces, devices, products, and platforms while remaining compatible with existing services, technologies, and workflows. CdM promotes online media control as an independent resource type in multimedia systems. With control as a driving force, CdM offers a highly flexible model, opening up for further innovations in automation, personalization, multi-device support, collaboration and time-driven visualization. Furthermore, CdM bridges the gap between continuous media and Web/native apps, allowing the combined powers of these platforms to be seamlessly exploited as parts of a single, consistent user experience. Extensive research in time-dependent, multi-device, data-driven media experiences supports CdM. In particular, CdM requires a generic and flexible concept for online, timeline-consistent media control, for which a candidate solution (State Trajectory) has recently been published. This paper makes the case for CdM, bringing the significant potential of this model to the attention of research and industry.

New digital technologies like activity trackers, nudge concepts, and approaches can inspire and improve personal health. There is increasing interest in employing such devices to monitor people's health and well-being. These devices can continually gather and examine health-related information from people and groups in their familiar surroundings. Context-aware nudges can assist people in self-managing and enhancing their health. In this protocol paper, we describe how we plan to investigate what motivates people to engage in physical activity (PA), what influences them to accept nudges, and how participant motivation for PA may be impacted by technology use.

Interactive applications are powerful tools for data exploration, visualization and collaboration. Applications featuring viewports are particularly expressive, offering controls for altering perspective by scrolling, panning, zooming or tilting a view. Still, interactivity is inherently live and manual, and often limited to a single interface. We propose to model interactivity as a data source. This way, interactivity may be transmitted from one interface to another, or broadcasted to a distributed audience. Interactivity could also be created or edited by AI-based algorithms, recorded from manual input, stored and made available for on demand playback, or shared in real-time in a multi-view setup or among collaborators in a group. To facilitate such opportunities, we propose State Trajectory, a unifying concept for local and online interactivity. State trajectories extend regular program variables with a temporal dimension and provide built-in support for persistence, real-time sharing, time-consistent recording and playback, and gradual transitions. A concept implementation demonstrates that state trajectories encapsulate significant complexity, yet with a low performance overhead. Using trajectories, support for real-time collaboration and time-shifted replays could be added to a 3'rd party map framework, with minimal modifications to the existing code base.

We propose a novel and adaptive feature space distillation method (AFSD) to reduce the communication overhead among distributed computers. The proposed method improves the Codistillation process by supporting longer update interval rates. AFSD performs knowledge distillates across the models infrequently and provides flexibility to the models in terms of exploring diverse variations in the training process. We perform knowledge distillation in terms of sharing the feature space instead of output only. Therefore, we also propose a new loss function for the Codistillation technique in AFSD. Using the feature space leads to more efficient knowledge transfer between models with a longer update interval rates. In our method, the models can achieve the same accuracy as Allreduce and Codistillation with fewer epochs.

Nudging provides a way to gently influence people to change behavior towards a desired goal, e.g., by moving towards a healthier or more environmentally friendly lifestyle. Personalized and context-aware digital nudging (named smart nudging) can be a powerful tool for efficient nudging by tailoring nudges to the current situation of each individual user. However, designing smart nudges is challenging, as different users may need different supports to improve their behavior. Determining the next nudge for a specific user must be done based on the user’s current situation, abilities, and potential for improvement. In this paper, we focus on the challenge of designing the next nudge by presenting a novel classification of nudges that distinguishes between (i) nudges that are impossible for the user to follow, (ii) nudges that are unlikely to be followed, and (iii) probable nudges that the user can follow. The classification is tailored to individual users based on user profiles, current situations, and knowledge of previous behaviors. This paper describes steps in the nudge design process and a novel set of principles for designing smart nudges.

In areas such as health, environment, and energy consumption, there is a need to do better. A common goal in society is to get people to behave in ways that are sustainable for the environment or support a healthier lifestyle. Nudging is a term known from economics and political theory, for influencing decisions and behavior using suggestions, positive reinforcement, and other non-coercive means. With the extensive use of digital devices, nudging within a digital environment (known as digital nudging) has great potential. We introduce smart nudging, where the guidance of user behavior is presented through digital nudges tailored to be relevant to the current situation of each individual user. The ethics of smart nudging and the transparency of nudging is also discussed. We see a smart nudge as a recommendation to the user, followed by information that both motivates and helps the user choose the suggested behavior. This paper describes such nudgy recommendations, the design of a smart nudge, and an architecture for a smart nudging system. We compare smart nudging to traditional models for recommender systems, and we describe and discuss tools (or approaches) for nudge design. We discuss the challenges of designing personalized smart nudges that evolve and adapt according to the user's reactions to the previous nudging and possible behavioral change of the user.

Large amounts of detailed electronic health data are being collected. Reuse of these data has enormous potential for scientific discoveries that enables the improvement of healthcare systems’ effectiveness, efficiency, and quality of care. However, health data reuse should protect the privacy interests of the stakeholders (i.e., patients and healthcare providers) and promote public good through research. This is particularly challenging when the data are distributed across several data custodians. This paper aims to give an overall overview of existing privacy-preserving techniques for distributed data reuse and their practical applications. We searched for review papers that are focused on privacy-preserving techniques for different stages of distributed data reuse, such as creating dataset that satisfy a given criteria, analyzing the dataset, and releasing statistical results. We analyzed the identified techniques in terms of privacy, data utility, efficiency, and scalability. Practical uses of the techniques are also discussed when there is actual use. Several privacy-preserving data reuse techniques have been identified. The techniques are developed for different stages of distributed data reuse based on de-identification, secure multi-party computation (SMC), or a combination of these two building blocks. Different combinations of the techniques need to be applied for the whole stages of distributed data reuse. Some of the surveyed techniques protect the privacy of data custodians in addition to individuals. The main challenge for de-identification based data reuse techniques is making a balance between utility and privacy. Whereas, efficiency and scalability are the main challenges for SMC based techniques. Enormous progress has been made towards making privacy-preserving reuse of distributed data possible. However, there are only few practical uses of the available techniques. the problem of distributed data reuse also requires governance, legal, and ethical frameworks, as well as the technical solutions. It is not clear whether consent, data-use agreement, and ethics review are required for practical uses of the techniques.

We present a social image recommender system that offers a hybrid filtering approach, combining content- and knowledge-based filtering with a novel social-based filtering, that selects images of social interest to the user, by e.g. being posted by close friends or family. User activity on social media is used when generating a user profile reflecting user interests and social context, and images are recommended according to a combination of social relevance and topic of interest. Our system handles both cross-source user profiling and image recommendation across social media, currently focusing on images from Facebook and Flickr.

Research in community health introduces challenges regarding analysis of the research data. It involves multiple actors in a varity of arenas, and it is often directed towards the local community and children and their families. The legal, ethical and privacy issues involved introduce constraints upon the analysis performed. SNOOP combined with the D2Worm declarative modelling and infrastructure architecture is a promising approach to support a wide range of possible privacy preserving analysis in community health research.

With the myriad of information resources available on the Web, personalization has become important to facilitate information retrieval and recommender systems that provide information and services adapted to the user's needs. A crucial component in any personalization approach is the availability of a user profile that reflects the user's interests, behaviour and intention. In this paper we exploit the fact that for many people the mobile phone is a close companion that follows the person everywhere and is used for many different tasks in the person's daily life. In particular, we focus on mining user activities on NFC-based services to collect information that can be included in a user profile. The paper describes the process of mining usage information from multiple NFC-based applications, both on a smart phone and on back-end systems. We also describe our experiments that lead to and support mining of NFC-based user information.

Mobile network operators' role as keystone players in the smartphone ecosystem is challenged by other actors and technologies that aim to reduce the importance of the Universal Integrated Circuit Card (also known as SIM card). Modern Universal Integrated Circuit Cards are Java Cards that also include a Global Platform conformant Secure Element, usually under the mobile operator's control. We argue that mobile operators still have the opportunity to defend their role by offering easy access for customers and service providers to the Secure Element on the Universal Integrated Circuit Card for storing data and executing applications with high demands for security. The mobile operators could let the customers or service providers own and manage their private Global Platform specified supplementary security domain on the Secure Element. Such access to supplementary security domains on the Universal Integrated Circuit Card can enable new ecosystems and new business models created around this asset. This paper describes a novel smartphone, customer and service provider oriented, technical approach to management of the secure element. We have designed and implemented SecurePlay, a client side, proxy based "lightweight" Trusted Service Manager prototype and have successfully used it to manage Secure Elements on Universal Integrated Circuit Cards in the Telenor operated mobile phone network in Norway. SecurePlay allow operators to cost efficiently enable end users' ownership and operation of their own private security. Implementation details of a proof-of-concept prototype are presented.