Abstracts Track 2021


Nr: 5
Title:

Fuzzy Clustering of Multivariate Time Series based on Quantile Dependence

Authors:

Ángel López-Oriona, Jose A. Vilar and Pierpaolo D'Urso

Abstract: A novel procedure to perform fuzzy clustering of multivariate time series generated from different dependence models is proposed. Different amounts of dissimilarity between the generating models or changes on the dynamic behaviours over time are some arguments justifying a fuzzy approach, where each series is associated to all the clusters with specific membership levels. Our procedure considers quantile-based cross-spectral features and consists of three stages: (i) each element is characterised by a vector of proper estimates of the quantile cross-spectral densities, (ii) principal component analysis is carried out to capture the main differences reducing the effects of the noise, and (iii) the squared Euclidean distance between the first retained principal components is used to perform clustering through the standard fuzzy C-means and fuzzy C-medoids algorithms. The performance of the proposed approach is evaluated in a broad simulation study where several types of generating processes are considered, including linear, nonlinear and dynamic conditional correlation models. Assessment is done in two different ways: by directly measuring the quality of the resulting fuzzy partition and by taking into account the ability of the technique to determine the overlapping nature of series located equidistant from well-defined clusters. The procedure is compared with the few alternatives suggested in the literature, substantially outperforming all of them whatever the underlying process and the evaluation scheme. Two specific applications involving air quality and financial databases illustrate the usefulness of our approach.

Area 1 - SmartWork

Nr: 3
Title:

healthyMe Mobile and iCare Portal: Lifestyle Interventions Management and Privacy-abiding Data Sharing with Carers

Authors:

Christiane Grünloh, Andreea Bondrea, Jasmijn Franke, Dennis Hofs and Boris van Schooten

Abstract: This demo will showcase 1) the modules of the healthyMe smartphone application, 2) the iCare portal and3) the Interventions Manager Service (IMS), all of which were developed as part of the H2020 SmartWork project. The healthyMe smartphone application is the main mobile entry point for the users to collect and visualise physiological, activity and lifestyle data. It is multi-lingual (English, Danish, Portuguese) and is available on Android and iOS. Each module (steps, sleep, heart rate, food diary, weight, exercises) has its own widget, presenting the collected data in daily, weekly and monthly overviews. The integration of external devices supports automatic retrieval of data on physical activity, sleep and heart rate (via the Fitbit integration) and body weight (via Withings integration). The food diary allows users to manually track their food intake, which raises their awareness of the total amount of energy consumed. The office-friendly exercise widget presents a library of video-guided exercises that have been recorded in collaboration with healthcare professionals. The videos allow users to safely perform physical exercises at home or at work at the time of their best convenience. The integrated filter allows the user to select exercises by body parts (shoulders, neck, back, arms, legs). The virtual coach “Amelia” guides users through the application, starting with an intake dialogue through which users can set their activity goals. Depending on their actual level of physical activity that is tracked later on, the goal is automatically adjusted. If a person is less active, the step goal will be adjusted and increased if a person reached their step goals. To prevent demotivation, the automatically adjusted goal is always slightly higher than was reached in the previous week and hence likely to be achievable for the person. The iCare portal is a service that allows (in)formal carers to support the older office worker reaching their health goals. Strong focus is placed on privacy and control in that the office worker can configure within the healthyMe service which data they want to share, from which period of time and with whom. After configuration, summaries of health-related information collected within the healthyMe service are visualised in a web-based portal. This way, the carer can monitor the health status of the office worker and provide support for the self-management of health conditions. The Interventions Manager Services (IMS) is a centralised component within the SmartWork platform that acts as a smart message hub for triggered interventions. From the back-end service side, the IMS can be called if any of the smart services developed within SmartWork decides that some intervention should be triggered. From the client side, the IMS lets the SmartWork client applications register themselves to be notified of triggered interventions. Through the IMS, all smart services have a single entry-point for delivering intervention triggers, and all client applications have a single entry-point for registering to receive triggers. Another motivation for the single entry-point was to avoid overloading the user with multiple notifications of triggered interventions at the same time. Currently, only one intervention is delivered at a given time, and in the future more sophisticated intervention prioritisation mechanisms can be implemented.

Nr: 6
Title:

Elaboration of Intervention Strategies for Health and Workability Management

Authors:

Tom C. Thomsen, Miriam Cabrita, Christiane Grünloh, Rita Kovordanyi, Otilia Kocsis and Willeke van Staalduinen

Abstract: This poster session describes how the H2020 Smartwork project has elaborated intervention strategies regarding individual health management among office workers at +55 of age. The approach has addressed user demands through questionaries mapping these onto six intervention domains: Interventions for health self-management; behavioural interventions to promote physical activity, healthy nutrition and wellbeing; cognitive function interventions to maintain cognitive capacity; work management interventions to reduce work related stress; work-related training interventions, facilitate explicit/implicit knowledge acquisition and intergenerational knowledge transfer; and work environment adaptations. Within the domains, interventions are motivated and referenced to the user needs, further shaped by motivation theories and user values to improve impact, setting goals for behavioural and wellbeing changes to achieve viable results for the individual user. The concept of intervention in this context is carefully defined, what it is and what causes an intervention. The approach enables further mapping of the interventions onto the envisaged SmartWork services and modules, hence establishing linkage from user needs to the services. Monitoring of physiological and behavioral parameters are carried out by sensors and wearables, fed into the Smartwork Artificial Intelligence system which unobtrusively and pervasively monitor health, behaviour, cognitive and emotional status of the worker. Additionally, workers will be asked regularly to report their findings and well-being at work for Machine Learning Purposes. By doing this, SmartWork is capable to identify and assess the functional and cognitive decline risks and trigger interventions which will appear on the devices the worker is using, e.g. such as trainings, relax techniques and advises.

Nr: 7
Title:

Using Morphic to Promote Inclusion in the Workplace

Authors:

Ignacio Peinado, Javier Hernández and Eva de Lera

Abstract: Morphic makes computers easier to use. Morphic unveils the usability and accessibility features present in the computer and is able to download and configure third-party software if required by the user. Once the user has configured a computer, they can apply this configuration in a different Morphic-enabled computer. Power users can create Morphic bundles with pre-configured configurations that can be installed or imported in any compatible computer. The SmartWork project marks the first time Morphic will be tested in Europe, and within a corporate environment. Morphic facilitates the inclusion of people with disabilities in the workplace by providing workers and employers with: - True digital equity. Workers who need assistive technology software are able to sit down at any computer, anywhere in their organization, and have the software they needs appear on that computer, configured to their needs and preferences. - Easy discovery of accessibility features and Assistive Technologies, which will hopefully lead to: - Bringing inclusive culture into corporate culture. Morphic can help companies build a more inclusive working culture by raising awareness about the whole range of accessibility needs and solutions, facilitating pre-built configuration bundles for different personae, that will foster empathy and will allow IT staff to involve users with special needs more quickly and efficiently. - Better use of company resources. With Morphic, each computer can be used by any worker in the organization, regardless of their needs and preferences. Finally, Morphic will not only help users with disabilities, but all workers may benefit from the curb cut effect. In the demo, we will present an implementation of some of Morphic’s key concepts that have been developed as part of the SmartWork project. In a computer with Morphic installed, we will key in with one user, and the computer will be configured according to the needs and preferences of the demo user, even installing software that was not installed before. Also, we will demonstrate how easy it is to change the settings of the computer using the Morphic Bar and store these settings in the user’s needs and preferences set. We will then key the demo user out and we will see how the computer will return to its original state, and the software required by the user will be uninstalled.