Recension av SoTL in action – illuminating critical moments of practice (Chick, 2018).

Machine learning (ML) is increasingly used in diverse fields, including animal behavior research. However, its application to ambiguous data requires careful consideration to avoid uncritical interpretations. This paper extends prior research on ringed mallards where sensors revealed their movements in southern Sweden, particularly in areas with small lakes. The primary focus is to distinguish the movement patterns of wild and farmed mallards. While well-known statistical methods can capture such differences, ML also provides opportunities to simulate behaviors outside of the core study span. Building on this, this study applies ML techniques to simulate these movements, using the previously collected data. It is crucial to note that unrefined application of ML can lead to incomplete or misleading outcomes. Challenges in the data include disparities in swimming and flying records, farmed mallards’ biased data due to feeding points, and extended intervals between data points. This research highlights these data challenges, while identifying discernible patterns, as well as proposing approaches to meet such challenges. The key contribution lies in separating incompatible data and, through different ML models, handle these separately to enhance the reliability of the simulation models. This approach ensures a more credible and nuanced understanding of mallard movements, demonstrating the importance of critical analysis in ML applications in wildlife studies.

Releasing farmed mallards into the wild is a common practice in wildlife management worldwide, involving millions of birds annually, and is mainly carried out to increase hunting opportunities. Ringing and previous research show that released mallards have low survival also outside the hunting season, and that survivors may compromise migration habits, morphology, and adaptations of the wild population. Detailed local movements of released mallards have not been studied before, despite the importance of spatiotemporal patterns for understanding the impact of releases and their utility for hunting. We studied local movements in August–October of 11 wild and 44 released mallards caught in the same wetland in southern Sweden and provided with GPS tags. Wild mallards moved longer distances than farmed, over the whole diel cycle, as well as during three out of four separate periods of the day (dawn, day, and dusk). Mallards of both origins moved the longest distances during dusk and dawn, and the shortest during the night. Males and females did not differ significantly in distance moved, regardless of origin (wild versus farmed). Our study demonstrates large differences in spatiotemporal movement patterns between wild and farmed mallards. The typical day of wild mallards included movements between wetlands in the landscape, likely to foraging sites known from previous experience. However, wild mallards frequently returned to the study wetland, probably attracted by supplementary bait. On the other hand, farmed mallards seldom left the study wetland, despite the possibility of accompanying wild birds to other sites. The sedentary behavior of farmed mallards and the fact that wild birds come to join them are both beneficial for hunting purposes. Limited movements of farmed mallards, together with their low survival, could also be positive as they limit hybridization between wild and farmed mallards, as well as dispersal of nutrients.

The avoidance of mortality in lung cancer is highly dependent on finding defects in the lungs early, to initiate effective treatments in time. Most often, lung disorders are diagnosed and treated using chest radiographs and CT scans. Methods based on machine learning can complement human observations and increase precisions of accuracy by mapping an CT image against a trained artificial neural network. The efficiency and accuracy of training such a network, however, depends on the availability of the performance of an underlying computer system, and the quality and size of images. The use of neural network structures with high intrinsic performance is therefore significant. This contribution focuses on comparisons between different Convolutional Neural Networks and formats on datasets to contribute to a good basis for decision-making in the context of possible lung cancer.

Image processing tasks have benefited from deep learning models based on convolutional neural networks. However, the success of image classification models is dependent on several factors such image quality, dataset size and class distribution. Achieving acceptable accuracies with datasets not meeting these requirements is challenging. Domain specific dataset augmentation techniques have been proposed to mitigate the problem. This paper investigates adaptation of multi-source datasets as an augmentation approach to improve accuracy of crack detection in bridge concrete structures from low quality images in limited and imbalanced datasets. While experimental results show that data augmentation can improve accuracy of detection, we anticipate achieving even better results by combining this approach with generative machine learning models in future research.

While Convolutional Neural Networks are generally considered efficient structures for image processing, that task has been shown to be non-trivial in cases of datasets of imbalanced and low-quality images. Here, it is not only significant to train an algorithm towards a high accuracy enough, but also to do that in a limited amount of time. This can be understood from that low-quality datasets may require several experiments on representations of hyperparameters for an efficient network. Such experiments may be costly, not only generally in terms of time, but also in terms of payments for cloud services as a development platform. This contribution builds upon previous investigations that showed promising techniques for attacking a problem with images of cracks in bridge concrete. Results from that work is here further improved through data augmentation techniques where simply inverting images is proven to improve the performance of the network. Besides this, results will be presented regarding a prognostication of the time required to train the proposed Convolutional Neural Networks.

PurposeThis study aims to address how a higher education pedagogical course in sustainable development (SD) for university educators affects their teaching efforts in providing sustainability matters for students. Design/methodology/approachWith the aim of improving that course, a case study approach was used to understand how the educators made use of the course in their teaching practice. Data were collected as written and oral feedback reflections and as semi-structured interviews at course completion. FindingsEducators clearly express that they understand the concept "about" SD, but there are only vague expressions of a developed teaching repertoire to address education "for" SD in their teaching practice. Research limitations/implicationsWhen it comes to the purposes of developing sustainability literacy among students, implications from the study furthermore address the needs for further clarifications on both structure and intent on the course presented in this contribution. Practical implicationsThe educators as well as their students will be exposed to, and trained in concepts, to prepare them to act in alignment with SD. This, in turn, meets requirements from higher education authorities concerning SD at higher education institutions. Social implicationsA core aim of the covered approach is to support student readiness in SD, and for those to become future agents of positive change. Originality/valueThis study has a focus on presenting how educators change the structures of courses and learning elements to approach SD in their teachings.

Intelligent transport systems (ITS) explore the benefits of wearable devices to monitor driver health and aid in emergencies and advertising. It integrates multitasking, automatically finding a tourist spot, ordering food, and smartly turning on and off a vehicle. In a medical emergency, passengers can access nearby medical facilities, which are not available in the current ITS. To sort out this deficiency, develop a wearable device for autonomous vehicle driver health and passenger interaction system. As part of the embedded system, the wearable gadget is affixed to the vehicle driver and can control the vehicle's motor operation, accelerometers, and gyroscopes. Drivers can travel more comfortably and engage co-passengers by directing promotion and advertisement. The experiment determines which age group of drivers is best suited for handling multiple services simultaneously.

Convolutional Neural Networks are among the most effective algorithms for image analysis applications. However, the accuracy of the algorithms depends on the availability of powerful computational resources and the quality of the images used to train the models. This paper investigates ways to build robust models to detect cracks in concrete structures using low resolution images and third-party datasets. Our experiments show that reducing image sizes by a factor of 4 does not significantly impact the accuracy. This is helpful to shorten execution time and hence lower cloud service costs. It is also observed that applying a model trained on one image dataset to detect cracks in images from a different source is not a trivial task.

Enligt Högskolelagen skall högskolorna i sin verksamhet främja en hållbar utveckling. FNs Agenda 2030 och dess 17 globala hållbarhetsmål belyser ytterligare högskolornas roll att utveckla och sprida kunskap om hållbar utveckling. Det ställs dock krav på högskolelärares insikter kring hållbarhetsbegreppet och iscensättandet av dess flervetenskaplighet i undervisningssituationen. Den här artikeln beskriver en fakultetsövergripande högskolepedagogisk kurs, Undervisning för hållbar utveckling, med dessa krav som utgångspunkt. Kursens genomförande och design presenteras, liksom syfte och lärdomar.

Purpose This paper aims to unveil how sustainability is integrated into the courses/programmes of higher education institutions. The research question addressed is: how do academics representing different disciplines cooperate and engage in the work of integrating sustainability into their teaching programmes. Design/methodology/approach This paper draws upon the notions of practise variation and institutional work from institutional theory and empirically focusses on the case of Kristianstad University (Sweden). This case is based on an autoethnographic approach and illustrates the experiences shared by six colleagues, representing different disciplines, engaged in implementing sustainability in their courses/programmes. Findings The findings highlight how academics representing different disciplines, with specific traditions and characteristics, face the sustainability challenge. Despite being bound by similar sustainable development goals, differences across disciplines need to be acknowledged and used as an asset if trans-disciplinarity is the ultimate goal. Research limitations/implications Although the intrinsic motivation of individuals to work with sustainability might be a strong driver, the implementation of sustainability within courses/programmes and across disciplines requires joint efforts and collective institutional work. Practical implications By highlighting how academics engage in the work of integrating sustainability, this study emphasizes that managers of higher education institutions need to account for the time and additional resources needed to ensure that academics effectively cope with sustainability. Intrinsic motivation may not last if organizational structures and leadership are not supportive on a practical level and in the long run. Social implications With the successful implementation of a holistic approach to sustainability, students will have better insights and understanding of both themselves and the surrounding society, laying the ground for an inclusive future society. Originality/value This paper emphasizes the gradual approach to be followed when sustainability becomes part of an organization-wide discourse. Dialogues within and across disciplines are needed to overcome silo thinking and stimulate cooperation within a trans-disciplinary approach.

This paper will discuss the future-ready university at the level of its future-ready teachers with regard to their teaching and learning practice for sustainable development. Academic institutions have both a role in promoting discussions of concern based on their specialized disciplines and a role in educating students to be future-ready to contribute to the society in a sustainable way. However, carrying out such roles with sufficient credibility may not be a matter of course for university teachers, who need sufficient insights into both sustainability per se and sustainable pedagogical teaching practice. This paper stresses the importance to the development of the future-ready university of cultivating sustainability, and provides an “educate the educators” project as an example.

Introduktion till Temanumret : laraktiviteter for att uppna varderingsformaga och forhallningssatt

Project-based educational forms are at the core of the CDIO concept, where students should be trained in contexts of complex enough tasks to prepare for the complexity of industry projects. Besides ...

The core of CDIO addresses criticism from engineering industry according engineering education having too much focus on theoretical training. Here, practice, and especially integrating theory and p ...

An important aspect of elderly people's habits can be connected to food and meal. To be able to make your own choices according good, and healthy food, even in situations of high age, and functional disabilities adds to quality of life. This contribution covers outcomes of the Active Ageing project, dealing with studies of elderly people's food situation, to be able to find suitable computer based support systems in order to provide appropriate food. Especially, background investigations will be covered, and prototype support systems will be outlined.