Search results

This study aims to propose novel edge device-tailored federated learning algorithms of local classifiers (stochastic gradient descent, support vector machines), namely, FL-lSGD and FL-lSVM. These algorithms are designed to address the challenge of large-scale ImageNet classification.

The authors’ FL-lSGD and FL-lSVM trains in a parallel and incremental manner to build an ensemble local classifier on Raspberry Pis without requiring data exchange. The algorithms load small data blocks of the local training subset stored on the Raspberry Pi sequentially to train the local classifiers. The data block is split into k partitions using the k-means algorithm, and models are trained in parallel on each data partition to enable local data classification.

Empirical test results on the ImageNet data set show that the authors’ FL-lSGD and FL-lSVM algorithms with 4 Raspberry Pis (Quad core Cortex-A72, ARM v8, 64-bit SoC @ 1.5GHz, 4GB RAM) are faster than the state-of-the-art LIBLINEAR algorithm run on a PC (Intel(R) Core i7-4790 CPU, 3.6 GHz, 4 cores, 32GB RAM).

Efficiently addressing the challenge of large-scale ImageNet classification, the authors’ novel federated learning algorithms of local classifiers have been tailored to work on the Raspberry Pi. These algorithms can handle 1,281,167 images and 1,000 classes effectively.

For privacy protection, federated learning based on data separation allows machine learning models to be trained on remote devices or in isolated data devices. However, due to the limited resources such as bandwidth and power of local devices, communication in federated learning can be much slower than in local computing. This study aims to improve communication efficiency by reducing the number of communication rounds and the size of information transmitted in each round.

This paper allows each user node to perform multiple local trainings, then upload the local model parameters to a central server. The central server updates the global model parameters by weighted averaging the parameter information. Based on this aggregation, user nodes first cluster the parameter information to be uploaded and then replace each value with the mean value of its cluster. Considering the asymmetry of the federated learning framework, adaptively select the optimal number of clusters required to compress the model information.

While maintaining the loss convergence rate similar to that of federated averaging, the test accuracy did not decrease significantly.

By compressing uplink traffic, the work can improve communication efficiency on dynamic networks with limited resources.

The paper proposes a privacy-preserving artificial intelligence-enabled video surveillance technology to monitor social distancing in public spaces.

The paper proposes a new Responsible Artificial Intelligence Implementation Framework to guide the proposed solution's design and development. It defines responsible artificial intelligence criteria that the solution needs to meet and provides checklists to enforce the criteria throughout the process. To preserve data privacy, the proposed system incorporates a federated learning approach to allow computation performed on edge devices to limit sensitive and identifiable data movement and eliminate the dependency of cloud computing at a central server.

The proposed system is evaluated through a case study of monitoring social distancing at an airport. The results discuss how the system can fully address the case study's requirements in terms of its reliability, its usefulness when deployed to the airport's cameras, and its compliance with responsible artificial intelligence.

The paper makes three contributions. First, it proposes a real-time social distancing breach detection system on edge that extends from a combination of cutting-edge people detection and tracking algorithms to achieve robust performance. Second, it proposes a design approach to develop responsible artificial intelligence in video surveillance contexts. Third, it presents results and discussion from a comprehensive evaluation in the context of a case study at an airport to demonstrate the proposed system's robust performance and practical usefulness.

The purpose of this research is to explore how one unsanctioned community sport organization (CSO), AM Hockey, sought to acquire legitimacy in a highly institutionalized minor hockey marketplace at various points in its organizational life cycle.

This study was guided by instrumental case study methodology. Twenty (20) AM Hockey stakeholders from a variety of roles (e.g. executives, program directors and coaches) were interviewed. Document analysis was also utilized to supplement the interviewees. Internal and public documents reflective of the CSO's creation and growth were obtained.

Findings revealed that the CSO had to navigate distinct phases of evolution including the Building, Growth, Competition and Stabilization phases. Although the four life cycle phases identified in this study share similarities with the phases identified by Lester et al. (2003), findings indicated that institutional work mechanisms must be understood in their context as they can vary over the life cycle of an organization. Therefore, start-up sports organizations must approach the pursuit of legitimacy as a continual process rather than something acquired and defended through maintenance work.

Developing legitimacy remains a central challenge for CSOs that seek to deliver alternative sport programming, yet it continues to be understudied. Ultimately, the long-term viability of an unsanctioned CSO in a federated sports system relies, in part, on its ability to continually determine the actions needed to achieve legitimacy within its environment.

The purpose of this study is to examine the state of research into adoption of machine learning systems within the health sector, to identify themes that have been studied and observe the important gaps in the literature that can inform a research agenda going forward.

A systematic literature strategy was utilized to identify and analyze scientific papers between 2012 and 2022. A total of 28 articles were identified and reviewed.

The outcomes reveal that while advances in machine learning have the potential to improve service access and delivery, there have been sporadic growth of literature in this area which is perhaps surprising given the immense potential of machine learning within the health sector. The findings further reveal that themes such as recordkeeping, drugs development and streamlining of treatment have primarily been focused on by the majority of authors in this area.

The search was limited to journal articles published in English, resulting in the exclusion of studies disseminated through alternative channels, such as conferences, and those published in languages other than English. Considering that scholars in developing nations may encounter less difficulty in disseminating their work through alternative channels and that numerous emerging nations employ languages other than English, it is plausible that certain research has been overlooked in the present investigation.

This review provides insights into future research avenues for theory, content and context on adoption of machine learning within the health sector.

Effective use of data is critically important for the provision of health services. A large proportion of employees in health organisations work in non-clinical roles and play a major part in organisational information flows. However, their practice, data-related capabilities and learning needs have been rarely studied. The purpose of this paper is to investigate issues of capabilities and learning needs related to employees' interactions with data in non-clinical work roles.

The study used a mixed-method approach. Qualitative methods were used to explore issues, and survey was administered to gather additional data.

Data use and related capabilities at the workplace are highly contextual. A range of general, core and data-specific capabilities, underpinned by transferable skills and personal traits, enable successful interactions with data. Continuous learning is needed in most areas related to data use.

The study was conducted in a large public-health organisation in Australia, which is not representative of unique organisations elsewhere. The study has implications for the provision of health services, workplace learning and education.

Findings have implications for organisational decisions related to data-use and workplace learning, and for formal education and lifelong learning.

The study contributes to closing a research gap in understanding interactions with data, capabilities and learning needs of employees in non-clinical work roles. Capabilities continuum presented in this paper can be used to inform education, training and service provision. The workplace-based results contribute to theoretical considerations of capabilities required for work in technology-rich environments.

This study aims to examine staff responses to management control systems (MCS) changes in an Australian university. Through the analysis of the category of staff responses, it aims to understand the perception gaps among the staff at different levels of the university.

Using a case study approach on an Australian university, data was collected from interviews with staff across three hierarchical levels to explore their behavioural responses.

This study finds that staff at all levels largely complied with MCS changes due to institutional enforcement. Top management emphasised aligning with government policies and funding, often using manipulation and compartmentalisation tactics in implementing the new MCS. Mid-level managers generally favour research strategies but feel excluded from decision-making and have limited influence over funding. They adopted a balancing tactic within a compromise strategy. Meanwhile, operating-level academics had mixed experiences, feeling largely powerless in influencing MCS while also showing instances of self-motivated compliance. Overall, the study reveals varying responses across different hierarchical levels, highlighting the complexities of MCS changes in staff behaviour and attitudes.

The insights from this study can guide university administrators and policymakers in understanding the intricate variations in staff reactions to institutional changes. By recognising the factors that drive compliance and defiance, institutions can better navigate and implement changes in MCS.

This research offers a unique perspective on the behavioural side of MCS changes in higher education. By focusing on varied hierarchical levels within a university, the study provides a granular understanding of individual responses, enriching the existing literature on MCS transitions in academia.

Developments regarding the use of artificial intelligence (AI) in transportation systems, one of the important stakeholders of tourism, are remarkable. However, no review thus far has provided a comprehensive overview of research on AI in transportation systems.

To fill this gap, this study uses the VOSviewer software to present a bibliometric review of the current scientific literature in the field of AI-related tourism research. The theme of AI in transportation systems was explored in the Web of Science database.

The original search yielded 642 documents, which were then filtered by parameters. For publications related to AI in transportation systems, the most cited documents, leading authors, productive countries, co-occurrence analysis of keywords and bibliographic matching of documents were examined. This report shows that there has been a recent increase in research on AI in transport systems. However, there is only one study on tourism. The country that contributed the most is China with 298 studies. The most used keyword in the documents was intelligent transportation system.

The bibliometric analysis of the existing work provided a valuable and seminal reference for researchers and practitioners in AI-related in transportation system.

The purpose of this study is to frame digital transformation (DT) within municipalities to improve the life cycles of urban infrastructure.

The study provides the results from a systematic review of the literature on concepts of DT and its implications for municipalities, barriers and challenges to DT, as well existing DT frameworks for municipalities and their built assets. This literature review leads to the development of a DT framework to help cities conduct a planned and federated DT beforehand. Then, workshops are conducted with two major Canadian municipalities.

The results of these studies point to the need for a dedicated DT framework for municipalities because of their particular context and their role and proximity to citizens. The theoretical framework develops 22 elements, which are divided among 6 categories. Through its application, the framework helps to identify and target the predominant issues hindering the DT of municipalities, specifically “legacy practices” and “data management.”

Limitations include limited experimental conditions and small sample size. Further work is needed to validate the framework. Other approaches are advocated to complement the data collection and analysis to generate more convincing results.

The theoretical framework was validated through two case studies on two large Canadian municipalities.

Municipalities maximize the value they provide to citizens and to be at the forefront of resilience and sustainability concerns. The use of technology, digital processes and initiatives helps cities to improve planning, optimize works and provide better services to citizens.

The framework is original in that it specifically aligns assets management with DT in a municipal context.