19683 research outputs found
Sort by
FLCP: federated learning framework with communication-efficient and privacy-preserving
No full textWithin the federated learning (FL) framework, the client collaboratively trains the model in coordination with a central server, while the training data can be kept locally on the client. Thus, the FL framework mitigates the privacy disclosure and costs related to conventional centralized machine learning. Nevertheless, current surveys indicate that FL still has problems in terms of communication efficiency and privacy risks. In this paper, to solve these problems, we develop an FL framework with communication-efficient and privacy-preserving (FLCP). To realize the FLCP, we design a novel compression algorithm with efficient communication, namely, adaptive weight compression FedAvg (AWC-FedAvg). On the basis of the non-independent and identically distributed (non-IID) and unbalanced data distribution in FL, a specific compression rate is provided for each client, and homomorphic encryption (HE) and differential privacy (DP) are integrated to provide demonstrable privacy protection and maintain the desirability of the model. Therefore, our proposed FLCP smoothly balances communication efficiency and privacy risks, and we prove its security against “honest-but-curious” servers and extreme collusion under the defined threat model. We evaluate the scheme by comparing it with state-of-the-art results on the MNIST and CIFAR-10 datasets. The results show that the FLCP performs better in terms of training efficiency and model accuracy than the baseline method
Foot, ankle, and leg problems in Australian primary care: consultation patterns, management practices, and costs
No full textOBJECTIVE: To explore consultation patterns, management practices, and costs of foot, ankle, and leg problems in Australian primary care. STUDY DESIGN: We analyzed data from the Bettering the Evaluation and Care of Health program, April 2000 to March 2016. Foot, ankle, and leg problems were identified using the International Classification of Primary Care, Version 2 PLUS terminology. Data were summarized using descriptive statistics examining general practitioner (GP) and patient characteristics associated with a foot, ankle, or leg problem being managed. Cost to government was estimated by extracting fees for GP consultations, diagnostic imaging, and pathology services from the Medicare Benefits Schedule (MBS) database. Costs for prescription-only medicines were extracted from the Pharmaceutical Benefits Schedule and for nonprescribed medications, large banner discount pharmacy prices were used. RESULTS: GPs recorded 1,568,100 patient encounters, at which 50,877 foot, ankle, or leg problems were managed at a rate of 3.24 (95% confidence intervals [CIs] 3.21-3.28) per 100 encounters. The management rate of foot, ankle, or leg problems was higher for certain patient characteristics (older, having a health care card, socioeconomically disadvantaged, non-Indigenous, and being English speaking) and GP characteristics (male sex, older age, and Australian graduate). The most frequently used management practice was the use of medications. The average cost (Australian dollars) per encounter was A256m. CONCLUSIONS: Foot, ankle, and leg problems are frequently managed by GPs, and the costs associated with their management represent a substantial economic impact in Australian primary care
High-throughput sequencing for plant virology diagnostics and its potential in plant health certification
No full textHigh-throughput sequencing (HTS) technologies have revolutionized plant virology through simultaneous detection of mixed viral infections. HTS advances have uncovered and improved understanding of virus biology, ecology, and evolution which is vital for viral disease management. Plant viruses continue to threaten global agricultural productivity and strict quarantine measures are essential to prevent the introduction and spread of virulent viruses around the world. The gradual decrease in HTS operational costs, including improved computational systems and automation through robotics, has facilitated the adoption of this tool for plant diagnostics, including its use in surveillance and quarantine programs. However, the speed of technology advancements and distinct HTS chemistries, laboratory procedures, data management, and bioinformatic analyses have proven challenging. In addition, the lack of viral species reference sequences, compared with the estimated number of distinct viral taxa, makes classification and identification of novel viruses difficult. There is a need for standardized HTS testing, especially within plant health programs. In this review, we consider the application of HTS in plant virology, explore the technical challenges faced and the opportunities for HTS in plant health certification. We propose standards for overcoming current barriers and for ensuring reliable and reproducible results. These efforts will impact global plant health by reducing the risk of introduction and the spread of damaging novel viruses
The integration of knowledge graph convolution network with denoising autoencoder
No full textThe knowledge graph convolution network (KGCN) is a recommendation model that provides a set of top recommendations based on knowledge graph developed between users, items, and their attributes. In this study, we integrate the KGCN model with denoising autoencoder (DAE) to improve its recommendation performance. A trained DAE is used to sample K-dimensional latent representation for each user, which then transforms that representation to generate a probability distribution over items. The relationship between acquired latent representation and the meta features is modelled using multivariate multiple regression (MMR) kernel. As a result, without the need for new configuration assessments, performance estimation of new data is pursued directly through MMR and the decoder of DAE. Empirically, we demonstrate that on real-world datasets, the proposed method substantially outperforms other state-of-the-art baselines. Movie-Lens 100K (ML-100K) and Movie-Lens 1M (ML-1M), two common MovieLens datasets, are used to verify the accuracy of the proposed approach. The results from experiments show significant improvement of 41.17% when the proposed method is applied on KGCN model. The proposed framework outperforms other state-of-the-art frameworks on Recall@K and normalized discounted cumulative gain (NDCG@K) metrics by achieving higher scores for Recall@5, Recall@10, NDCG@1, and NDCG@10
“Navigating in a maze without a map”. Partners’ experiences of hyperemesis gravidarum- a qualitative study
No full textObjective: A supportive environment for women with Hyperemesis Gravidarum is crucial but not always provided. There is a lack of research regarding Hyperemesis Gravidarum, its impact on the family, and the partner's perception of supporting their spouse. Thus, this study aims to explore partners’ experiences of Hyperemesis Gravidarum during their spousés pregnancy. Methods: Data were gathered through 13 individual, semi-structured, in-depth, digital interviews with partners of women who had experienced Hyperemesis Gravidarum and analysed with Qualitative Content Analysis. The partners were recruited through advertisement on a social media platform and were exclusively males, representing 8 of 21 Swedish regions. The mean age was 34, and they had, on average, 1 previous child. The mean time from the experience to the interview was 12 months. Findings: The main theme, “Navigating in a maze without a map”, explains partners’ situation as stressful and demanding when their spouse suffers from Hyperemesis Gravidarum, with insufficient support and guidance from healthcare providers. The analysis resulted in three themes: “Standing alone with a demanding responsibility”, “Being in a lottery when facing healthcare”, and “Climbing the mountain together.” The themes display challenges within everyday life and healthcare, as well as strained relations within the family. Conclusion: Partners experience a need to support their spouse in every aspect of daily life and advocate for adequate healthcare. Healthcare professionals must support and acknowledge the partners' struggles during the demanding situation with Hyperemesis Gravidarum
Building a place-based model for learning about Indigenous cultures: A case study of an immersive Kakadu experience
No full textSubmitted in total fulfillment of the requirements for the degree of Master of Outdoor Education and Environment to the School of Education, La Trobe University, Victoria, Australia.</p
“A Girl Cannot Become a Man Through Boasts”: Constructing Gender in Medieval Turkic Epic and Hagiography
No full textA thesis submitted in partial fulfilment of the requirements for the degree of Master of Arts to the Department of Archaeology and History, School of Humanities and Social Sciences, La Trobe University, Victoria, Australia.</p
The Limitations of Partisan Cues in Influencing Belief in Political Conspiracy Theories
No full textSubmitted in total fulfillment of the requirements for the degree of Master of Psychological Science to the School of Psychology and Public Health, La Trobe University, Victoria, Australia.</p
Interprofessional Working: A Program of Research
No full textSubmitted in the total fulfilment of the requirements for the degree of Doctor of Philosophy to the School of Nursing and Midwifery, La Trobe University, Victoria, Australia.</p
Employable Me: Australian Higher Education and the Employability Agenda
No full textAbstract: Few issues have attracted as much policy interest in the tertiary sector as graduate employability. Graduate employability positions universities and their students as key players in the national economy. At the same time, the standard conception of graduate employability, as it has evolved from human capital theory and modified by neoliberal ideology, has met with significant criticism. This paper reports on our analysis of the strategic plans of Australia’s 42 operating universities current in 2018 to better understand (1) the extent to which employability was embedded in each university’s strategic priorities and (2) the ways in which employability was characterised in those plans. Our paper provides empirical evidence of the way in which Australian universities universally and uniformly adopted a particular model of employability, simultaneously claiming its distinctiveness. Our analysis suggests the need for Australian universities to take a more thoughtful and nuanced approach to graduate employability.</p