A Fair Federated Graph Learning Algorithm Based on Gradient and Prototype Contributions

Aim/Purpose: This article addresses the lack of research on BIPOC (Black, Indigenous, People of Color) PhD students’ experiences in collaborative learning processes. It aims to fill this gap by using collaborative autoethnography to analyze the experiences of four BIPOC doctoral students who participated as co-researchers in a year-long collaborative program evaluation self-study of their academic program. Background: The authors discuss their experience in a student leadership role in the collaborative program evaluation process of their doctoral program as a form of collaborative learning, how their identities influenced the processes they participated in, and the themes of what they learned during the evaluation process. The article ends with recommendations to improve the experience and impact of BIPOC doctoral students’ participation and better align the skills learned at the PhD level with the broader job market. Methodology: The study uses the qualitative method of collaborative autoethnography to analyze the perspectives and experiences of four BIPOC doctoral students within a broader cultural context. Contribution: This article provides insight on how to improve the experience of BIPOC doctoral students’ participation in the collaborative process and better align the skills learned at the PhD level with the broader job market. It provides four recommendations: first, align student training with job market realities; second, dismantle systemic and structural barriers; third, respond to the effects of social location; and last, negotiate power through consensus building. Findings: Five themes emerged from the reflections. First, consider how important planning and good communication are to the collaborative learning process in terms of preparation, group dynamics, and resource availability. Second, power disparities existed among faculty, student researchers, and committees. Third, consider how the researchers’ identities affected their interpretations and interactions at the individual, group, and organizational levels. Fourth, the importance of negotiating between diverse perspectives and interests; and finally, what they gained from participating in the collaborative review process. Recommendations for Practitioners: The authors provided four recommendations for improving the experience of BIPCO doctoral students’ participation in the collaborative process and better aligning the skills learned at the PhD level with the broader job market. First, align student training with job market realities. Second, dismantle systemic and structural barriers. Third, respond to the effects of social location. Last, negotiating power through consensus building. Recommendation for Researchers: Researchers should work toward creating an accepted framework for the collaborative program evaluation process to provide a road map of the journey instead of just providing the compass. Additionally, they should design and conduct a full-scale collaborative program evaluation to engage BIPOC students from the beginning. Finally, they should conduct a participatory-collaborative-inclusive academic program evaluation at different levels of higher education. Impact on Society: This article provides insight on how to improve the experience of BIPOC doctoral students’ participation in the collaborative process. It also demonstrates how collaborative learning assists in aligning the skills learned at the PhD level with the broader job market. Future Research: Future research should work toward creating an accepted framework for the collaborative program evaluation process to provide a road map of the journey instead of just providing the compass. Additionally, it should design and conduct a full-scale collaborative program evaluation to engage BIPOC students from the beginning. Finally, future research could conduct a participatory-collaborative-inclusive academic program evaluation at different levels of higher education.

Taking a byte out of big data

Nowadays a lot of massive data is stored and typically, the data itself contains a lot of non-trivial but useful information. Data mining techniques can be used to discover this information which can help the companies for decision-making. However, in real life applications, data is massive and is stored over distributed sites. One of my major research topics is to protect privacy over this kind of data. Previously, the important characteristics, issues and challenges related to management of the large amount of data has been explored. Various open source data analytics frameworks that deal with large amount of Data analytics workloads have been discussed. Comparative study between the given frameworks and suitability of the same has been proposed. Digital universe is flooded with huge amount of data generated by number of users worldwide. These data are of diverse in nature, come from various sources and in many forms. To keep with the desire to store and analyze ever larger volumes of complex data, relational databases vendors have delivered specialized analytical platforms that come in many shapes and sizes from software only to analytical services that run in third party hosted environments. In addition new technologies have emerged to address exploding volumes of complex data, including web traffic, social media content and machine generated data including sensor data, global positioning system data.Big data is defined as large amount of data which requires new technologies and architectures so that it becomes possible to extract value from it by capturing and analysis process. Due to such large size of data it becomes very difficult to perform effective analysis using the existing traditional techniques. Big data has become a prominent research field, especially when it comes to decision making and data analysis. However, Big data due to its various properties like volume, velocity, variety, variability, value and complexity put forward many challenges. Since Big data is a recent upcoming technology in the market which can bring huge benefits to the business organizations, it becomes necessary that various challenges and issues associated in bringing and adapting to this technology are brought into light. Another challenge is that data collection may not have enough accuracy which will lead to a non-consistent analysis that can critically affect the decision based on this analysis. Moreover, it is clearly apparent that organizations need to employ data-driven decision making to gain competitive advantage. Processing, integrating and interacting with more data should make it better data, providing both more panoramic and more granular views to aid strategic decision making. This is made possible via Big Data exploiting affordable and usable Computational and Storage Resources. Many offerings are based on the Map-Reduce and Hadoop paradigms and most focus solely on the analytical side. Nonetheless, in many respects it remains unclear what Big Data actually is; current offerings appear as isolated silos that are difficult to integrate and/or make it difficult to better utilize existing data and systems. Since data is growing at a huge speed making it difficult to handle such large amount of data (Exabyte). The main difficulty in handling such large amount of data is because that the volume is increasing rapidly in comparison to the computing resources. The Big data term which is being used now a days is kind of misnomer as it points out only the size of the data not putting too much of attention to its other existing properties.If data is to be used to make accurate decisions in time it becomes necessary that it should be available in accurate, complete and timely manner. This makes the data management and governance process bit complex adding the necessity to make Data open and make it available to government agencies in standardized manner with standardized APIs, metadata and formats thus leading to better decision making, business intelligence and productivity improvements.This paper presents a discussion and evaluation for the most prominent techniques used in the processes of data collection and analysis in order to identify the privacy defects in them that affects the accuracy of big data. Depending on the results of this analysis, recommendations were provided for improving data collection and analysis techniques that will help to avoid if not all then most of the problems facing the use of big data in decision making. Keywords: Big Data, Big Data Challenges, Big Data Accuracy, Big Data Collection, Big Data Analytics.

This paper studies how the integration of group awareness tools in the knowledge management system called KnowCat (Knowledge Catalyser), which promotes collaborative knowledge construction, may both foster the students' perception about the meaningfulness of visualization of group awareness information and promote better collaborative processes as well as enhance better task performance. Forty‐seven university students participated in a research study, where one group of 23 students used KnowCat without the awareness console (non‐awareness group); the other 24 students used KnowCat with the awareness console (awareness group). Both groups used KnowCat during one semester. Data analysis revealed that the awareness group means were higher than those of the non‐awareness group in terms of participation, cognitive and metacognitive learning activities, and task performance. Moreover, students revealed that knowing what, where and how much their classmates were contributing acted as positive feedback by encouraging participation and orienting their own behaviour and their contribution to the collaborative work. In this paper, we claim that the visualization group awareness information in KnowCat, a utility allowing students to visualize and track what, where, how much and how often other participants contributed to the KnowCat knowledge area, had a positive impact on the students' collaborative behaviour.

This paper presents the conclusions of a study on the impact of ICT on collaborative learning processes in Libyan Higher Education (LHE). The quantitative analysis of the answers to a questionnaire (completed by Libyan full-time lecturers at the universities of Tripoli, Garyounis, Gharian and Ezawia) shows the necessity to design and develop more classroom activities and interactive online applications, enabling the development of team-building skills required by employers. The influence of limited Internet bandwidths in Libya on collaborative learning processes in HE is then presented. It is obvious that HE institutions need to develop proactive strategies that envisage and anticipate learners future learning needs and requirements in this transition period of moving towards an increasingly digitalized, networked and knowledge-based society. The paper also contains the analysis of a SWOT model considering the factors that must be considered in relation to collaborative learning within the university teaching process, such as intelligent multimedia, Internet technologies, and knowledge management. The employment of modern technology will enable the development of innovative and inspiring collaborative learning environments where lecturers are expert designers of intellectual experiences for students, who become active participants to the learning processes.

Digital era produces vast amounts of data which leads to advances in data storage and analysis and the emergence of the concept of big data. Big data will be an important resource used in academic research and to make data driven decision making. This paper provides an overview about the big data concept and its application in libraries. This paper aims to create an awareness amongst librarians about big data, its applications and the role of librarian in big data scenario.

There is a exponential growth of data usage, specially due to the proliferation of connected applications with personalized models for different applications. In this context, Federated Learning (FL) emerges as a promising solution to enable collaborative model training while preserving the privacy and autonomy of participating clients. In a typical FL scenario, clients exhibit significant heterogeneity in terms of data distribution and hardware configurations. In this way, randomly sampling clients in each training round may not fully exploit the local updates from heterogeneous clients, resulting in lower model accuracy, slower convergence rate, degraded fairness, etc. In addition, malicious users could disseminate incorrect weights, which may decrease the accuracy of aggregated models and increase the time for convergence in FL. In this article, we introduce Resilience-aware Client Selection Mechanism for non-IID data and malicious clients in FL environment, called RICA. The proposed mechanism employs data size and entropy as criteria for client selection. In addition, RICA relies Centroid-Based Kernel Alignment (CKA) to identify and exclude potentially malicious clients. Our evaluation shows an improvement of 125% in Accuracy values in a scenario of malicious clients, which means the RICA+CKA demonstrates a more stable and resilient approach, reaching 90% accuracy in a few rounds compared to the default average approach, reached only around 30%. Therefore, results of the behavior of RICA+CKA in different datasets show the evaluation of different numbers of clients reaching around 90% while the other approach does not pass the 50% Accuracy.

With the proliferation of frequency-using devices and the advent of the era of big data, spectrum management and control are faced with challenges of effectiveness and accuracy. Modulation classification technology is the foundation and key part of spectrum management and control. Therefore, the effectiveness of modulation classification technology in big data scenario is very important. In this paper, we consider not only the validity of the classification model under the background of big data, but also the dynamics of noise in the complex electromagnetic environment. So we construct a big dataset containing different signals under different MSNR, and use the big data to drive the deep learning model, and finally get the classification result. The proposed method can realize modulation classification only by training one model, which avoids the redundancy of model training in previous algorithms. The simulation results demonstrate the effectiveness and reliability of the proposed method.

Purpose One of the major issues faced by academic health science centers (AHSCs) is the need for mechanisms to foster the integration of research, clinical, and educational activities to achieve the vision of evidence-informed decision making (EIDM) and optimal client care. The paper aims to discuss this issue. Design/methodology/approach This paper synthesizes literature on organizational learning and collaboration, evidence-informed organizational decision making, and learning-based organizations to derive insights concerning the nature of effective workplace learning in AHSCs. Findings An evidence-informed model of collaborative workplace learning is proposed to aid the alignment of research, clinical, and educational functions in AHSCs. The model articulates relationships among AHSC academic functions and sub-functions, cross-functional activities, and collaborative learning processes, emphasizing the importance of cross-functional activities in enhancing collaborative learning processes and optimizing EIDM and client care. Cross-functional activities involving clinicians, researchers, and educators are hypothesized to be a primary vehicle for integration, supported by a learning-oriented workplace culture. These activities are distinct from interprofessional teams, which are clinical in nature. Four collaborative learning processes are specified that are enhanced in cross-functional activities or teamwork: co-constructing meaning, co-learning, co-producing knowledge, and co-using knowledge. Practical implications The model provides an aspirational vision and insight into the importance of cross-functional activities in enhancing workplace learning. The paper discusses the conceptual and empirical basis to the model, its contributions and limitations, and implications for AHSCs. Originality/value The model's potential utility for health care is discussed, with implications for organizational culture and the promotion of cross-functional activities.

The data in the power Internet of Things has the characteristics of multiple types, large amounts of data, and obvious heterogeneity. In order to solve the storage and scheduling problem of massive heterogeneous data in the power grid, this paper designs a hybrid heterogeneous data storage scheduling architecture based on container technology. In order to solve the problem of sensor data analysis with unknown types and dynamic changes, this paper uses the K-Means clustering method to propose a data analysis process based on the idea of unsupervised learning to realize the analysis and adaptation of massive sensor data. In addition, in order to improve scheduling efficiency in big data scenarios, a two-layer scheduling architecture based on YARN and Kubernetes is proposed to achieve unified scheduling at the task level and node level, thereby improving cluster resource utilization. The heterogeneous data storage scheduling architecture proposed in this paper combines the container technology to meet the needs of massive data storage and processing for low latency and improved interrupt response rate and can be applied to power Internet of Things scenarios.KeywordsBig dataTask schedulingContainerKubernetesYARN

Decentralized storage powered by blockchain is becoming a new trend that allows data owners to outsource their data to remote storage resources offered by various storage providers. Unfortunately, unqualified storage providers easily encounter unpredictable downtime due to security threats, such as malicious attacks or system failures, which is unacceptable in many real-time or data-driven applications. As a result, continuous data integrity should be guaranteed in decentralized storage, which ensures that data is intact and available for the entire storage period. However, this requires frequent checking for long time periods and incurs heavy burdens of both communication and computation, especially in a big data scenario. In this paper, we propose an efficient continuous big data integrity checking approach for decentralized storage. We design a data-time sampling strategy that randomly checks the integrity of multiple files at each time slot with high checking probability. Furthermore, to tackle the fairness problem derived from the sampling strategy, we propose a fair approach by designing an arbitration algorithm with the verifiable random function. Security analysis shows the security of our approach under the random oracle model. Evaluation and experiments demonstrate that our approach is more efficient in the big data scenario compared with the state-of-the-arts.

Although artificial intelligence has become part of everyone’s real life, a trust crisis against such systems is occurring, thus increasing the need to explain black-box predictions, especially in the military, medical, and financial domains. Modern eXplainable Artificial Intelligence (XAI) techniques focus on benchmark datasets, but the cognitive applicability of such solutions under big data settings is still unclear due to memory or computation constraints. In this paper, we extend a model-agnostic XAI methodology, named Cluster-Aided Space Transformation for Local Explanation (CASTLE), to be able to deal with high-volume datasets. CASTLE aims to explain the black-box behavior of predictive models by combining both local (i.e., based on the input sample) and global (i.e., based on the whole scope for action of the model) information. In particular, the local explanation provides a rule-based explanation for the prediction of a target instance as well as the directions to update the likelihood of the predicted class. Our extension leverages modern big data technologies (e.g., Apache Spark) to handle the high volume, variety, and velocity of huge datasets. We have evaluated the framework on five datasets, in terms of temporal efficiency, explanation quality, and model significance. Our results indicate that the proposed approach retains the high-quality explanations associated with CASTLE while efficiently handling large datasets. Importantly, it exhibits a sub-linear, rather than exponential, dependence on dataset size, making it a scalable solution for massive datasets or in any big data scenario.

Sampling scheme-based classification rule mining method using decision tree in big data environment

Deep Learning Models for Multiple Face Mask Detection under a Complex Big Data Environment

The field of forestry has changed substantially in the last 100 years as scientists and managers have grappled with ways to best manage forests and adapt to changing knowledge, needs, and climates. We believe a path forward may be through using an ecosystem wellbeing framework where both community and environment wellbeing must be achieved to meet ecosystem wellbeing goals. To achieve this, we used a collaborative learning process where managers, researchers, tribes, and stakeholders engage with one another to ask and answer questions about options and effects of management choices. We also propose the use of the field of ethnoforestry, or a people-focused forest management, as the necessary way to apply the input, knowledge, and feedback gained through the collaborative learning process. We offer two examples of ways ethnoforestry is being tested on the Olympic Peninsula through an operational-scale experiment that seeks to meet the needs of communities and the environment, while producing revenue for trust land beneficiaries and meeting late seral habitat requirements.