Search results

This study aims to summarize the critical issues in medical federated learning and applicable solutions. Also, detailed explanations of how federated learning techniques can be applied to the medical field are presented. About 80 reference studies described in the field were reviewed, and the federated learning framework currently being developed by the research team is provided. This paper will help researchers to build an actual medical federated learning environment.

Since machine learning techniques emerged, more efficient analysis was possible with a large amount of data. However, data regulations have been tightened worldwide, and the usage of centralized machine learning methods has become almost infeasible. Federated learning techniques have been introduced as a solution. Even with its powerful structural advantages, there still exist unsolved challenges in federated learning in a real medical data environment. This paper aims to summarize those by category and presents possible solutions.

This paper provides four critical categorized issues to be aware of when applying the federated learning technique to the actual medical data environment, then provides general guidelines for building a federated learning environment as a solution.

Existing studies have dealt with issues such as heterogeneity problems in the federated learning environment itself, but those were lacking on how these issues incur problems in actual working tasks. Therefore, this paper helps researchers understand the federated learning issues through examples of actual medical machine learning environments.

In response to food supply constraints resulting from coronavirus disease 2019 (COVID-19) restrictions, in the year 2020, the project developed automated household Aquaponics units to guarantee food self-sufficiency. However, the automated aquaponics solution did not fully comply with data privacy and portability best practices to protect the data of household owners. The purpose of this study is to develop a data privacy and portability layer on top of the previously developed automated Aquaponics units.

Design Science Research (DSR) is the research method implemented in this study.

General Data Protection and Privacy Regulations (GDPR)-inspired principles empowering data subjects including data minimisation, purpose limitation, storage limitation as well as integrity and confidentiality can be implemented in a federated learning (FL) architecture using Pinecone Matrix home servers and edge devices.

The literature reviewed for this study demonstrates that the GDPR right to data portability can have a positive impact on data protection by giving individuals more control over their own data. This is achieved by allowing data subjects to obtain their personal information from a data controller in a format that makes it simple to reuse it in another context and to transmit this information freely to any other data controller of their choice. Data portability is not strictly governed or enforced by data protection laws in the developing world, such as Zimbabwe's Data Protection Act of 2021.

Privacy requirements can be implemented in end-point technology such as smartphones, microcontrollers and single board computer clusters enabling data subjects to be incentivised whilst unlocking the value of their own data in the process fostering competition among data controllers and processors.

The use of end-to-end encryption with Matrix Pinecone on edge endpoints and fog servers, as well as the practical implementation of data portability, are currently not adequately covered in the literature. The study acts as a springboard for a future conversation on the topic.

Within digital innovation, there are two significant consequences of the pervasiveness of digital technology: (1) the increasing connectivity is enabling a wider reach and scope of innovation structures, such as innovation networks and (2) the unprecedented availability of digital data is creating new opportunities for innovation. Accordingly, there is a growing domain for studying data-driven innovation (DDI), especially in contemporary contexts of innovation networks. The purpose of this study is to explore how DDI processes take form in a specific type of innovation networks, namely federated networks.

A multiple case study design is applied in this paper. We draw our analysis from data collected over six months from four cases of DDI. The within-analysis is aimed at constructing the DDI process instance in each case, while the crosscase analysis focuses on pattern matching and cross-case synthesis of common and unique characteristics in the constructed processes.

Evidence from the crosscase analysis suggests that the widely accepted four-phase digital innovation process (including discovery, development, diffusion and post-diffusion) does not account for the explorative nature of data analytics and DDI. We propose an extended process comprising an explicit exploration phase before development, where refinement of the innovation concept and exploring social relationships are essential. Our analysis also suggests two modes of DDI: (1) asynchronous, i.e. data acquired before development and (2) synchronous, i.e. data acquired after (or during) development. We discuss the implications of these modes on the DDI process and the participants in the innovation network.

The paper proposes an extended version of the digital innovation process that is more specifically suited for DDI. We also provide an early explanation to the variation in DDI process complexities by highlighting the different modes of DDI processes. To the best of our knowledge, this is the first empirical investigation of DDI following the process from early stages of discovery till postdiffusion.

In federated search, a query is sent simultaneously to multiple resources and each one of them returns a list of results. These lists are merged into a single list using the results merging process. In this work, the authors apply machine learning methods for results merging in federated patent search. Even though several methods for results merging have been developed, none of them were tested on patent data nor considered several machine learning models. Thus, the authors experiment with state-of-the-art methods using patent data and they propose two new methods for results merging that use machine learning models.

The methods are based on a centralized index containing samples of documents from all the remote resources, and they implement machine learning models to estimate comparable scores for the documents retrieved by different resources. The authors examine the new methods in cooperative and uncooperative settings where document scores from the remote search engines are available and not, respectively. In uncooperative environments, they propose two methods for assigning document scores.

The effectiveness of the new results merging methods was measured against state-of-the-art models and found to be superior to them in many cases with significant improvements. The random forest model achieves the best results in comparison to all other models and presents new insights for the results merging problem.

In this article the authors prove that machine learning models can substitute other standard methods and models that used for results merging for many years. Our methods outperformed state-of-the-art estimation methods for results merging, and they proved that they are more effective for federated patent search.

This paper aims to study the relationship between employment and tourism activities as well as economic variables for the 32 states of Mexico for the period 1999-2014.

To study the case of Mexico, the authors use panel data and cointegration panel data. They also use geographic information systems to observe changes over time between the variables, which is useful in the empirical evidence.

The main results obtained by the models are as following: domestic tourism is the variable with the greatest impact on the generation of direct employment in the tourism sector, a finding supported by both methodologies; economic growth (measured by state gross domestic product) also directly impacts the generation of employment; and the cointegration of the panels causes a long-term equilibrium among the states and some variables.

The model used leaves out other variables that may influence the performance of the tourist activity. In addition, given the availability of official and homogeneous information, it only covers what has been documented up to 2014.

The aim is to measure the impact of tourism on the variables at the state level, where the economic activities could be based on public policies, as well as the importance of tourism activities in generating employment. In this sense, the impact would be in channeling efforts to support the main economic activities and could serve as a starting point for the evaluation of programs to promote domestic tourism.

This paper reviews the relationship that exists between tourism activity and its effect on other variables, especially employment. It is the first time that these topics are studied for the Mexican economy.

Recent advancement in technological development has encouraged distance learning institutions to be more productive and creative in effectively utilizing the Learning Management System (LMS). Among the many measures employed is the integration of federated search engine into the LMS which allows for a more productive and wider scope of information retrieval through the provisions of library resources and services. The purpose of this paper is to report one such case study in Wawasan Open University exploring the integration of federated search engine (EBSCO Discovery Service (EDS) widget) into the learning spaces of LMS. Widgets resemble apps that enable the integration of EDS functionality in providing access for students to retrieve library learning resources from the convenience of the LMS, excluding the need to log onto the library.

This paper presents a discussion that highlights the development and conjectural implementation of a framework on the integration of the EDS widget into the University’s LMS. Data collection includes meta-analysis data from the micro- and macro-level infrastructure that make up the framework, namely, end-user layer, system layer and data management layer.

Findings from this study addressed significant importance to the library in promoting effective search and utilization of information needs. The findings will also make clear recommendations in developing effective collaborations between the library and faculties. Although the implementation of this framework is still in a developmental stage, this study still provides pertinent information in validating the integration of EDS into the University’s LMS.

While serious limitations are not anticipated, possible concerns do exist with programming algorithms in the integration of EDS into the LMS. These challenges will be reported in the paper as reference for future replications of study

One key implication is the increase in the usage of the library resources and the potential to reach a larger audience of remote library users.

The primary advantage is to minimize the need for multiple gateway login while ensuring the library to monitor relevant library databases activities throughout the system check of the LMS.

This study advances a reconceptualization of data and information which overcomes normative understandings often contained in data policies at national and international levels. This study aims to propose a conceptual framework that moves beyond subject- and collective-centric normative understandings.

To do so, this study discusses the European Union (EU) and China’s approaches to data-driven technologies highlighting their similarities and differences when it comes to the vision underpinning how tech innovation is shaped.

Regardless of the different attention to the subject (the EU) and the collective (China), the normative understandings of technology by both actors remain trapped into a positivist approach that overlooks all that is not and cannot be turned into data, thus hindering the elaboration of a more holistic ecological thinking merging humans and technologies.

Revising the philosophical and political debate on data and data-driven technologies, a third way is elaborated, i.e. federated data as commons. This third way puts the subject as part by default of a collective at the centre of discussion. This framing can serve as the basis for elaborating sociotechnical alternatives when it comes to define and regulate the mash-up of humans and technology.

This paper aims to study the node modeling, multi-agent architecture and addressing method for the material conscious information network (MCIN), which is a large-scaled, open-styled, self-organized and ecological intelligent network of supply–demand relationships.

This study models the MCIN by node model definition, multi-agent architecture design and addressing method presentation.

The prototype of novel E-commerce platform based on the MCIN shows the effectiveness and soundness of the MCIN modeling. By comparing to current internet, the authors also find that the MCIN has the advantages of socialization, information integration, collective intelligence, traceability, high robustness, unification of producing and consuming, high scalability and decentralization.

Leveraging the dimensions of structure, character, knowledge and experience, a modeling approach of the basic information can fit all kinds of the MCIN nodes. With the double chain structure for both basic and supply–demand information, the MCIN nodes can be modeled comprehensively. The anima-desire-intention-based multi-agent architecture makes the federated agents of the MCIN nodes self-organized and intelligent. The MCIN nodes can be efficiently addressed by the supply–demand-oriented method. However, the implementation of the MCIN is still in process.

This paper lays the theoretical foundation for the future networked system of supply–demand relationship and the novel E-commerce platform.

The authors believe that the MCIN, first proposed in this paper, is a transformational innovation which facilitates the infrastructure of the future networked system of supply–demand relationship.

Medical research requires biological material and data collected through biobanks in reliable processes with quality assurance. Medical studies based on data with unknown or questionable quality are useless or even dangerous, as evidenced by recent examples of withdrawn studies. Medical data sets consist of highly sensitive personal data, which has to be protected carefully and is available for research only after the approval of ethics committees. The purpose of this research is to propose an architecture to support researchers to efficiently and effectively identify relevant collections of material and data with documented quality for their research projects while observing strict privacy rules.

Following a design science approach, this paper develops a conceptual model for capturing and relating metadata of medical data in biobanks to support medical research.

This study describes the landscape of biobanks as federated medical data lakes such as the collections of samples and their annotations in the European federation of biobanks (Biobanking and Biomolecular Resources Research Infrastructure – European Research Infrastructure Consortium, BBMRI-ERIC) and develops a conceptual model capturing schema information with quality annotation. This paper discusses the quality dimensions for data sets for medical research in-depth and proposes representations of both the metadata and data quality documentation with the aim to support researchers to effectively and efficiently identify suitable data sets for medical studies.

This novel conceptual model for metadata for medical data lakes has a unique focus on the high privacy requirements of the data sets contained in medical data lakes and also stands out in the detailed representation of data quality and metadata quality of medical data sets.

The purpose of this work is to bridge FL and blockchain technology through designing a blockchain-based smart agent system architecture and applying in FL. and blockchain technology through designing a blockchain-based smart agent system architecture and applying in FL. FL is an emerging collaborative machine learning technique that trains a model across multiple devices or servers holding private data samples without exchanging their data. The locally trained results are aggregated by a centralized server in a privacy-preserving way. However, there is an assumption where the centralized server is trustworthy, which is impractical. Fortunately, blockchain technology has opened a new era of data exchange among trustless strangers because of its decentralized architecture and cryptography-supported techniques.

In this study, the author proposes a novel design of a smart agent inspired by the smart contract concept. Specifically, based on the proposed smart agent, a fully decentralized, privacy-preserving and fair deep learning blockchain-FL framework is designed, where the agent network is consistent with the blockchain network and each smart agent is a participant in the FL task. During the whole training process, both the data and the model are not at the risk of leakage.

A demonstration of the proposed architecture is designed to train a neural network. Finally, the implementation of the proposed architecture is conducted in the Ethereum development, showing the effectiveness and applicability of the design.

The author aims to investigate the feasibility and practicality of linking the three areas together, namely, multi-agent system, FL and blockchain. A blockchain-FL framework, which is based on a smart agent system, has been proposed. The author has made several contributions to the state-of-the-art. First of all, a concrete design of a smart agent model is proposed, inspired by the smart contract concept in blockchain. The smart agent is autonomous and is able to disseminate, verify the information and execute the supported protocols. Based on the proposed smart agent model, a new architecture composed by these agents is formed, which is a blockchain network. Then, a fully decentralized, privacy-preserving and smart agent blockchain-FL framework has been proposed, where a smart agent acts as both a peer in a blockchain network and a participant in a FL task at the same time. Finally, a demonstration to train an artificial neural network is implemented to prove the effectiveness of the proposed framework.