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.

The purpose of this paper is to present how, in May 2008, the Ad Hoc Committee on Federated Search was formed to prepare a preliminary report on federated searching for a special meeting of Librarians Academic Council at Memorial University Libraries. The primary purpose is to discuss current implementation of federated searching at this institution, explore what other institutions have done, examine federated search technologies, and offer recommendations for the future of this resource.

Information was drawn from a recent usability study, an informal survey was created, and a literature/technology review was conducted.

These four recommendations were proposed and unanimously accepted: actively develop the current federated search implementation by developing a web presence supporting “federated search in context”, re‐evaluating the need for consortial purchase of a federated search tool, continuing to assess the current federated search marketplace with an eye to choosing a next‐generation federated search tool that includes effective de‐duping, sorting, relevancy, clustering and faceting, and that the selection, testing, and implementation of such a tool should involve broad participation from the Memorial University Libraries system.

Provided is an inside look at one institution's experience with implementing a federated search tool. The paper should be of interest to anyone working in academic libraries, particularly the areas of administration, public services, and systems.

Seeks to describe library federated search engines MetaLib and WebFeat as research tools by comparing MetaLib with WebFeat and by highlighting their strengths and weaknesses against Google and Google Scholar.

This study tested MetaLib and WebFeat from various libraries; attended vendor demos and asked vendors questions; reviewed literature; and interviewed system administrators of MetaLib and WebFeat.

MetaLib and WebFeat have fundamental differences between them. They cannot compete with Google in speed, simplicity, ease of use, and convenience, nor can they be truly one‐stop shopping. Their strengths lie in the contents they search as well as in the objective way they retrieve and display results. With the federated search engines, information literacy education is still relevant.

The comprehensive comparisons of MetaLib and WebFeat from the perspectives of both users and system administrators are original. It helps libraries make decisions when they select federated search engines, and it gives libraries realistic expectations of federated search engines compared with Google.

The purpose of this paper is to compare the efficacy of two approaches to finding articles on a topic by measuring student perceptions of the ease‐of‐use of the search process and the perceived relevancy of search results retrieved using both a single multidisciplinary database and a federated search tool.

Students are asked to search both a federated search tool and a single multidisciplinary database, record their searches and respond to a series of quantitative and qualitative questions regarding their experiences with searching both search tools.

Study results indicate a slight preference for federated searching over single database searching based on perceived relevancy of results and likeliness of future use. Study data supports equal promotion of single database searching and federated searching to undergraduate students.

The results of this paper have practical implications for reference and instruction librarians teaching undergraduate students and library users in general to find the most effective, efficient and manageable approach to finding articles on a topic.

Previous research comparing federated searching to other research methods uses prescribed topics outside of an actual class setting, while this is a naturalistic study in which students searched for articles on a research topic of their own choosing for a required research assignment. A previous study compared federated searching to navigating and searching numerous databases. This paper compares a federated search tool to a single multidisciplinary database.

As the storage and processing requirement of digital information is increasing on the cloud, it is very difficult for the single cloud provider (CP) to meet the resource requirement. Multiple providers form a federation for the execution of users’ requests. For the federated cloud, this paper aims to address the issue distribution of users’ request for resources and revenue among the providers by offering fair and stable distribution models for the federated cloud.

This paper uses cooperative game (CG)-theoretical models, i.e. Shapley–Shubik power index (SSPI) and Banzhaf power index (BPI) for distribution. Performance is analysed using variance and monotonicity using a case study.

Numerical analysis is done using two scenarios. Monotonicity is evaluated. Results show that SSPI performs better as compared to BPI in terms of fairness accuracy and the framework provide the fair distribution of revenue among providers in the federated cloud.

The proposed framework works efficiently under the specific defined conditions.

Paper provides the fair distribution. It assist the centralised cloud exchange in managing the users’ request in such a way every CPs, in the federated cloud will get an equal chance of serving the users’ request. The framework also provides the stable federation. Proposed work provides less rejection rate of users’ request. Finally, it assists the providers in increasing their profits in the federation.

This paper presents a CG theoretic-based framework for the distribution of resources required and revenue. The framework analysed the performance of distribution models by considering the variance and monotonicity for multiple users’ requests.

In the digital age, organizations want to build a more powerful machine learning model that can serve the increasing needs of people. However, enhancing privacy and data security is one of the challenges for machine learning models, especially in federated learning. Parties want to collaborate with each other to build a better model, but they do not want to reveal their own data. This study aims to introduce threats and defenses to privacy leaks in the collaborative learning model.

In the collaborative model, the attacker was the central server or a participant. In this study, the attacker is on the side of the participant, who is “honest but curious.” Attack experiments are on the participant’s side, who performs two tasks: one is to train the collaborative learning model; the second task is to build a generative adversarial networks (GANs) model, which will perform the attack to infer more information received from the central server. There are three typical types of attacks: white box, black box without auxiliary information and black box with auxiliary information. The experimental environment is set up by PyTorch on Google Colab platform running on graphics processing unit with labeled faces in the wild and Canadian Institute For Advanced Research-10 data sets.

The paper assumes that the privacy leakage attack resides on the participant’s side, and the information in the parameter server contains too much knowledge to train a collaborative machine learning model. This study compares the success level of inference attack from model parameters based on GAN models. There are three GAN models, which are used in this method: condition GAN, control GAN and Wasserstein generative adversarial networks (WGAN). Of these three models, the WGAN model has proven to obtain the highest stability.

The concern about privacy and security for machine learning models are more important, especially for collaborative learning. The paper has contributed experimentally to private attack on the participant side in the collaborative learning model.

The federated filter created by Carlson has been widely used in multi-sensor integrated navigation. Compared with no-reset federated filter, the reset mode has greater sub-filters’ performance, but faults of any subsystem would affect other healthy subsystems via global fusion and the sub-optimality of sub-filters’ estimation has influence on fault detection sensitivity. It’s a challenge to design a robust reset federated filter.

The time-varying observation noise is designed to reduce proportions of observation information in faulty sub-filters. A new dynamic information distribution algorithm based on optimal residual chi-square detection function is presented to reduce proportions of faulty sub-filters’ estimation in information fusion filter.

The robust filtering algorithm represents a filtering strategy for reset federated filter. Compared with fault isolation, the navigation result is smoother by using this algorithm. It has significant benefits in avoiding faulty sensors’ contamination and the performance of federated filter is greatly improved.

The approach described in this paper provides a new method to deal with federated reset filter’s faulty problems. This new robust federated filter algorithm possesses a great potential for various applications.

The approach described in this paper can be used in multi-sensor integrated navigation with no fewer than three sensors.

Compared with conventional approach of fault isolation, the proposed algorithm does not destroy the continuity and integrity of the filtering process. It improves the performance of the federated filter by reducing proportions of faulty observation information. It also reduces the influence of sub-optimality on fault detection sensitivity.

This paper aims to shed light on federated search engine solutions by documenting how Westminster College implemented a new system after a failed attempt.

The librarians became interested in a new federated search solution after a previous vendor was unable to provide a satisfactory product. After seeing demonstrations from several vendors, the librarians decided on WebFeat. This paper chronicles that decision and its implementation, which surprisingly took less than three months.

The librarians found WebFeat to be a superior product in comparison with the first federated search engine. WebFeat made an immediate impact on how the students and faculty searched for information. The seamless integration into the library web page made it easier for users to search databases and the library catalog without confusion and frustration.

The value of this paper comes from the fact that the library became a “federated search casualty” after adopting a system in 2003 that did not live up to expectation and eventually had to be terminated. WebFeat was the next choice, and it has performed above expectations. The librarians witnessed the failure of one federated search engine solution and the success of another over the past three years.

This paper aims to examine the behaviour of the Egyptian scholars while accessing electronic resources through two federated search tools. The main purpose of this article is to provide guidance for federated search tool technicians and support teams about user issues, including the need for training.

Log files were exploited to examine the behaviour of users of information retrieval systems. This study examined two log files extracted from federated search tools available to the Egyptian scholars' community for accessing electronic resources. A data mining approach was implemented to investigate user behaviour through deep analysis of these logs.

Results show that: none of the available tools provide error messages for dummy queries; most of the Egyptian scholars had short queries; Boolean operators are not used in about 50 per cent of the queries; federated search tools do not provide techniques for query reformation; the optimal days for system maintenance are the non‐weekend vacations; and early morning is the best time for maintenance.

To maximise the value of the federated search tools by understanding user trends when utilising federated search tools. The study shows that more attention should be given to the search capabilities through ongoing training and awareness in order to maximise the benefit from the available resources and tools.

The hypothetical value of the federated search tools has not been previously examined and analysed to understand user trends.

Purpose –  – To explore the searchability and search results obtained from the Central Search federated search engine product.Design/methodology/approach – Common keywords are used to test how Central Search processes the search query and how it categorizes search results by subject.Findings – Central Search offers a wide number of results but is not as effective in sorting results by subject.Research limitations/implications – The literature on federated searching is limited because of the products have only recently been released. More study is necessary to determine how effective Central Search and related federated search engines will be for more refined, subject‐specific searches.Practical implications – This assessment emphasizes some of the problematical aspects of federated searching in an academic library environment.Originality/value – Responds to the need for a practical demonstration of how federated search products such as Central Search actually perform searches and how they processes the myriad results they retrieve.