COVID-19-like pandemics are a major threat to the global health system that causes a lot of deaths across ages. Large-scale medical images (i.e., X-rays, computed tomography (CT)) dataset is favored to the accuracy of deep learning (DL) in the screening of COVID-19-like pneumonia. The cost, time, and efforts for acquiring and annotating, for instance, large CT datasets make it impossible to obtain large numbers of samples from a single institution. The research attentions have been moved toward sharing medical images from numerous medical institutions. However, owing to the necessity to preserve the privacy of the data of a patient, it is challenging to build a centralized dataset from many institutions, especially during the pandemic. More. The difference in the data acquisition process from one institution to another brings another challenge known as distribution heterogeneity. This paper presents a novel federated learning framework, called Federated Multi-Site COVID-19 (FEDMSCOV), for efficient, generalizable, and privacy-preserved segmentation of COVID-19 infection from multi-site data. In FEDMSCOV, a novel is local drift smoothing (LDS) module encodes the input from feature space to frequency space, aiming to suppress the modules that are not conducive to generalization. Given the smoothed local updated, FEDMSCOV presents a novel Mixture-of-Expert (MoE) scheme to resolve global shift in parameters. An adapted differential privacy method is applied to design and protect the privacy of local updates during the training. Experimental evaluation on a large-scale multi-institutional COVID-19 dataset demonstrated the efficiency of the proposed framework over competing learning approaches with statistical significance.
Results of research into the use of fuzzy set theory for multicriteria decision making are presented. The application of a multicriteria approach is associated with the need to solve: 1) problems in which solution consequences cannot be estimated with a single criterion; 2) problems which can be solved on the basis of a single criterion, however their unique solutions are not achieved because the uncertainty of information produces decision uncertainty regions. In accordance with this, two classes of models are considered with using the Bellman-Zadeh approach and the technique of fuzzy preference relations, respectively, to their analysis. Advantages opened up by their application are discussed. The results are of a universal character, have found applications, and are illustrated by considering problems of power engineering.
The architecture for a hierarchical fuzzy neural attitude controller for satellites based on fuzzy gain scheduling is presented. A brief overview of fuzzy controller theory leading up to the creation of this new algorithm for attitude control is given. Inferencing based on decision tables for fuzzy controllers is made possible with the introduction of AND as well as OR fuzzy neurons. The construction of a hierarchical fuzzy neural attitude controller for satellites becomes the result of patching local (linear) controllers. The attitude control policy is then expressed as a series of "if-then" rules written in terms of local controller parameters and fuzzy relations used to invoke specific control actions. Each fuzzy relation is primarily correlated with the capabilities of a local linear model.
Multiagent systems are inherently associated with their distributivity, which enforces a great deal of communication mechanisms. To effectively arrive at meaningful solutions in a vast array of problem-solving tasks, it becomes imperative to establish a sound machinery of reconciling findings which might form partial solutions to an overall problem. In this paper, we focus on a broad category of problems of collaborative data analysis realized by a collection of agents having access to their individual data and exchanging findings through their collaboration activities. Such problems of data analysis arise in the context of building a global view at a certain phenomenon (process) by viewing it from different perspectives (and thus engaging various collections of attributes by various agents). Our goal is to develop some interaction between the agents so that they could form an overall perspective, where the knowledge available locally is shared and reconciled. The underlying format of knowledge built by the agents is that of information granules and fuzzy sets in particular. We develop a comprehensive optimization scheme and discuss its two-phase nature in which the communication phase of the granular findings intertwines with the local optimization being realized by the agents at the level of the individual datasite and exploits the evidence collected from other sites. We show how the mechanism of fuzzy granulation realized in the form of a well-known fuzzy c-means (FCM) clustering can be augmented to support collaborative activities required by the agents. For this purpose, we introduce augmented versions of the original objective function used in the FCM and derive algorithmic details. We also discuss an issue of optimizing the strength of collaborative linkages, so that the reconciled findings attain the highest level of consistency (agreement). The presented experimental studies include some synthetic data and selected data sets coming from the Machine Learning repository.
An important contribution of the Semantic Web is a new format of data representation called Resource Description Framework (RDF). In RDF, every piece of information is represented as a triple: subject-property-object. In general, subjects and objects can be shared between multiple RDF triples, and all triples can constitute a densely interlinked network. RDF becomes a very popular format of representing data on the web. As of September 2012, the last available data, more than 31 billions of triples exist on the web. In the paper, we propose a system - called T2R - for automatic acquisition of syntactic and semantic relations among terms from a plain text. These relations are expressed in the form of RDF triples. The proposed method is independent of any prior knowledge and domain specific patterns, and is applicable to any textual resources. The system implementing the approach is capable of identifying grammatical structure of an input sentence and analysing its semantics to generate meaningful RDF triples. We evaluate this approach by proving the quality of our results through case studies.
From the Publisher: A clear-cut, practical approach to software development! Emphasizing both the design and analysis of the technology, Peters and Pedrycz have written a comprehensive and complete text on a quantitative approach to software engineering. As you read the text, youll learn the software design practices that are standard practice in the industry today. Practical approaches to specifying, designing and testing software as well as the foundations of Software Engineering are also presented. Key Features *Thorough coverage is provided on the quantitative aspects of software Engineering including software measures, software quality, software costs and software reliability. *A complete case study allows students to trace the application of methods and practices in each chapter. *Examples found throughout the text are in C++ and Java. *A wide range of elementary and intermediate problems as well as more advanced research problems are available at the end of each chapter. *Students are given the opportunity to expand their horizons through frequent references to related web pages.
In this paper, a concept of fuzzy semientropy is proposed to quantify the downside uncertainty. Several properties of fuzzy semientropy are identified and interpreted. By quantifying the downside risk with the use of semientropy, two mean-semi-entropy portfolio selection models are formulated, and a fuzzy simulation-based genetic algorithm is designed to solve the models to optimality. We carry out comparative analyses among the fuzzy mean-entropy models and the fuzzy mean-semi-entropy models and demonstrate that the mean-semi-entropy models can significantly improve the dispersion of investment. Several illustrative examples using stock dataset from the real-world financial market (China Shanghai Stock Exchange) also show the effectiveness of the models.
The design and application of neural networks are examined in the context of a specific decision-making problem, the selection of an appropriate layout for a rural natural gas distribution system. The neural networks considered exploit one-stage classes of single-layer and multilayer architectures driven by logical constructs arising from the theory of fuzzy sets. The application is a multicriterion optimization problem. The neural networks' ability to model the decision-making process is illustrated with an example set of alternative layouts of rural natural gas systems.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
In this study, we present a new idea dealing with the analysis of fuzzy random variables (FRVs) being treated as samples of data. The proposed concept can be used to model various real-life situations where uncertainty is not only present in the form of randomness but also comes in the form of imprecision described in terms of fuzzy sets. We propose a hybrid approach, which combines a convex hull approach (called Beneath-Beyond algorithm) with a fuzzy random regression analysis. Falling under the umbrella of intelligent data analysis (IDA) tool, this approach is suitable for real-time implementation of data analysis. For a fuzzy random data set, we include simulation results and highlight two main advantages, namely a decrease of required analysis time and a reduction of computational complexity. This emphasizes that the proposed IDA approach becomes an efficient way for real-time data analysis.
