Stefania Tomasiello

University of Salerno | UNISA · Department of Industrial Engineering (DIIn)

In this paper, we shed light on the use of two types of fuzzy genetic algorithms, which stand out from the literature due to the innovative ideas behind them. One is the Gendered Fuzzy Genetic Algorithm, where the crossover mechanism is regulated by the gender and the age of the population to generate offspring through proper fuzzy rules. The other...

As neural network-based localization algorithms are becoming popular, there is a need to shorten the training time and the localization time for sustainability and efficiency purposes. To address such issues, the fuzzy transform (or F-transform for short) is employed here for the first time in a neural network-based localization algorithm. The F-tr...

In an era where image manipulation is easily accessible, detecting digital image forgery has become more challenging. The manuscript delves into the essential issue of picture forgery and discusses the use of the standard Adaptive Neuro-Fuzzy Inference System (ANFIS) and some of its variants for identifying and analyzing altered digital image conte...

The International Conference on Mathematical Analysis and Applications in Science and Engineering –ICMA2SC’24 will take place at the beautiful city of Porto, Portugal, in June 20th-June 22nd 2024. This conference is dedicated to the memory of Prof JA Tenreiro Machado, who passed away in October 2021. Its aim is to bring together researchers in ev...

This work aims at providing a concise review of various agri-food models that employ fractional differential operators. In this context, various mathematical models based on fractional differential equations have been used to describe a wide range of problems in agri-food. As a result of this review, we found out that this new area of research is f...

As a common generalization of intuitionistic fuzzy sets (IFSs), Pythagorean fuzzy sets (PFSs) and Fermatean fuzzy sets (FFSs), generalized orthopair fuzzy sets (GOFSs) have received worldwide attention in past few years. These extended fuzzy sets are all built on the basis that the information described by the membership grade (MG) and non-membersh...

The differential quadrature method is a well-known numerical approach for solving ordinary and partial differential equations. This work introduces an explicit form for the approximate solution using differential quadrature rules. Analogies with Taylor's expansion are presented. Some properties are formally discussed. An interpretation of the appro...

The proliferation of fake news has raised concerns regarding its detection, posing a significant challenge. Motivated by the ongoing discussion on the sustainability of machine learning algorithms, this paper discusses the usefulness of data reduction for fake news detection. This is accomplished by using the fuzzy transform (or F-transform for sho...

Artificial Intelligence (AI) is increasingly pervading everyday life since it can be used to solve high-complexity problems, as well as determine optimal solutions, in various domains and for numerous applications [...]

Motivated by the increasing interest in machine learning algorithms for data-driven applications in agri-food addressing sustainability issues and by the ongoing discussion on the interpretability and sustainability of such algorithms, we compare congruently the performance of some state-of-the-art techniques and a new version (here proposed for th...

The proliferation of fake news has raised concerns regarding its detection, posing a significant challenge. Motivated by the ongoing discussion on the sustainability of machine learning algorithms, this paper discusses the use- fulness of data reduction for fake news detection. This is accomplished by using the Principal Component Analysis (PCA) wh...

Granular computing, an attractive branch of artificial intelligence, focuses on constructing, processing and communicating information granules. Although various useful structures involving fuzzy sets, rough sets and their extensions have been discussed in relation to this literature, there is still a research gap regarding the connections between...

In this short note, the performance of two kinds of physics-guided computing schemes, namely the Hamiltonian Neural Network and the Port-Hamiltonian Neural Network, are discussed through the predicted dynamics of two coupled Duffing oscillators. First, we propose a new error bound which holds for both types of networks. Then, we numerically investi...

In this short paper, we compare for the first time the performance of some approaches for evapotranspiration prediction, in terms of interpretability, accuracy and training time. The considered techniques are Decision Trees (DTs) and the Adaptive Network-based Fuzzy Inference System with fractional Tikhonov regularization (ANFIS-T), which are known...

The sustainable supply chains optimization is a high-dimensional multi-objective optimization problem. The involved costs can be categorized as economic, environmental, and social. Metaheuristics can be used for tackling this kind of problem efficiently. This short note deals with a comparative analysis of the main metaheuristics (according to rece...

In this article, we introduce a variant of the adaptive network-based fuzzy inference system (ANFIS). The proposed variant does not use backpropagation and grid partitioning, but the least-squares method with fractional Tikhonov regularization. The fractional regularization is a generalization of the standard regularization and is applied here to t...

We investigate the adverse effect of noisy labels in a training dataset on a neural network’s precision in an image classification task. The importance of this research lies in the fact that most datasets include noisy labels. To reduce the impact of noisy labels, we propose to extend the binary cross-entropy by dynamical clipping, which clips all...

This chapter is devoted to fuzzy relations. Fuzzy relations on classical and fuzzy sets are discussed. Composition of fuzzy relations are presented. Fuzzy relational equations are also introduced.

This chapter starts off by briefly recalling Artificial Neural Networks. A general Fuzzy Neural Netowrk scheme is discussed. The Adaptive Neuro-Fuzzy Inference System (ANFIS) and one of its variants are presented.

In this chapter, the direct and inverse fuzzy transform is presented. Its application to data compression is discussed. Its use as a reduction technique to preprocess data presented to machine learning algorithms is also discussed.

The general structure of fuzzy inference systems is introduced in this chapter. The Mamdani and Takagi-Sugeno-Kang systems are discussed.

In this chapter, the fuzzy logic basic connectives are introduced. Linguistic variables are also presented, along with linguistic modifiers. Fuzzy propositions, modus ponens and generalized modus ponens are discussed.

This chapter starts off with the Zadeh’s extension principle. The concept of fuzzy number is then introduced, by also presenting its parametric form. Different types of fuzzy numbers are discussed. Concepts from interval mathematics are briefly recalled. Methods for performing arithmetic operations with fuzzy numbers are illustrated.

In this chapter, the basic notions of granular computing are presented. The concept of information granule is briefly illustrated. Selected models of granular neural networks are discussed, by also providing an application example.

This chapter is devoted to introduce fuzzy sets and the main related notions. The operations on fuzzy sets are also presented. Different types of fuzzy sets are briefly recalled.

In this paper, we discuss a variant of the Adaptive Neuro-Fuzzy Inference System (ANFIS) to predict two performance attributes, i.e. total cost to serve and return on working capital, following the Supply Chain Operations Reference (SCOR) model on the basis of the state of the art, This variant is based on fractional Tikhonov regularization, a kind...

We present a computing scheme as a variant of a recently proposed granular recurrent neural network. Being deduced from a generic system of partial differential equations, this variant is able to capture the spatiotemporal variability of some datasets and problems. The convergence of the computing scheme has been formally discussed. Some preliminar...

This special issue was conceived to explore the latest advancements in the field of computational techniques for solving forward and inverse problems [...]

A survey on the technologies employed in the modern agriculture and agri-food supply chains lately appeared, but only one paper using a fuzzy-based approach was cited. The aim of the present mini-review is to complement the above-mentioned survey and to show the application of different fuzzy-based approaches for agri-food supply chains. Agri-food...

Link prediction is an emerging and fast-growing applied research area. In a network, it is possible to predict the next link which is going to be formed. The usefulness of link prediction modeling has been proved in several fields and applications, such as biomedicine, recommending systems, and social media. In this short paper, we discuss the pote...

In this paper, we consider supply chains modelled by partial and ordinary differential equations, for densities of parts on suppliers and queues among consecutive arcs, respectively. The considered numerical schemes, whose properties are discussed, foresee the upwind method for goods densities, described by conservation laws, and a Differential Qua...

In this note, a numerical algorithm which combines fuzzy transform and Picard like recursions is adapted for solving weakly singular Volterra integral equations (WSVIs) with proportional delay. Under certain hypothesis, the proposed method provides a non-recursive approximate solution based on operational matrices and vector of unknown quantities....

In this work, an efficient algorithm is proposed for solving the system of Volterra integral equations based on wavelet Galerkin method. This problem is reduced to a set of algebraic equations using the operational matrix of integration and wavelet transform matrix. For linear type, the computational effort decreases by thresholding. The convergenc...

In this short paper, analogies and differences between a type of fuzzy transform and a type of autoencoder, both based on a least-squares optimization, will be discussed. Such schemes have been recently introduced in the literature in different contexts. In particular, here the data compression application will be considered. As it will be shown, t...

In this paper, we discuss a new kind of stability, that is finite-time stability, for uncertain differential equations, by formalizing some properties. As a possible application, we define a new class of uncertain multi-agent systems, according to the Liu's uncertainty theory, as a counterpart of stochastic multi-agent systems. We formalize the gov...

In this paper, we formally discuss a computational scheme, which combines a local weighted regression model with fuzzy transform (or F-transform for short). The latter acts as a reduction technique on the cardinality of the learning problem, resulting in a more efficient algorithm. We tested the proposed approach first through two typical benchmark...

In this paper, we present a new granular classifier in two versions (iterative and non–iterative), by adopting some ideas originating from a kind of Functional Link Artificial Neural Network and the Functional Network schemes. These two architectures are substantially the same: they both use a function basis instead of the usual activation function...

The paper proposes an extension of stability analysis methods for a class of impulsive reaction–diffusion Cohen–Grossberg delayed neural networks by addressing a challenge namely stability of sets. Such extended concept is of considerable interest to numerous systems capable of approaching not only one equilibrium state. Results on uniform global a...

Fuzzy transform is a relatively recent fuzzy approximation method, mainly used for image and general data processing. Due to the growing interest in the application of fuzzy transform over the last years, it seems proper providing a review of the technique. In this paper, we recall F-transform-based compression methods for data and images. The rela...

In this paper, we present and investigate the analytical properties of a new set of orthogonal basis functions derived from the block-pulse functions. Also, we present a numerical method based on this new class of functions to solve nonlinear Volterra-Fredholm integral equations. In particular, an alternative and efficient method based on the forma...

The development of interpretable and readable diagnosis support models in the medical field is becoming an active research area. Neuroimaging technology has been widely used in the study of various brain diseases, supported by several kind of machine learning algorithms. Such algorithms, in spite of their accuracy, have often a lack of interpretabi...

In this paper, we formally deduce a new computational model, with a recurrent structure, by means of data granulation. The proposed scheme can be regarded as an Echo State Network (ESN), with an additional granular layer. ESNs have been recently revisited in the context of deep learning. In view of such a state-of-the-art, and coherently with the c...

Discrete nonlinear systems have become increasingly important in the past decades. To understand them, new analytical and numerical methodologies have been proposed. More accurate predictions of solutions of the systems related physical problems are investigated. Qualitative and quantitative features of discrete nonlinear modeling problems and thei...

Based on the nonloocal elasticity theory, this paper deals with the dynamic instability analysis of cantilevered single-walled carbon nanotube with concentrated mass, located at a generic position, and subject to a follower force at the free end. Accounting for the small scale effect, the governing equations of motion are derived using an alternati...

In this paper, we propose a general scheme of Functional Network, by considering granularity of information and time delay. Functional Networks (FNs) are a relatively recent alternative to standard Neural Networks (NNs). They have shown better performance in comparison to performance of NNs. Data granulation used in the development of NNs allows fo...

Purpose This paper aims to capture the effective behaviour of nonlinear coupled advection-diffusion-reaction systems and develop a new computational scheme based on differential quadrature method. Design/methodology/approach The developed scheme converts the coupled system into a system of ordinary differential equations. Finally, the obtained sys...

Editorial note.

We present a decision support system for seamless and self-regulating learning. The decision support system presents a degree of novelty in supporting learners since it allows to: (1) understand the concepts that a learner may have acquired during her/his daily life activities, and (2) make the learner aware of these concepts and enforcing learning...

Purpose The purpose of the present paper is to investigate the nonconservative instability of a single-walled carbon nanotube (SWCNT) with an added mass through nonlocal theories. The governing equations are discretized by means of the differential quadrature (DQ) rules, as introduced by Bellman and Casti. DQ rules have been largely used in engine...

Ubiquitous recommender systems facilitate users on{location by personalized rec- ommendations of items in the proximity via mobile devices. Due to a high variability of situations and preferences, an efficient resource processing is needed in order to assist the user in a proper way. In this paper, we consider a recommender system, able to learn pr...

Antimirov’s partial derivatives are used in classical automata theory for the conversion of regular expressions to finite automata, tree regular expressions to tree automata, and ω−regular expressions to Büchi automata. In this paper, we describe a new variant of the Antimirov’s partial derivatives for the conversion of fuzzy regular expressions to...

In this study, we propose a fuzzy-based approach aiming at finding numerical solutions to some classical problems. We use the technique of fuzzy transform to solve a second-order ordinary differential equation with boundary conditions. To determine the optimal coefficients in expansion of a function, we minimize the integral squared error in 2-norm...

Fuzzy transform is a promising approximation technique. In this paper, we use it to approximate a delayed function in a two-neuron system. We formally study the effect of this approximation, by elucidating the dynamics of the resulting discrete system. A linear stability analysis has been performed, with a first investigation on the possible bifurc...

In this paper, we propose a computational scheme for the problem of wind power forecasting. Such scheme combines a local weighted regression model with fuzzy transform. The latter provides a way to reduce the cardinality of the learning problem, resulting in a more efficient algorithm. Numerical examples show the effectiveness of the proposed appro...

In the last years, many papers on Granular Neural Networks appeared, aiming at achieving a higher degree of transparency in the network architecture and a better understanding of the involved steps. This paper is a first attempt to emphasize and exploit the notion of granularity into Functional Networks. Functional Networks are a relatively recent...

Function reconstruction is one of the valuable properties of the Fuzzy Transform and its inverse. The quality of reconstruction depends on how dense the fuzzy partition of the function domain is. However, the partition should be denser where the function exhibits faster variations, while the partition can be less dense where the function is moving...

In this paper, we discuss the main scientific aspects of a Multi-Agent System (MAS), which was designed for monitoring Smart Grids (SGs) with assessment of optimal settings obtained through approximate Optimal Power Flow (OPF) solutions. The consideration behind the approach is that large historical operation datasets are usually available in SGs a...

In this paper, a numerical approach for the simulation of a dynamical model with damping defined by the Riemann–Liouville fractional derivative and with uncertainty, that is fuzziness, is discussed. The proposed method exploits differential quadrature rules and a Picard– like recursion. The convergence is formally discussed. Some example applicatio...

In this paper, a fuzzy based paradigm for data compression aimed at reducing the computational burden of data analysis in smart grids is proposed. In the smart grid context, it is challeging achieving an efficient use of the channel communication bandwidth and a reduced need of the storage space for operational data. Thus, we discuss a fuzzy based...

In this work, we propose an exponential-type discretization of the well-known Fisher's equation from population dynamics. Only non-negative, bounded and monotone solutions are physically relevant in this note, and the discretization that we provide is able to preserve these properties. The method is a modified explicit exponential technique which h...

A Picard-like approach which has been used to solve a class of Volterra integro-differential equations, is extended in this manuscript to solve fuzzy fractional differential equations. Such technique uses quadrature rules and Picard's iterations in the fuzzy context. In spite of this, it is conceived to become a non-recursive scheme, in terms of op...

Departing from a numerical method designed to solve ordinary differential equations, in this manuscript we extend such approach to solve problems involving fuzzy partial differential equations. The method proposed in this work is a non-recursive technique that combines differential quadrature rules and a Picard-like scheme in order to obtain genera...

In this paper, the problem of second{order consensus in multi-agent systems with sampled position data is handled. Due to the discrete nature of the information transmission among agents, there is an increasing interest in such kind of systems. By thinking of collecting sampled position data in a certain time interval, that is more general than con...

In this paper, a numerical scheme for a parameter identification problem is presented. The problem here considered is the identification of the stiffness of structural elements and a new procedure to solve it is proposed. This procedure involves not only the usual Newton{like iterative algorithm for nonlinear least squares problems, but it also pro...

The paper focuses on simulation results for a real supply network, dealing with tomatoes production, which is widespread in Southern Italy. The dynamics of the system is studied via differential equations, that involve either parts on arcs or queues that consider the exceeding goods. Two different numerical schemes are compared to test the approxim...

Fuzzy transforms (or F-transforms for short) are an approximation technique recently introduced. The main application is referred to image and data compression. There are really few works devoted to the use of F-transform for solving ordinary differential equations. In the present paper, an F-transform based Picard-like numerical scheme is proposed...

Events that deal with Group Decision Making are continuously studied in order to provide a suitable representation of different opinions, with the aim of reaching the consensus of all experts involved in decision processes. In this paper, the authors, focusing on employees’ evaluations inside Italian companies, propose an extension of a fuzzy conse...

In this paper, we present our results related to the definition of a methodology that combines augmented reality (AR) with semantic techniques for the creation of digital stories associated with museum exhibitions. In contrast to traditional AR approaches, we augment real-world elements by supplementing contents of a museum exhibition with addition...

In this paper a new offline model-free approximate Q-iteration is proposed. Following the idea of Fitted Q-iteration, we use a computational scheme based on Functional Networks, which have been proved to be a powerful alternative to Neural Networks, because they do not require a large number of training samples. We state a condition for the converg...

In this paper, we discuss a numerical approach for the simulation of a model for supply chains based on both ordinary and partial differential equations. Such methodology foresees differential quadrature rules and a Picard--like recursion. In its former version, it was proposed for the solution of ordinary differential equations and is here extende...

Joining data compression and encryption is a way to keep secure data, as discussed by the current literature. While data compression responds to the great demand on data storage and transmission techniques, the encryption allows to handle some important parameters in a secure way. In wireless sensor networks the usual transform–based compression is...

Governance can be considered as the way in which companies are directed and controlled. In any governance process, effective experience reuse is an increasingly important asset, representing source of competitive advantage in making decisions. As an example, an adequate level of experience is needed for taking decisions related to resource allocati...

A Picard-like scheme using quadrature and differential quadrature rules, formerly introduced to solve integro-differential equations, is herein adapted to solve the problem of an oscillator with damping defined by the Riemann-Liouville fractional derivative and with fuzzy initial conditions. Considering fuzzy initial conditions has the meaning of a...

Handling the continuity of learning experience across different activities and contexts is a key challenge for seamless learning. Current context and activity recognition techniques work well in fixed environments where sensors deployment and data are known but are not adaptable to dynamic and changing situations when, for instance, a learner moves...

Purpose- Within the Enterprise contexts, one of the most important aspects is the realization of intelligent services (inspired by the paradigm Service-Dominant Logic, SDL), which obey the wishes and constraints of final users. To achieve this aim in an efficient way, various efforts are considered, from technological, economical and modeling point...

In wireless sensor networks a large amount of data is collected for each node. The challenge of transferring these data to a sink, because of energy constraints, requires suitable techniques such as data compression. Transform-based compression, e.g. Discrete Wavelet Transform (DWT), are very popular in this field. These methods behave well enough...

This note introduces a new computational scheme based on a numerical Picard-like method, which combines successive approximations with differential quadrature rules in order to improve the computational efficiency. Despite of its recursive nature, this scheme can generate the approximate solution in terms of operational matrices and vectors of know...

The main barrier to a mainstream adoption of Semantic Web and Linked Data is the difficulty for users to search and retrieve the required information in this huge network of data. This work proposes a novel approach for Ubiquitous Browsing and Searching Linked Data. The proposed approach lays on a conceptual communication model, namely Interactive...

This paper introduces Functional Networks in a fuzzy environment. We formally define a Fuzzy Functional Network for application in some regression problems and prove a sufficient condition on the solution of our regression model. Several numerical examples, based on simulated data and literature data, show the good performance of our approach: solu...

A numerical Picard–like method, which combines successive approximations with differential quadrature, is discussed for application to the stability of columns. Unlike other methods, the present method satisfies all the boundary conditions without any additional computation for each iteration and it does not require any subdivision of the problem d...

This paper discusses a new computational scheme based on functional networks and applies it to the problem of classification and quantification of gas species in a mixture. A generalized functional network as a new classifier is proposed to improve the potentialities of the standard functional network classifier. Both methodology and learning algor...

In this paper an alternative approach for identification problems is discussed. Unlike existing methods, this new approach combines in a general way finite differences and function approximation and is herein used for the identification of a particular system in structural dynamics, that is the damped Duffing oscillator subject to a swept-sine exci...

Functional networks (FNs) are a promising numerical scheme that produces accurate solutions for several problems in science and engineering with less computational effort than other conventional numerical techniques such as neural networks. By using domain knowledge in addition to data knowledge, functional networks can be regarded as a generalizat...

In this paper, a new DQ-based compact step-by-step integration method is proposed. Analytical proof of stability is presented. The method is unconditionally stable and not affected by algorithmic damping. Besides, sixth-order convergence can be achieved. A classical nonlinear model is studied as example application. Compared to other similar proced...

In this paper a numerical Picard-like method, which combines successive approximations with integral and differential quadrature, is considered to solve some Volterra integro-differential equations where rational functions are involved. Under certain conditions, the method provides solutions in explicit form, without any recursive computation, and...

In this paper, three numerical methods to solve Volterra integro-differential equations containing rational functions are discussed. The first one is the Differential Quadrature Method and, to the best knowledge of the author, it has never been applied to this kind of problem; the second one is a new version of the Iterative Differential Quadrature...

In this paper, the numerical stability of an iterative method based on differential quadrature (DQ) rules when applied to solve a two-dimensional (2D) wave problem is discussed. The physical model of a vibrating membrane, with different initial conditions, is considered. The stability analysis is performed by the matrix method generalized for a 2D...