Informatica

Structural break detection is an important time series analysis task. It can be treated as a multi-objective optimization problem, in which we ought to find a time series segmentation such that time series theoretical models constructed on each segment are well-fitted and the segments are long enough to bear meaningful information. Metaheuristic optimization can help us solve this problem. This paper introduces a suite of new cost functions for the structural break detection task. We demonstrate that the new cost functions allow for achieving quantitatively better precision than the cost functions employed in the literature of this domain. We show particular advantages of each new cost function. Furthermore, the paper promotes the use of Particle Swarm Optimization (PSO) in the domain of structural break detection, which so far has relied on the Genetic Algorithm (GA). Our experiments show that PSO outperforms GA for many analysed time series examples. Last but not least, we introduce a non-trivial generalization of the top-performing state-of-the-art approach to the structural break detection problem based on the Minimum Description Length (MDL) rule with autoregressive (AR) model to MDL ARIMA (autoregressive integrated moving average) model.

As musculoskeletal illnesses continue to increase, practical computerised muscle modelling is crucial. This paper addresses this concern by proposing a mathematical model for a dynamic 3D geometrical surface representation of muscles using a Radial Basis Function (RBF) approximation technique. The objective is to obtain a smoother surface while minimising data use, contrasting it from classical polygonal (e.g. triangular) surface mesh models or volumetric (e.g. tetrahedral) mesh models. The paper uses RBF implicit surface description to describe static surface generation and dynamic surface deformations based on its spatial curvature preservation during the deformation. The novel method is tested on multiple data sets, and the experiments show promising results according to the introduced metrics.

Focusing on the problems of failing to make full use of spatial context information and limited local receptive field when U-Net is utilized to solve MRI brain tumour segmentation, a novel 3D multi-scale attention U-Net method, i.e. MAU-Net, is proposed in this paper. Firstly, a Mixed Depth-wise Convolution (MDConv) module is introduced in the encoder and decoder, which leverages various convolution kernels to extract the multi-scale features of brain tumour images, and effectively strengthens the feature expression of the brain tumour lesion region in the up and down sampling. Secondly, a Context Pyramid Module (CPM) combining multi-scale and attention is embedded in the skip connection position to achieve the combination of local feature enhancement at multi-scale with global feature correlation. Finally, MAU-Net adopts Self-ensemble in the decoding process to achieve complementary detailed features of sampled brain tumour images at different scales, thereby further improving segmentation performance. Ablation and comparison experiment results on the publicly available BraTS 2019/2020 datasets well validate its effectiveness. It respectively achieves the Dice Similarity Coefficients (DSC) of 90.6%/90.2%, 82.7%/82.8%, and 77.9%/78.5% on the whole tumour (WT), tumour core (TC) and enhanced tumour (ET) segmentation. Additionally, on the BraTS 2021 training set, the DSC for WT, TC, and ET reached 93.7%, 93.2%, and 88.9%, respectively.

This paper presents a multiple-criteria choice model, the circular intuitionistic fuzzy (C-IF) ELECTRE, designed to resolve C-IF ambiguities through built-in circular functions. Joint generalized scoring functions establish contrast relationships between C-IF evaluation values, facilitating concordance and discordance analyses for option ranking. The efficacy of C-IF ELECTRE I and II—leveraging tools such as the prioritization Boolean matrix, average outflows and inflows, and overall net flow—is validated through a multi-expert supplier evaluation, with outcomes benchmarked against alternative methods. A comparative analysis explores the impact of parameter variations, underscoring how integrating C-IF sets with ELECTRE enhances decision-making in complex, multifaceted environments.

The Hamiltonian cycle and path problems are fundamental in graph theory and useful in modelling real-life problems. Research in this area is directed toward designing better and better algorithms for general problems, but also toward defining new special cases for which exact polynomial-time algorithms exist. In the paper, such new classes of digraphs are proposed. The classes include, among others, quasi-adjoint graphs, which are a superclass of adjoints, directed line graphs, and graphs modelling a DNA sequencing problem.

In this paper, we propose a novel image deblurring approach that utilizes a new mask based on the Grünwald-Letnikov fractional derivative. We employ the first five terms of the Grünwald-Letnikov fractional derivative to construct three masks corresponding to the horizontal, vertical, and diagonal directions. Using these matrices, we generate eight additional matrices of size $5\times 5$ for eight different orientations: $\frac{k\pi }{4}$, where $k=0,1,2,\dots ,7$. By combining these eight matrices, we construct a $9\times 9$ mask for image deblurring that relates to the order of the fractional derivative. We then categorize images into three distinct regions: smooth areas, textured regions, and edges, utilizing the Wakeby distribution for segmentation. Next, we determine an optimal fractional derivative value tailored to each image category to effectively construct masks for image deblurring. We applied the constructed mask to deblur eight brain images affected by blur. The effectiveness of our approach is demonstrated through evaluations using several metrics, including PSNR, AMBE, and Entropy. By comparing our results to those of other methods, we highlight the efficiency of our technique in image restoration.

This paper examines ranking reversal (RR) in Multiple Criteria Decision Making (MCDM) using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). Through a mathematical analysis of min-max and max normalization techniques and distance metrics (Euclidean, Manhattan, and Chebyshev), the study explores their impact on RR, particularly when new, high-performing alternatives are introduced. This research provides insight into the causes of RR, offering a framework that clarifies when and why RR occurs. The findings help decision-makers select appropriate techniques, promoting more consistent and reliable outcomes in real-world MCDM applications.

In this paper, firstly, we propose two new GTHFNs-prioritized aggregation operators called generalized trapezoidal hesitant fuzzy number prioritized weighted average operator and generalized trapezoidal hesitant fuzzy number prioritized weighted geometric operator. Secondly, we investigate the fundamental properties of the operators in detail such as idempotency, boundedness and monotonicity. Thirdly, we propose a method based on the developed GTHF-numbers prioritized aggregation operators for solving an MADM problem with GTHF-numbers. Fourthly, we give a numerical example of the developed method. Finally, a comparative analysis is given with some existing methods in solving an MADM problem with GTHF-numbers.

There are different deep neural network (DNN) architectures and methods for performing augmentation on time series data, but not all the methods can be adapted for specific datasets. This article explores the development of deep learning models for time series, applies data augmentation methods to conveyor belt (CB) tension signal data and investigates the influence of these methods on the accuracy of CB state classification. CB systems are one of the essential elements of production processes, enabling smooth transportation of various industrial items, therefore its analysis is highly important. For the purpose of this work, multi-domain tension data signals from five different CB load weight conditions (0.5 kg, 1 kg, 2 kg, 3 kg, 5 kg) and one damaged belt condition were collected and analysed. Four DNN models based on fully convolutional network (FCN), convolutional neural network combined with long short-term memory (CNN-LSTM) model, residual network (ResNet), and InceptionTime architectures were developed and applied to classification of CB states. Different time series augmentations, such as random Laplace noise, drifted Gaussian noise, uniform noise, and magnitude warping, were applied to collected data during the study. Furthermore, new CB tension signals were generated using a TimeVAE model. The study has shown that DNN models based on FCN, ResNet, and InceptionTime architectures are able to classify CB states accurately. The research has also shown that various data augmentation methods can improve the accuracy of the above-mentioned models, for example, the combined addition of random Laplace and drifted Gaussian noise improved FCN model’s baseline (without augmentation) classification accuracy with 2.0 s-length signals by 4.5% to 92.6% ± 1.54%. FCN model demonstrated the best accuracy and classification performance despite its lowest amount of trainable parameters, thus demonstrating the importance of selecting and optimizing the right architecture when developing models for specific tasks.