Informatica

Traditional loss functions such as mean squared error (MSE) are widely employed, but they often struggle to capture the dynamic characteristics of high-dimensional nonlinear systems. To address this issue, we propose an improved loss function that integrates linear multistep methods, system-consistency constraints, and prediction-phase error control. This construction simultaneously improves training accuracy and long-term stability. Furthermore, the introduction of recursive loss and interpolation strategies brings the model closer to practical prediction scenarios, broadening its applicability. Numerical simulations demonstrate that this construction significantly outperforms both mean square error and existing custom loss functions in terms of performance.

Nowadays sustainability and transportation concepts have been incorporated by the authorities and engineers. The indicator of this situation is the introduction of hybrid vehicles into the market. For the consumers, the purchasing process of hybrid vehicles is not easy because of the many alternatives with different brands including different properties. This process is considered a multi criteria problem with multi alternatives. This paper aims to develop a solution methodology for this problem of a company. The proposed methodology integrates the Interval Valued Intuitionistic Fuzzy (IVIF) sets and two Multi Criteria Decision Making (MCDM) methods; Analytic Hierarchy Process (AHP) and the Multi Attributive Border Approximation Area Comparison (MABAC). With the help of IVIF sets, the fuzziness in the structures of the decision problem and decision-making process is overcome. The IVIF AHP evaluation has revealed the importance that consumers attach to the criteria. According to the IVIF AHP results, each of the criteria has a similar weight. According to the IVIF MABAC results, the ranking order of the hybrid vehicle alternatives is specified as A₁–A₂–A₃–A₅–A₄. The advantage of the integrated IVIF AHP and IVIF MABAC approach is that it helps in evaluating the most suitable alternatives when there is a disagreement about the relative suitability of the criteria and requires less numerical calculations. The results and the comparative analysis conducted in the study also support this situation.

Existing fuzzy inference systems are generally based on ordinary fuzzy sets, which do not let the second and third dimensions of the other fuzzy sets extensions to be employed. This paper suggests a decision-making approach by utilizing the fuzzy inference systems (FIS) based on spherical fuzzy sets (SFS). We prefer spherical fuzzy sets to consider the indecision degree together with membership and non-membership degrees in the proposed FIS. During the defuzzification of SF inference system, the indecision degree is distributed over membership and non-membership degree in balance regarding to indecision degree by using a special transformation function. By applying the proposed approach on FIS, it aims to cover hesitancies and uncertainties caused by insufficient assessments of the decision makers more effectively. The proposed decision-making approach is tested with a real-world application in the field of maintenance work order prioritization for scheduling. Finally, the result of the suggested approach based on SFS is compared with the risk assessment matrix technique (RAM) existing in the literature and Picture Fuzzy Inference Systems (PiFIS). It is observed that the proposed Spherical Fuzzy Inference System (SFIS) is more efficient than RAM and PiFIS methods.

Proliferation of wearable healthcare devices has created the need to deliver artificial intelligence applications for these resource-constrained devices to achieve faster, localized decision-making, by bringing computation closer to the data sources, for improved responsiveness and privacy. This contribution presents the results of an experimental evaluation of artificial neural network compression techniques, including quantization, structured pruning, and knowledge distillation, applied to multi-label classification of electrocardiogram (ECG) signals. The experiments were carried out on the PTB-XL dataset using three deep learning models, i.e. an LSTM-based recurrent neural network, a 1D convolutional neural network, and a 1D residual neural network. The results show how the compression methods impact model quality and highlight opportunities to reduce model size and accelerate inference, thereby enabling effective deployment on resource-constrained, edge devices.

Seismic hazard analysis plays a vital role in evaluating the potential earthquake risk in a given region. Northeast India is one of the most seismically active zones due to its tectonic positioning at the collision boundary of the Indian and Eurasian plates. This study aims to implement a comprehensive Seismic Hazard Assessment (SHA) framework using Fuzzy Multi-Criteria Decision Making (MCDM) techniques to improve the accuracy and reliability of Peak Ground Acceleration (PGA) estimates in Northeast India. The methodology integrates Trapezoidal Fuzzy Full Consistency Method (TrF-FUCOM) and Neutrosophic-TOPSIS under Single Valued Neutrosophic Set (SVNS) environment (Neutrosophic-TOPSIS), effectively addressing the limitations of traditional seismic hazard assessment methods, particularly in selecting and weighting Ground Motion Prediction Equations (GMPEs). An extensive earthquake catalogue covering the period from 1762 to 2024 has been analysed, and after declustering, fault zones have been delineated based on earthquake density along active faults. The analysis provides a detailed spatial distribution of Peak Ground Acceleration (PGA) across the region, with the highest PGA value reaching 1.43g using the Deterministic Seismic Hazard Assessment (DSHA) method. The findings of this study offer crucial insights for disaster preparedness, urban planning, and the design of earthquake-resistant infrastructure, helping to mitigate seismic risks and enhance the resilience of communities in Northeast India.

In the legal domain, ontologies organize legal concepts and their relationships, while knowledge graphs connect these concepts to specific entities in legal documents. This study proposes a solution for integrating ontology and knowledge graph, called Legal-Onto model, to construct a knowledge base of an intelligent retrieval system in the legal domain. The Legal-Onto model combines ontology as the conceptual layer and knowledge graphs as the implementation layer for representing the content of legal documents. This relational model is integrated with a structure of knowledge graph to identify relations between concepts and entities extracted from ontology in the determined domain. Moreover, this research addresses inherent challenges in semantic-based knowledge-driven search. The specific objective is to accurately extract relevant information from legal documents to respond to entered queries. The experimental results show that this method is more effective than state-of-the-art methods in natural language processing and large language models, which are without specific legal domain knowledge.

New generation battery technology investments play a key role in the transition process from fossil fuels to renewable energy. The main problem related to the subject is that decision makers experience uncertainty about which of these numerous criteria affecting investment performance are prioritized. The lack of comprehensive models in the literature for systematically prioritizing these criteria creates a significant gap. The aim of this study is to determine the priority strategies to increase the performance of new generation battery technology investments. In this context, an innovative decision-making model is developed by integrating multi-facet fuzzy sets, logarithmic least-squares and WASPAS techniques. This study makes a significant contribution to the literature by prioritizing the performance indicators of new generation battery technology investments via an innovative decision-making model. The development of multi-facet fuzzy sets in this study provides an important contribution to the literature. Moreover, dynamic decision-making opportunity is provided by redefining membership degrees with different parameter sets for each scenario. This provides the opportunity to make clearer decisions based on scenarios and dynamic evaluations in complex decision-making processes. The main findings of the study indicate that circularity and compatibility with existing manufacturing infrastructure are priorities in improving the performance of these projects.

Forward-looking coding has recently been introduced as a source modelling paradigm that exploits predictions of forthcoming symbols. In this paper, we extend this methodology to word-level alphabets, enabling improved compression performance for large and variable-length symbol sets. We present a space-efficient scheme for encoding header information, with particular emphasis on the accurate representation of symbol frequency distributions. In addition, we propose an alternative ordering strategy for word-based dictionaries that leverages the adaptive nature of forward-looking compression. We further show how these techniques can be integrated with a word-based Prediction by Partial Matching model of order one, while avoiding the zero-frequency problem. Experimental results confirm the effectiveness of the proposed approach across multiple datasets.

The open data movement has led to the widespread sharing of data across all sectors, offering great potential for innovation and informed decision-making. Nevertheless, open data quality remains a key challenge. This study provides a systematic overview of 16 recent methodologies for data quality assessment, emphasizing their alignment with ISO/IEC 25012 and ISO 8000 standards, FAIR principles, 5-Star Linked Open Data System, and DCAT vocabulary. We also highlight foundational work and identify adaptable methods suitable for the Slovenian open data portal. By recommending practical approaches, this work provides a strategic basis for improving data quality in regional and national platforms, supporting improved data utilization and transparency for end users.