You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.

Search for Articles:

Title / Keyword

Author / Affiliation / Email

Journal

Article Type

Advanced Search

Section

Special Issue

Volume

Issue

Number

Page

Logical OperatorOperator

Search Text

Search Type

Information-Driven Team Collaboration in RoboCup Rescue
Analysis of Japanese Twitter Posts Related to COVID-19 Vaccination Focusing on Frequently Occurring Words and Emotional Expressions
Uncertainty-Aware Machine Learning for NBA Forecasting in Digital Betting Markets
FDSDS: A Fuzzy-Based Driver Stress Detection System for VANETs Considering Interval Type-2 Fuzzy Logic and Its Performance Evaluation
AI/ML Based Anomaly Detection and Fault Diagnosis of Turbocharged Marine Diesel Engines: Experimental Study on Engine of an Operational Vessel

Journal Description

Information

Information is a scientific, peer-reviewed, open access journal of information science and technology, data, knowledge, and communication, published monthly online by MDPI. The International Society for the Study of Information (IS4SI) is affiliated with Information and its members receive discounts on the article processing charges.

Open Access— free for readers, with article processing charges (APC) paid by authors or their institutions.
High Visibility: indexed within Scopus, ESCI (Web of Science), Ei Compendex, dblp, and other databases.
Journal Rank: JCR - Q2 (Computer Science, Information Systems) / CiteScore - Q2 (Information Systems)
Rapid Publication: manuscripts are peer-reviewed and a first decision is provided to authors approximately 20.9 days after submission; acceptance to publication is undertaken in 3.6 days (median values for papers published in this journal in the second half of 2025).
Recognition of Reviewers: reviewers who provide timely, thorough peer-review reports receive vouchers entitling them to a discount on the APC of their next publication in any MDPI journal, in appreciation of the work done.
Journal Cluster of Information Systems and Technology: Analytics, Applied System Innovation, Cryptography, Data, Digital, Informatics, Information, Journal of Cybersecurity and Privacy and Multimedia.

Impact Factor: 2.9 (2024); 5-Year Impact Factor: 3.0 (2024)

Imprint Information Journal Flyer Open Access ISSN: 2078-2489

Latest Articles

18 pages, 8736 KB

Open AccessArticle

Data-Driven Model Reference Neural Control for Four-Leg Inverters Under DC-Link Voltage Variations

by Ana J. Marín-Hurtado, Andrés Escobar-Mejía and Eduardo Giraldo

Information 2026, 17(2), 171; https://doi.org/10.3390/info17020171 (registering DOI) - 7 Feb 2026

The Four-Leg Three-Phase Voltage Source Inverter (4LVSI) is a versatile solution for integrating renewable energy sources (RESs) into distribution networks, as it compensates unbalanced voltages and currents while providing a path for zero-sequence components. Accurate current control is essential to ensure power quality [...] Read more.

The Four-Leg Three-Phase Voltage Source Inverter (4LVSI) is a versatile solution for integrating renewable energy sources (RESs) into distribution networks, as it compensates unbalanced voltages and currents while providing a path for zero-sequence components. Accurate current control is essential to ensure power quality and reliable operation under these conditions. Conventional controllers such as proportional–integral, resonant, or feedback-linearization methods achieve acceptable tracking under static dc-link conditions, but their performance degrades when dc-link voltage dynamics arise due to renewable-source fluctuations. This paper proposes a data-driven model reference neural control (MRNC) strategy for a four-leg inverter connected to RESs, explicitly accounting for dc-link voltage variations. The proposed controller reformulates the classical Model Reference Adaptive Control (MRAC) as a lightweight single-layer neural network whose adaptive weights are updated online using the Recursive Least Squares (RLS) algorithm. In this framework, the dc-link variations are not modeled explicitly but are implicitly learned through the data-driven adaptation process, as their influence is captured in the neural network regressors formed from real-time input–output measurements. This allows the controller to continuously identify the inverter dynamics and compensate the effect of dc-link fluctuations without requiring additional observers or prior modeling. The proposed approach is validated through detailed time-domain simulations and real-time Hardware-in-the-Loop (HIL) experiments implemented at a 10 kHz switching frequency. The results indicated that the RLS-based MRNC controller achieved the lowest steady-state current error, reducing it by approximately 1.85% and 1% compared to the Proportional-Resonant (PR) and One-Step-Ahead (OSAC) controllers, respectively. Moreover, under dc-link voltage variations, the proposed controller significantly reduced the current overshoot, achieving decreases of 5.9 A and 6.36 A relative to the PR controller. Full article

(This article belongs to the Special Issue Feature Papers in Information in 2024–2025)

19 pages, 2661 KB

Open AccessArticle

Data-Driven Reconstruction of the Singapore Stone: A Numerical Imputation Method of Epigraphic Restoration

by Tehreem Zahra, Francesco Perono Cacciafoco and Muhammad Tayyab Zamir

Information 2026, 17(2), 170; https://doi.org/10.3390/info17020170 (registering DOI) - 7 Feb 2026

One of the key artefacts of epigraphy in Southeast Asia is the Singapore Stone inscription, which is, unfortunately, in a poor condition. There are huge spaces that separate the readable characters, rendering the text incomplete. This renders a traditional reconstruction and interpretation by [...] Read more.

One of the key artefacts of epigraphy in Southeast Asia is the Singapore Stone inscription, which is, unfortunately, in a poor condition. There are huge spaces that separate the readable characters, rendering the text incomplete. This renders a traditional reconstruction and interpretation by philologists extremely challenging. We consider epigraphic restoration as a data-restoration task in this paper. We represent the inscription as a system of categorical symbols, in keeping with the original spatial disposition of characters and spaces. Our model is trained in a conservative, data-driven manner using the observed symbols to learn the local transition statistics, and it takes advantage of this information to make plausible predictions of the most likely characters in missing sequences that are short and well-constrained. The procedure generates a probabilistic hypothesis of restoration, which can be audited, as opposed to one definitive reading. The validation of masked-character recovery demonstrates that the model has a mean top-one error of 53.3%, which represents a significantly worse performance compared with simple baseline methods. The process is focused on interaction and transparency with experts. It relies upon assurance scores and prioritised alternative completions of each proposed reconstruction, as a useful means to produce hypotheses in computational epigraphy and the digital humanities. Full article

► Show Figures

Graphical abstract

20 pages, 1819 KB

Open AccessArticle

Entropy, Information, and the Curvature of Spacetime in the Informational Second Law

by Florian Neukart, Eike Marx and Valerii Vinokur

Information 2026, 17(2), 169; https://doi.org/10.3390/info17020169 - 6 Feb 2026

We develop an informational extension of spacetime thermodynamics in which local entropy production is coupled to spacetime curvature within an effective covariant framework. Spacetime is modeled as a continuum limit of finite-capacity information registers, giving rise to a coarse-grained entropy field whose gradients [...] Read more.

We develop an informational extension of spacetime thermodynamics in which local entropy production is coupled to spacetime curvature within an effective covariant framework. Spacetime is modeled as a continuum limit of finite-capacity information registers, giving rise to a coarse-grained entropy field whose gradients define an informational flux. Within a nonminimally coupled scalar–tensor formulation, the resulting field equations imply that the local divergence of this flux is sourced by the Ricci scalar, establishing a direct relation between curvature and entropy production. The corresponding integral form links cumulative entropy generation to the integrated spacetime curvature over a causal region. In stationary limits, the framework reproduces the Bekenstein–Hawking entropy of horizons, while in homogeneous expanding cosmologies it yields monotonic entropy growth consistent with the observed arrow of time. The construction remains compatible with unitarity at the microscopic level and with holographic entropy bounds in the stationary limit. Numerical solutions in flat FLRW backgrounds are used as consistency checks of the coupled evolution equations and confirm the expected curvature–entropy behavior across cosmological epochs. Overall, the results provide a thermodynamically consistent interpretation of curvature as a geometric source of irreversible information flow, without modifying the underlying gravitational field equations. Full article

(This article belongs to the Section Information Theory and Methodology)

28 pages, 1672 KB

Open AccessSystematic Review

Gamification in Digital Mental Health Interventions: A Systematic Review of the Engagement–Efficacy–Ethics Trilemma

by Harold Ngabo-Woods, Larisa Dunai, Isabel Seguí Verdú and Valentina Tîrșu

Information 2026, 17(2), 168; https://doi.org/10.3390/info17020168 - 6 Feb 2026

Digital Mental Health Interventions (DMHIs) offer a scalable solution to the global mental health crisis, yet their real-world impact is often hampered by low user engagement. Gamification has been widely adopted as a strategy to enhance adherence, but its implementation creates a complex [...] Read more.

Digital Mental Health Interventions (DMHIs) offer a scalable solution to the global mental health crisis, yet their real-world impact is often hampered by low user engagement. Gamification has been widely adopted as a strategy to enhance adherence, but its implementation creates a complex and often unacknowledged “Engagement–Efficacy–Ethics Trilemma”. This systematic review synthesises the current literature to deconstruct this trilemma, arguing that an uncritical focus on maximising engagement can fail to improve—or may even undermine—clinical efficacy, while simultaneously introducing significant ethical risks. Our analysis reveals a persistent “Engagement–Efficacy Gap”, where increased usage of mobile health applications (mHealth apps) does not consistently translate to better therapeutic outcomes. Furthermore, we map the ethical landscape, identifying potential harms such as manipulation, psychological distress, and privacy violations that arise from persuasive design. The roles of Artificial Intelligence (AI) in personalising these experiences and Human–Computer Interaction (HCI) in mediating user responses are critically examined as key factors that both amplify and potentially mitigate the tensions of the trilemma. The findings indicate a pressing need for a paradigm shift toward an integrated approach that concurrently evaluates engagement, efficacy, and ethical integrity. We conclude by proposing a framework for responsible innovation, emphasising theory-driven design, co-design with users, and prioritising intrinsic motivation to harness the potential of gamified DMHIs safely and effectively. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, a systematic search was conducted across Scopus, Web of Science, MEDLINE, and PsycINFO for studies published between 2015 and 2025. Full article

► Show Figures

Figure 1

27 pages, 9344 KB

Open AccessArticle

Unitary Operations with Five and More Qubits: Roadmaps and Effective Quantum Circuits

by Artyom M. Grigoryan

Information 2026, 17(2), 167; https://doi.org/10.3390/info17020167 - 6 Feb 2026

This article presents the general method of QR decomposition of

r

-qubit operations,

r \geq 3

, by means of quantum signal-induced heap transformations (QsiHT). These are quantum analogues of discrete signal-induced heap transformations, which are generated by given signals and paths, [...] Read more.

This article presents the general method of QR decomposition of

r

-qubit operations,

r \geq 3

, by means of quantum signal-induced heap transformations (QsiHT). These are quantum analogues of discrete signal-induced heap transformations, which are generated by given signals and paths, or orders, of processing the data. The case of the 5-qubit operations is described in detail, and a recurrent form of calculation of all 5-qubit QsiHTs from the 4-qubit QsiHTs is given. For that, roadmaps and quantum circuits are presented for all 31 5-point QsiHTs that are used in the QR decomposition. New roadmaps, namely the schemes with paths for performing basic operations on qubits, and corresponding quantum circuits, are also described for the 4- and 3-qubit transformations. All QsiHTs use fast paths, which allow us to calculate the QR decomposition only on disjoint bit planes. As a result, we build the quantum circuits for 5- and more-qubit operations without permutations, only elementary rotations. Unitary operations with real numbers are considered. In the general case, the method of compositing all the roadmaps and quantum schemes for the calculation of any

r

-qubit operation by only

{(2}^{r} - 1) 2^{r - 1}

elementary rotations is described. Full article

(This article belongs to the Section Information Theory and Methodology)

► Show Figures

Graphical abstract

19 pages, 416 KB

Open AccessArticle

Hybrid Intelligence in Requirements Education: Preserving Student Agency in Refining User Stories with Generative AI

by Leon Sterling and Eduardo Oliveira

Information 2026, 17(2), 166; https://doi.org/10.3390/info17020166 - 6 Feb 2026

Generative Artificial Intelligence (Gen AI) offers significant potential to support requirements engineering (RE) education; however, its integration poses challenges regarding accuracy and student engagement. While Gen AI cannot independently specify requirements without hallucinating or overstepping scope, it can serve as a powerful partner [...] Read more.

Generative Artificial Intelligence (Gen AI) offers significant potential to support requirements engineering (RE) education; however, its integration poses challenges regarding accuracy and student engagement. While Gen AI cannot independently specify requirements without hallucinating or overstepping scope, it can serve as a powerful partner in a hybrid intelligence workflow. In this paper, we address the challenge of translating high-level motivational models into detailed user stories, a process that is traditionally labour-intensive for novices. We introduce a structured, human-in-the-loop workflow that uses Gen AI to refine and polish user stories while strictly preserving student agency. By grounding the output from Gen AI in a validated motivational model, the workflow minimises the risk of metacognitive offloading, requiring students to actively critique and validate the initially generated requirements. Our analysis of instructional artefacts demonstrates that Gen AI helps in three ways: suggesting structural improvements, offering alternative professional phrasing, and enhancing readability. However, we also identify risks of intent drift and scope expansion, reinforcing the need for rigorous human oversight. The findings advocate for a pedagogical approach where the Gen AI system acts as a reflective assistant rather than an autonomous generator. Full article

(This article belongs to the Special Issue Using Generative Artificial Intelligence Within Software Engineering)

► Show Figures

Figure 1

42 pages, 2797 KB

Open AccessReview

Decoding Technical Diagrams: A Survey of AI Methods for Image Content Extraction and Understanding

by Nick Bray, Michael Hempel, Matthew Boeding and Hamid Sharif

Information 2026, 17(2), 165; https://doi.org/10.3390/info17020165 - 6 Feb 2026

With artificial intelligence (AI) rapidly increasing in popularity and presence in everyday life, new applications utilizing AI are being explored across virtually all domains, from banking and healthcare to cybersecurity to generative AI for images, voice, and video content creation. With that trend [...] Read more.

With artificial intelligence (AI) rapidly increasing in popularity and presence in everyday life, new applications utilizing AI are being explored across virtually all domains, from banking and healthcare to cybersecurity to generative AI for images, voice, and video content creation. With that trend comes an inherent need for increased AI capabilities. One cornerstone of AI applications is the ability of generative AI to consume documents and utilize their content to answer questions, generate new content, correlate it with other data sources, and more. No longer constrained to text alone, we now leverage multimodal AI models to help us understand visual elements within documents, such as images, tables, figures, and charts. Within this realm, capabilities have expanded exponentially from traditional Optical Character Recognition (OCR) approaches towards increasingly utilizing complex AI models for visual content analysis and understanding. Modern approaches, especially those leveraging AI, are now focusing on interpreting more complex diagrams such as flowcharts, block diagrams, Unified Modeling Language (UML) diagrams, electrical schematics, and timing diagrams. These diagram types combine text, symbols, and structured layout, making them challenging to parse and comprehend using conventional techniques. This paper presents a historical analysis and comprehensive survey of scientific literature exploring this domain of visual understanding of complex technical illustrations and diagrams. We explore the use of deep learning models, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and transformer-based architectures. These models, along with OCR, enable the extraction of both textual and structural information from visually complex sources. Despite these advancements, numerous challenges remain, however. These range from hallucinations, where the content extraction system produces outputs not grounded in the source image, which leads to misinterpretations, to a lack of contextual understanding of diagrammatic elements, such as arrows, grouping, and spatial hierarchy. This survey focuses on five key diagram types: flowcharts, block diagrams, UML diagrams, electrical schematics, and timing diagrams. It evaluates the effectiveness, limitations, and practical solutions—both traditional and AI-driven—that aim to enable the extraction of accurate and meaningful information from complex diagrams in a way that is trustworthy and suitable for real-world, high-accuracy AI applications. This survey reveals that virtually all approaches struggle with accurately extracting technical diagram information. It also illustrates a path forward. Pursuing research to further improve their accuracy is crucial for supporting and enabling various applications, including complex document question answering and Retrieval Augmented Generation (RAG), document-driven AI agents, accessibility applications, and automation. Full article

(This article belongs to the Special Issue Intelligent Image Processing by Deep Learning, 2nd Edition)

► Show Figures

Figure 1

21 pages, 2626 KB

Open AccessArticle

Evaluating Augmented Reality Activities Designed Within the 5E Model in Biology Education

by Christina Volioti, Christos Orovas, Sophia Rapti, Paschalina Gkiosi, Theodosios Sapounidis and Euclid Keramopoulos

Information 2026, 17(2), 164; https://doi.org/10.3390/info17020164 - 6 Feb 2026

The 5E model (1. Engagement; 2. Exploration; 3. Explanation; 4. Elaboration; 5. Evaluation) is a well-known instructional framework for promoting active learning. Augmented reality (AR) has been integrated into the educational process to support interactive learning. Building on these foundations, this study examines [...] Read more.

The 5E model (1. Engagement; 2. Exploration; 3. Explanation; 4. Elaboration; 5. Evaluation) is a well-known instructional framework for promoting active learning. Augmented reality (AR) has been integrated into the educational process to support interactive learning. Building on these foundations, this study examines the usability of an AR application developed within the framework of the 5E instructional model to support biology in high school. The proposed AR app consists of a structured sequence of activities aligned with the 5E stages. Moreover, a usability study was conducted to evaluate the app’s usability among 195 users across three participant groups emerging from various educational levels and with different backgrounds: education students (novice adults), engineering students (expert adults), and 1st year high school students (students). The findings indicate perceived acceptable usability, with younger and less experienced users indicating the need for more guidance. Rather than evaluating learning outcomes or pedagogical effectiveness, this study focuses on investigating usability perceptions and interactions of the AR app designed within the 5E learning model. The findings highlight usability-related issues relevant to the design of AR applications and emphasize the importance of combining user-centered design with instructional frameworks in secondary education. Full article

(This article belongs to the Section Information and Communications Technology)

► Show Figures

Figure 1

21 pages, 2169 KB

Open AccessArticle

Enhancing Early Detection of Alzheimer’s Disease via Vision Transformer Machine Learning Architecture Using MRI Images

by Wided Hechkel, Marco Leo, Pierluigi Carcagnì, Marco Del-Coco and Abdelhamid Helali

Information 2026, 17(2), 163; https://doi.org/10.3390/info17020163 - 6 Feb 2026

Computer-aided diagnosis (CAD) systems based on deep learning have shown significant potential for Alzheimer’s disease (AD) stage classification from Magnetic Resonance Imaging (MRI). Nevertheless, challenges such as class imbalance, small sample sizes, and the presence of multiple slices per subject may lead to [...] Read more.

Computer-aided diagnosis (CAD) systems based on deep learning have shown significant potential for Alzheimer’s disease (AD) stage classification from Magnetic Resonance Imaging (MRI). Nevertheless, challenges such as class imbalance, small sample sizes, and the presence of multiple slices per subject may lead to biased evaluation and statistically unreliable performance, particularly for minority classes. In this study, a Vision Transformer (ViT)-based framework is proposed for multi-class AD classification using a Kaggle dataset containing 6400 MRI slices across four cognitive stages. A subject-wise data-splitting strategy is employed to prevent information leakage between the training and testing sets, and the statistical unreliability of near-perfect scores in underrepresented classes is critically examined. An ablation study is conducted to assess the contribution of key architectural components, demonstrating the effectiveness of self-attention and patch embedding in capturing discriminative features. Furthermore, attention-based visualization maps are incorporated to highlight brain regions influencing the model’s decisions and to illustrate subtle anatomical differences between MildDemented and VeryMildDemented cases. The proposed approach achieves a test accuracy of 97.98%, outperforming existing methods on the same dataset while providing improved interpretability. It supports early and accurate AD stage identification. Full article

(This article belongs to the Special Issue Computer Vision, Pattern Recognition, and Machine Learning in Italy—Second Edition)

► Show Figures

Graphical abstract

18 pages, 1137 KB

Open AccessArticle

Evaluating ChatGPT’s Cognitive Performance in Chemical Engineering Education

by Salman Shahid and Shaun Walmsley

Information 2026, 17(2), 162; https://doi.org/10.3390/info17020162 - 5 Feb 2026

Large Language Models (LLMs) now occupy a prominent role in science, engineering, and higher education. Their capacity to generate step-wise solutions, conceptual explanations, and problem-solving pathways creates new opportunities—but also new risks—for Chemical Engineering learners. Despite widespread informal use, few empirical studies have [...] Read more.

Large Language Models (LLMs) now occupy a prominent role in science, engineering, and higher education. Their capacity to generate step-wise solutions, conceptual explanations, and problem-solving pathways creates new opportunities—but also new risks—for Chemical Engineering learners. Despite widespread informal use, few empirical studies have evaluated LLM performance using a systematically designed dataset mapped directly to Bloom’s Taxonomy. This study evaluates the competency of ChatGPT in solving Chemical Engineering problems mapped to Bloom’s Taxonomy. A diverse dataset of undergraduate-level problems spanning six cognitive domains was used to assess the model’s reasoning across ascending levels of cognitive complexity. Each response was evaluated for accuracy and categorized into five error types. Although ChatGPT demonstrated considerable potential across a range of topics, the analysis also revealed important challenges and limitations that inform best practices for integrating LLMs into Chemical Engineering education. Results show significant differences in ChatGPT performance across Bloom levels, revealing three distinct tiers of capability. Strong performance was observed at lower cognitive levels (Remember–Apply), while substantial degradation occurred at Analyze, Evaluate, and specially Create. The findings provide a nuanced, empirically grounded understanding of current LLM capability limits, with practical recommendations for educators integrating LLMs into engineering curricula. Full article

(This article belongs to the Special Issue Advances in Digital Learning Processes: Innovation, Ethics and Governance)

► Show Figures

Graphical abstract

21 pages, 4705 KB

Open AccessArticle

Computational and Graph-Theoretic Analysis of Legislative Networks: New Zealand’s Mental Health Act as a Case Study

by Iman Ardekani, Maryam Ildoromi, Neda Sakhaee, Sewmini Gunawardhana and Parmida Raeis

Information 2026, 17(2), 161; https://doi.org/10.3390/info17020161 - 5 Feb 2026

This paper presents a computational framework for constructing and analysing a focal legislative citation network. A depth-limited expansion strategy generates subgraphs of the network that capture the local structural environment of a seed Act while avoiding the global hub dominance present in whole-corpus [...] Read more.

This paper presents a computational framework for constructing and analysing a focal legislative citation network. A depth-limited expansion strategy generates subgraphs of the network that capture the local structural environment of a seed Act while avoiding the global hub dominance present in whole-corpus analyses. Centrality measures and community detection show how the seed Act’s perceived influence changes with network radius. To incorporate semantic information, we develop and apply an Large Language Model (LLM)-assisted topic modelling method in which representative keywords and LLM-generated summaries form a compact text representation that is converted into a Term Frequency-Inverse Document Frequency (TF–IDF) document–term matrix. Although demonstrated on New Zealand’s mental health legislation, the framework generalises to any legislative corpus or jurisdiction. Integrating graph-theoretic structure with LLM-assisted semantic modelling provides a scalable approach for analysing legislative systems, identifying domain-specific clusters, and supporting computational studies of legal evolution and policy impact. Full article

(This article belongs to the Section Information Theory and Methodology)

► Show Figures

Figure 1

22 pages, 1515 KB

Open AccessArticle

Model for Diversifying iGaming Through Financial Derivatives

by Petko Iliev and Daniela Orozova

Information 2026, 17(2), 160; https://doi.org/10.3390/info17020160 - 5 Feb 2026

The present study analyzes the possibilities for diversification in the iGaming sector through the integration of concepts derived from financial derivatives theory. The main idea is the development of a model introducing a mechanism for buying and selling bets between two clients as [...] Read more.

The present study analyzes the possibilities for diversification in the iGaming sector through the integration of concepts derived from financial derivatives theory. The main idea is the development of a model introducing a mechanism for buying and selling bets between two clients as a means of early position closure—an analog to option trading in capital markets. The model is structured in three phases and four conditions, forming eight scenarios with varying probabilities and expected returns. The analysis demonstrates that, under appropriate parameters, the innovation can be potentially profitable for clients and acceptable for the bookmaker, who may offset potential losses through an increased number of registrations and an enhanced corporate image. The proposed conceptual framework provides a theoretical foundation for the creation of a secondary market in iGaming, which could lead to greater market efficiency, increased liquidity, and the rationalization of player behavior. The results emphasize the significance of an interdisciplinary approach combining game theory, behavioral economics, and financial engineering as a basis for sustainable development and competitive advantage in the dynamically evolving iGaming industry. Full article

► Show Figures

Figure 1

19 pages, 694 KB

Open AccessArticle

A Spanish Language Proficiency Dataset for AI Evaluation

by Anselmo Peñas, Álvaro Rodrigo, Javier Fruns-Jiménez, Inés Soria-Pastor, Sergio Moreno-Álvarez, Alberto Pérez and Julio Reyes-Montesinos

Information 2026, 17(2), 159; https://doi.org/10.3390/info17020159 - 5 Feb 2026

Benchmarking Spanish reading comprehension remains challenging due to the scarcity of proficiency-calibrated resources grounded in authentic human assessments. We introduce IC-UNED-RC-ES, a benchmark comprising more than 6000 items derived from Instituto Cervantes examinations, converted to a machine-readable format while preserving exam structure, proficiency [...] Read more.

Benchmarking Spanish reading comprehension remains challenging due to the scarcity of proficiency-calibrated resources grounded in authentic human assessments. We introduce IC-UNED-RC-ES, a benchmark comprising more than 6000 items derived from Instituto Cervantes examinations, converted to a machine-readable format while preserving exam structure, proficiency levels, and scoring criteria. Unlike many existing resources, IC-UNED-RC-ES includes a diverse set of exercise formats, combining common multiple-choice questions with new formats such as matching and fill-in-the-gap, which support a broader assessment of reading skills. The benchmark supports evaluation at both the item and exam levels and includes an exercise taxonomy with category-specific metrics. Baseline results with current AI systems reveal a strong difficulty effect (a 15-point drop from lower to advanced levels) and substantial variation across exercise types, with inference- and discourse-heavy categories reaching only 41%. IC-UNED-RC-ES provides a human-aligned, interpretable testbed for diagnosing strengths and weaknesses in Spanish reading comprehension and for tracking progress across model generations. Full article

(This article belongs to the Section Artificial Intelligence)

► Show Figures

Graphical abstract

17 pages, 980 KB

Open AccessArticle

Dual-View Sign Language Recognition via Front-View Guided Feature Fusion for Automatic Sign Language Training

by Siyuan Jing and Gaorong Yan

Information 2026, 17(2), 158; https://doi.org/10.3390/info17020158 - 5 Feb 2026

The foundation of an automatic sign language training (ASLT) system lies in word-level sign language recognition (WSLR), which refers to the translation of captured sign language signals into sign words. However, two key issues need to be addressed in this field: (1) the [...] Read more.

The foundation of an automatic sign language training (ASLT) system lies in word-level sign language recognition (WSLR), which refers to the translation of captured sign language signals into sign words. However, two key issues need to be addressed in this field: (1) the number of sign words in all public sign language datasets is too small, and the words do not match real-world scenarios, and (2) only single-view sign videos are typically provided, which makes solving the problem of hand occlusion difficult. In this work, we design an efficient algorithm for WSLR which is trained on our recently released NationalCSL-DP dataset. The algorithm first performs frame-level alignment of dual-view sign videos. A two-stage deep neural network is then employed to extract the spatiotemporal features of the signers, including hand motions and body gestures. Furthermore, a front-view guided early fusion (FvGEF) strategy is proposed for effective fusion of features from different views. Extensive experiments were carried out to evaluate the algorithm. The results show that the proposed algorithm significantly outperformed existing dual-view sign language recognition algorithms. Compared with several state-of-the-art methods, the proposed algorithm achieves Top-1 accuracy on the NationalCSL6707 dataset that is 10.29 and 11.38 higher than MViT and CNN + Transformer, respectively. Full article

(This article belongs to the Special Issue Advancing AI Applications in Education and Engineering: A Multidisciplinary Perspective)

► Show Figures

Graphical abstract

20 pages, 888 KB

Open AccessArticle

Writing Accuracy: How AI-Assisted Writing Instruction Can Support EFL Undergraduate Students

by Hana Almutairi, Abdullah A. Alfaifi and Mohammad Saleem

Information 2026, 17(2), 157; https://doi.org/10.3390/info17020157 - 5 Feb 2026

Recent research suggests that artificial intelligence (AI) tools allow EFL (English as a Foreign Language) learners to exert greater control over their language-learning process. Furthermore, these tools enhance their language skills by providing them with elements often absent in traditional classroom settings, such [...] Read more.

Recent research suggests that artificial intelligence (AI) tools allow EFL (English as a Foreign Language) learners to exert greater control over their language-learning process. Furthermore, these tools enhance their language skills by providing them with elements often absent in traditional classroom settings, such as autonomy and individual pace of learning. Specifically, AI-based tools, such AI chatbots, have the potential to facilitate learning and streamline tasks for both students and instructors in language-learning contexts. These digital companions (i.e., chatbots) can be methodically crafted and equipped with the required essential materials to support students in practising language skills independently, regardless of time or location. The current study presents an experiment conducted with undergraduate students at a university in Saudi Arabia to assess the effectiveness of a customised AI chatbot, WritePro (GPT-4), in improving their writing skills. Learners in the experimental group were instructed to use WritePro to navigate through their writing stages, focusing their queries on four key components: content and organisation, grammar mechanics, vocabulary usage, and sentence structure. The findings showed that WritePro serves as an effective tool for EFL learners to overcome several challenges in developing writing competencies. Therefore, the significance of these findings lies in the potential of AI tools to enable EFL instructors to effectively integrate chatbots into classroom instruction, supporting the development of students’ writing skills. Furthermore, these findings can be used as a basis for advocacy with university policymakers regarding the use of AI chatbots in language classrooms. Full article

(This article belongs to the Special Issue Advances in Digital Learning Processes: Innovation, Ethics and Governance)

► Show Figures

Graphical abstract

20 pages, 668 KB

Open AccessArticle

Two Pathways to Digital Flourishing: How Meaning and Positivity Orientations Shape Online Behavior and Well-Being

by Ofer I. Atad and Pninit Russo-Netzer

Information 2026, 17(2), 156; https://doi.org/10.3390/info17020156 (registering DOI) - 4 Feb 2026

As social media becomes increasingly embedded in daily life, understanding the psychological mechanisms that shape users’ digital experiences is essential for promoting healthy, sustainable digital behavior. This study examines two motivational life-orientation strategies—prioritizing meaning and prioritizing positivity—and investigates how they are differentially associated [...] Read more.

As social media becomes increasingly embedded in daily life, understanding the psychological mechanisms that shape users’ digital experiences is essential for promoting healthy, sustainable digital behavior. This study examines two motivational life-orientation strategies—prioritizing meaning and prioritizing positivity—and investigates how they are differentially associated with patterns of social media use, digital flourishing, and psychological well-being. A sample of 414 adults completed validated measures of digital and psychological outcomes, including a Hebrew adaptation of the Digital Flourishing Scale (DFS), and measures of prioritizing meaning and prioritizing positivity. Across seven regression models, prioritizing meaning was consistently associated with more adaptive digital outcomes, including lower screen time, reduced FoMO, higher digital flourishing, greater self-compassion, and lower psychopathology. In contrast, prioritizing positivity was associated with greater social media engagement and elevated social comparison but showed no associations with digital flourishing or psychological well-being. These findings support a dual-pathway framework, suggesting that meaning-oriented individuals tend to interact with digital platforms more intentionally and resiliently, whereas positivity-oriented individuals engage in more affect-driven and evaluative patterns that do not translate into well-being benefits. The study advances current understanding of digital behavior by identifying motivational factors that shape how users navigate online environments and highlights implications for designing digital well-being interventions and platform features. Full article

(This article belongs to the Special Issue Information Behaviors: Social Media Challenges and Analytics)

► Show Figures

Graphical abstract

25 pages, 1516 KB

Open AccessArticle

Comparative Benchmarking of Deep Learning Architectures for Detecting Adversarial Attacks on Large Language Models

by Oleksandr Kushnerov, Ruslan Shevchuk, Serhii Yevseiev and Mikołaj Karpiński

Information 2026, 17(2), 155; https://doi.org/10.3390/info17020155 - 4 Feb 2026

The rapid adoption of large language models (LLMs) in corporate and governmental systems has raised critical security concerns, particularly prompt injection attacks exploiting LLMs’ inability to differentiate control instructions from untrusted user inputs. This study systematically benchmarks neural network architectures for malicious prompt [...] Read more.

The rapid adoption of large language models (LLMs) in corporate and governmental systems has raised critical security concerns, particularly prompt injection attacks exploiting LLMs’ inability to differentiate control instructions from untrusted user inputs. This study systematically benchmarks neural network architectures for malicious prompt detection, emphasizing robustness against character-level adversarial perturbations—an aspect that remains comparatively underemphasized in the specific context of prompt-injection detection despite its established significance in general adversarial NLP. Using the Malicious Prompt Detection Dataset (MPDD) containing 39,234 labeled instances, eight architectures—Dense DNN, CNN, BiLSTM, BiGRU, Transformer, ResNet, and character-level variants of CNN and BiLSTM—were evaluated based on standard performance metrics (accuracy, F1-score, and AUC-ROC), adversarial robustness coefficients against spacing and homoglyph perturbations, and inference latency. Results indicate that the word-level 3_Word_BiLSTM achieved the highest performance on clean samples (accuracy = 0.9681, F1 = 0.9681), whereas the Transformer exhibited lower accuracy (0.9190) and significant vulnerability to spacing attacks (adversarial robustness

ρ_{s p a c i n g} = 0.61

). Conversely, the Character-level BiLSTM demonstrated superior resilience (

ρ_{s p a c i n g} = 1.0

,

ρ_{h o m o g l y p h} = 0.98

), maintaining high accuracy (0.9599) and generalization on external datasets with only 2–4% performance decay. These findings highlight that character-level representations provide intrinsic robustness against obfuscation attacks, suggesting Char_BiLSTM as a reliable component in defense-in-depth strategies for LLM-integrated systems. Full article

(This article belongs to the Special Issue Public Key Cryptography and Privacy Protection)

► Show Figures

Graphical abstract

24 pages, 5772 KB

Open AccessArticle

Method for Generating Pseudo-NDVI from RVI Derived from Satellite-Borne SAR Imagery Data Using CycleGAN and pix2pix Models

by Kohei Arai, Ria Maruta and Hiroshi Okumura

Information 2026, 17(2), 154; https://doi.org/10.3390/info17020154 - 3 Feb 2026

Continuous vegetation monitoring is essential for predicting crop varieties and yields; however, optical satellite data are frequently unavailable due to cloud cover. To overcome this limitation, this study proposes a method for generating pseudo-NDVI (Normalized Difference Vegetation Index) imagery from RVI (Radar Vegetation [...] Read more.

Continuous vegetation monitoring is essential for predicting crop varieties and yields; however, optical satellite data are frequently unavailable due to cloud cover. To overcome this limitation, this study proposes a method for generating pseudo-NDVI (Normalized Difference Vegetation Index) imagery from RVI (Radar Vegetation Index) derived from Synthetic Aperture Radar (SAR) data using Generative Adversarial Networks (GANs). Two architectures—pix2pixHD (supervised) and CycleGAN (unsupervised)—were evaluated using Sentinel-1 and Sentinel-2 data under identical conditions. By introducing RVI as an intermediate feature instead of directly converting SAR backscatter to NDVI, the proposed method enhanced physical interpretability and improved correlation with NDVI. Quantitative results show that pix2pix achieved higher accuracy (SSIM = 0.5667, PSNR = 22.24 dB, RMSE = 20.54) than CycleGAN (SSIM = 0.5240, PSNR = 19.54 dB, RMSE = 28.02), with further improvement when combining VV and VH polarization data. Although the absolute accuracy remains moderate, this approach enables continuous annual NDVI time series reconstruction for crop monitoring under persistent cloud conditions, demonstrating clear advantages over conventional direct SAR-to-NDVI conversion methods. Full article

(This article belongs to the Special Issue Next-Generation Vision Systems in Agriculture—Toward Explainable and Trustworthy AI)

► Show Figures

Figure 1

39 pages, 2492 KB

Open AccessSystematic Review

Cloud, Edge, and Digital Twin Architectures for Condition Monitoring of Computer Numerical Control Machine Tools: A Systematic Review

by Mukhtar Fatihu Hamza

Information 2026, 17(2), 153; https://doi.org/10.3390/info17020153 - 3 Feb 2026

Condition monitoring has come to the forefront of intelligent manufacturing and is particularly important in Computer Numerical Control (CNC) machining processes, where reliability, precision, and productivity are crucial. The traditional methods of monitoring, which are mostly premised on single sensors, the localized capture [...] Read more.

Condition monitoring has come to the forefront of intelligent manufacturing and is particularly important in Computer Numerical Control (CNC) machining processes, where reliability, precision, and productivity are crucial. The traditional methods of monitoring, which are mostly premised on single sensors, the localized capture of data, and offline interpretation, are proving too small to handle current machining processes. Being limited in their scale, having limited computational power, and not being responsive in real-time, they do not fit well in a dynamic and data-intensive production environment. Recent progress in the Industrial Internet of Things (IIoT), cloud computing, and edge intelligence has led to a push into distributed monitoring architectures capable of obtaining, processing, and interpreting large amounts of heterogeneous machining data. Such innovations have facilitated more adaptive decision-making approaches, which have helped in supporting predictive maintenance, enhancing machining stability, tool lifespan, and data-driven optimization in manufacturing businesses. A structured literature search was conducted across major scientific databases, and eligible studies were synthesized qualitatively. This systematic review synthesizes over 180 peer-reviewed studies found in major scientific databases, using specific inclusion criteria and a PRISMA-guided screening process. It provides a comprehensive look at sensor technologies, data acquisition systems, cloud–edge–IoT frameworks, and digital twin implementations from an architectural perspective. At the same time, it identifies ongoing challenges related to industrial scalability, standardization, and the maturity of deployment. The combination of cloud platforms and edge intelligence is of particular interest, with emphasis placed on how the two ensure a balance in the computational load and latency, and improve system reliability. The review is a synthesis of the major advances associated with sensor technologies, data collection approaches, machine operations, machine learning, deep learning methods, and digital twins. The paper concludes with what can and cannot be performed to date by providing a comparative analysis of what is known about this topic and the reported industrial case applications. The main issues, such as the inconsistency of data, the lack of standardization, cyber threats, and old system integration, are critically analyzed. Lastly, new research directions are touched upon, including hybrid cloud–edge intelligence, advanced AI models, and adaptive multisensory fusion, which is oriented to autonomous and self-evolving CNC monitoring systems in line with the Industry 4.0 and Industry 5.0 paradigms. The review process was made transparent and repeatable by using a PRISMA-guided approach to qualitative synthesis and literature screening. Full article

► Show Figures

Figure 1

21 pages, 1840 KB

Open AccessArticle

Predicting Cybersecurity Behaviours in Higher Education Institutions: A Data-Driven Analysis of Policy, Culture, and Motivation in the UAE Context

by Said Badreddine, Hamsa Al Ammari, Tariq Alwada’n, Azmi AlAzzam, Asem Omari, Abdullah Almogahed, Ghassan Malkawi and AbdulQuddus Mohammed

Information 2026, 17(2), 152; https://doi.org/10.3390/info17020152 - 3 Feb 2026

This study examines the determinants of cybersecurity behaviour among employees in higher education institutions (HEIs) in the United Arab Emirates (UAE) through an integrated framework combining the Protection Motivation Theory (PMT), the Theory of Planned Behaviour (TPB), and the COM-B model. Using a [...] Read more.

This study examines the determinants of cybersecurity behaviour among employees in higher education institutions (HEIs) in the United Arab Emirates (UAE) through an integrated framework combining the Protection Motivation Theory (PMT), the Theory of Planned Behaviour (TPB), and the COM-B model. Using a convergent mixed-methods design, survey data from 250 participants and interviews with 18 staff members were analysed to identify behavioural drivers and construct a predictive model. Results showed that policy familiarity, security culture, and intrinsic motivation were the strongest predictors of secure behaviour, collectively explaining nearly 70% of variance. The predictive neural network model achieved high accuracy (F1 = 91.99%, R² = 0.94), with interpretability analysis ranking policy familiarity as the top determinant. Qualitative insights reinforced these findings, revealing that policy clarity, leadership engagement, and supportive culture enable consistent compliance. The study contributes a validated, theory-driven predictive framework for HEIs, bridging behavioural science and cybersecurity governance. The findings are based on self-reported cybersecurity behaviour and should be interpreted accordingly. It concludes that effective institutional cybersecurity depends on clear policies, inclusive culture, and sustained motivation rather than fear-based enforcement. Full article

► Show Figures

Figure 1

More Articles...

Submit to Information Review for Information

Journal Menu

Journal Browser

► Journal Browser

Highly Accessed Articles

View More...

Latest Books

More Books and Reprints...

E-Mail Alert

News

6 November 2025
MDPI Launches the Michele Parrinello Award for Pioneering Contributions in Computational Physical Science

2 February 2026
MDPI INSIGHTS: The CEO's Letter #31 - MDPI 30 Years, 500 Journals, UK Summit, Z-Forum Conference, APE

2 February 2026
Acknowledgment to the Reviewers of Information in 2025

More News & Announcements...

Topics

Propose a Topic

Topic in Applied Sciences, Computers, Electronics, Information, J. Imaging

Visual Computing and Understanding: New Developments and Trends Topic Editors: Wei Zhou, Guanghui Yue, Wenhan Yang
Deadline: 31 March 2026

Topic in Applied Sciences, Information, Systems, Technologies, Electronics, AI

Challenges and Opportunities of Integrating Service Science with Data Science and Artificial Intelligence Topic Editors: Dickson K. W. Chiu, Stuart So
Deadline: 30 April 2026

Topic in Electronics, Future Internet, Technologies, Telecom, Network, Microwave, Information, Signals

Advanced Propagation Channel Estimation Techniques for Sixth-Generation (6G) Wireless Communications Topic Editors: Han Wang, Fangqing Wen, Xianpeng Wang
Deadline: 31 May 2026

Topic in AI, Applied Sciences, Education Sciences, Electronics, Information

Explainable AI in Education Topic Editors: Guanfeng Liu, Karina Luzia, Luke Bozzetto, Tommy Yuan, Pengpeng Zhao
Deadline: 30 June 2026

More Topics

Conferences

Propose a Conference Collaboration

26–30 October 2026 The 2nd International Online Conference of the Journal Philosophies: Philosophical Unification of Diverse Forms of Inquiry (IOCPh 2026)

11–13 November 2026 The 1st Online Conference on Technologies

More Conferences...

Special Issues

Propose a Special Issue

Special Issue in Information

Advanced Retrieval-Augmented Generation Systems Based on Large Language Models Guest Editors: Marina Bagic Babac, Andrea Giovanni Nuzzolese
Deadline: 15 February 2026

Special Issue in Information

Deep Learning for Advanced Malware Detection Guest Editors: Prasanthi Sreekumari, Georgios Kambourakis
Deadline: 15 February 2026

Special Issue in Information

Emerging Research in Computational Creativity and Creative Robotics Guest Editors: Sanaz Nikghadam-Hojjati, José Barata, Eda Marchetti
Deadline: 28 February 2026

Special Issue in Information

Emerging Trends in Artificial Intelligence and Machine Learning for Advanced Collaborative Manufacturing Systems Guest Editors: Valentina De Simone, Valentina Di Pasquale, Raffaele Iannone, Fabio Fruggiero
Deadline: 28 February 2026

More Special Issues

Topical Collections

Topical Collection in Information

Knowledge Graphs for Search and Recommendation Collection Editors: Pierpaolo Basile, Annalina Caputo

Topical Collection in Information

Augmented Reality Technologies, Systems and Applications Collection Editors: Ramon Fabregat, Jorge Bacca-Acosta, N.D. Duque-Mendez

Topical Collection in Information

Natural Language Processing and Applications: Challenges and Perspectives Collection Editor: Diego Reforgiato Recupero

Back to TopTop